Livv
Décisions

EC, October 26, 2004, No M.3436

COMMISSION OF THE EUROPEAN COMMUNITIES

Decision

Continental/Phoenix

EC n° M.3436

26 octobre 2004

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Community, Having regard to the Agreement on the European Economic Area, and in particular Article 57(2)(a) thereof, Having regard to Council Regulation (EC) No 139-2004 of 20 January 2004 on the control of concentrations between undertakings (1), and in particular Article 26(2) thereof, Having regard to Council Regulation (EEC) No 4064-89 of 21 December 1989 on the control of concentrations between undertakings (2), as last amended by Regulation (EC) No 1310-97 of 30 June 1997 (3), and in particular Article 8(2) thereof, Having regard to the Commission Decision of 29 June 2004 to initiate proceedings in this case, Having given the undertakings concerned the opportunity to make known their views on the objections raised by the Commission, Having regard to the opinion of the Advisory Committee on Concentrations (4), Having regard to the final report of the Hearing Officer (5), Whereas:

1. On 12 May 2004 Continental AG, Hanover ("Continental"), notified the Commission that it intended to acquire control, within the meaning of Article 3(1)(b) of Regulation (EEC) No 4064-89 ("the Merger Regulation"), of the whole of Phoenix AG, Hamburg ("Phoenix"), through a public takeover bid made on 26 April 2004. Following examination of the notification, the Commission found that the notified operation fell within the scope of the Merger Regulation and raised serious doubts as to its compatibility with the Common market.

2. On 29 June 2004, therefore, the Commission decided to initiate proceedings pursuant to Article 6(c) of the Merger Regulation and Article 57 of the EEA Agreement.

3. After a thorough investigation of the case, the Commission has now reached the conclusion that, while the notified proposal as such is likely to strengthen a dominant position through which effective competition in a substantial part of the Common market would be significantly impeded, the undertakings made by the parties remove the competition reservations about the merger identified by the Commission.

I. THE PARTIES

4. Continental is a German limited company whose business activity includes the manufacture and distribution of technical rubber products such as conveyor belts, air spring systems, rubber-metal parts for vibration damping and hoses. Continental also produces and distributes tyres and brakes for cars and utility vehicles.

5. Phoenix is also a German limited company which produces and distributes technical rubber products (in particular conveyor belts, air spring systems, rubber-metal parts for vibration damping and hoses). Phoenix also manufactures products for sound insulation.

II. THE PLANNED OPERATION

6. The notified merger is to take place through the acquisition by Continental of the majority of the shares in Phoenix through a public takeover bid. The bid was published on 26 April 2004. On 28 June, Continental acquired 75.51% of the shares in Phoenix.

III. CONCENTRATION

7. Through the planned transaction, Continental will acquire sole control of Phoenix. This will result in a concentration within the meaning of Article 3(1)(b) of the Merger Regulation.

IV. COMMUNITY DIMENSION

8. The relevant undertakings achieved a combined aggregate worldwide turnover in 2003 of more than EUR 5 billion (6) (Continental: EUR 11 534.4 million; Phoenix: EUR 1 153.0 million). Continental and Phoenix each have an aggregate Community-wide turnover of more than EUR 250 million (Continental: EUR [...]* million; Phoenix: [...]*), but do not achieve more than two-thirds of their aggregate Community turnover within one and the same Member State. The notified merger accordingly has a Community dimension (Article 1(2) of the Merger Regulation).

V. COMPETITION ASSESSMENT

9. The activities of Continental and Phoenix overlap in a large number of individual product markets. Following detailed examination of the merger, the Commission has concluded that the concentration between the undertakings would create or strengthen a dominant position of the merging parties in particular in the areas of air spring technology and conveyor belts, significantly impeding effective competition in the Common market or a substantial part thereof.

1. RELEVANT MARKETS

(a) AIR SPRINGS FOR UTILITY VEHICLES (OEM/OES)

(aa) Relevant product markets

10. Continental and Phoenix (7) produce various types of air springs. Air springs are a widely used suspension system in utility vehicles (8). Like other suspension systems (e.g. leaf springs or coil springs made from steel), the purpose of air springs is to absorb vertical movements in the wheels and axle caused by unevenness in the road surface. The use of air springs between the axle and the chassis prevents irregularities in the road surface from being transmitted to the vehicle chassis and at the same time ensures that vibrations do not cause the wheels to lose contact with the road surface. Air springs consist of air bellows made of rubber and, as a rule, a metal cover plate and a metal or composite piston (9) to align and seal the rubber bellows. Air springs contain a column of compressed air. Through the compressed air system available in utility vehicles, the internal pressure can be adjusted to the relevant circumstances. Through adjustment of the internal pressure, the cushioning effect in a utility vehicle can be modified and its loading height varied.

Air springs and leaf and coil springs do not constitute a single product market

11. Continental takes the view that all the various types of suspension system (leaf, coil and air springs) constitute a single product market. However, the market survey carried out by the Commission showed that air springs constitute a separate product market that has to be differentiated clearly from other suspension systems (e.g. leaf and coil springs).

12. From the point of view of the market participants surveyed, air springs differ from coil and leaf springs so significantly in terms of technology and application that air springs cannot, because of their different design and smaller size, be exchanged by the vehicle manufacturer for coil or leaf springs without substantial modifications being made to the vehicle (10). Furthermore, the use of coil and leaf springs results in significant losses in ride comfort and to increased fuel consumption and wear and tear on roads. In Europe (11), consequently, air springs are nowadays no longer regarded as interchangeable with steel springs for most applications (12). In the view of the manufacturers surveyed, steel springs are nowadays considered only in cases where a particularly robust spring is required (13), for example in the case of building site vehicles or in areas with very bad road surfaces. Even if occasionally models can be delivered with steel or alternatively air springs, the steel and the air spring model are clearly different products, which are each designed for different requirements (e.g. for different regions). All the manufacturers approached as part of the market survey (14) stated accordingly that air springs are not interchangeable with steel springs. The great majority of customers also indicated that they would not switch from air springs to steel springs (15).

13. The technology used in the production of air springs also differs fundamentally from the technology required for the production of steel spring systems, since know-how in the manufacture of the rubber element used is crucial in the production of air springs. Consequently, the main competitors in the air spring market are undertakings that specialise in the manufacture of rubber products. Most steel spring manufacturers for their part are not active in air spring or rubber production, which is why competitive pressure from a switch in production by steel spring manufacturers should not be expected. All in all, therefore, the Commission takes the view that steel springs should not be taken into account in the competition assessment of the air spring market for utility vehicles and belong to another market.

Distinction between the original equipment manufacturer/original equipment supplier market and the independent aftermarket

14. Continental also proposes in its notification that a distinction be made as regards the relevant product market between the OEM (16) and OES (17) market on the one hand and the IAM (18) on the other. The market survey confirmed that market conditions in the OEM/OES and IAM areas differed so substantially that both markets must be assessed separately.

15. Whereas the OEM/OES market is characterised by close cooperation between spring manufacturers and customers in product development, the independent aftermarket hinges more on the ability to have a broad range of different springs in stock and to be able to supply them as quickly as possible. The installation or sale of a replacement spring requires no close cooperation with the spring manufacturer, since the appropriate springs for each model can be obtained from the manufacturer. As the data necessary for installation (measurements, geometry, etc.) are often known, third-party firms can also copy the necessary springs. It is not the engineering performance of the supplier, but his price which is the key criterion here. Moreover, the customer groups on the two markets are basically different. It is the job of the independent aftermarket dealers to have as big a range of replacement parts as possible in stock. They therefore keep a supply of mostly relatively small quantities from several manufacturers. OEM/OES customers, on the other hand, regularly purchase only products for their own brand, but in relatively large quantities. Not only are the differing competition conditions reflected in the different price level on both markets (19). The different market conditions also become clear from a glance at the respective competitive strengths of the participants in both markets. Thus Continental and Phoenix/Vibracoustic have a significantly weaker market position in the IAM market than in the OEM/OES business (20).

No differentiation between sub-markets for springs for tractor cabs and buses and springs for axles and trailers

16. A number of the market participants surveyed suggested that other separate product markets should be identified, with a distinction being made, for example, between air springs for tractor cabs and buses and for axles and trailers. There do indeed seem to be differences between the two segments at least in so far as manufacturers and customers regularly cooperate more closely in the development of vehicle springs (development partnerships) than in the development of trailer springs.

17. On the other hand, the technical differences between both product segments are so small that all the manufacturers of products for one segment are able to produce springs for the other segment as well or at least could do so without having to make any major changes in production lines. All the leading air spring manufacturers operate in both segments, and many large customers purchase both types of spring (21). The Commission therefore takes the view that it is not necessary to make a distinction between separate product markets for air springs for tractor cabs and bus manufacturers on the one hand and axle and trailer manufacturers on the other (22).

(bb) Geographic market for utility vehicle air springs (OEM/OES): Europe (23)

18. In Continental's view, both the OEM/OES market and the independent aftermarket are world markets. It argues that this is the case, at least as far as the OEM/OES market is concerned, on the grounds that air spring customers are international vehicle, axle and trailer manufacturers which seek and purchase their products on a worldwide basis. However, the results of the market survey have confirmed the Commission in the view which it took when it decided to initiate proceedings that the market for air springs for utility vehicles (OEM/OES) should be defined as European.

(1) Substantial market entry barriers impede access for suppliers without European production facilities in Europe

19. In its detailed market survey, the Commission found that there are a whole series of market entry barriers that substantially impede or in the long term render impossible access for manufacturers without production facilities in Europe.

The product specifications of air springs used outside Europe differ appreciably from those of European air springs, so that production specifically geared to the European market is necessary

20. The Commission's market survey showed that the properties of air springs for utility vehicles outside Europe differ to a not inconsiderable extent from the properties of air springs for the European market. This is due to a variety of factors.

21. Roads in Europe differ significantly from the generally worse roads in Asia, for example. In the NAFTA area too, where road surface quality at least outside built-up areas corresponds on average to European roads, road conditions differ amongst other things due to the fact that roads are much longer and straighter than in Europe. In general, utility vehicles also make much longer trips with more uniform speeds than in Europe (24) and travel more on motorways and less in towns and cities than European vehicles (25). Suspension for European customers is consequently designed for more manoeuvrable vehicles with better handling than the suspension of US vehicles (26).

22. Differences also exist as regards the statutory rules governing the use of air springs. Whereas, according to the information provided by US producers, there are no such rules in the United States, statutory rules in some European countries allow heavy goods vehicles to be used on roads only if they are equipped with air springs (27).

23. Furthermore, the way heavy goods vehicles are made in the United States differs from the way European heavy goods vehicles are made. Thus, the fact that heavy goods vehicles in the United States are, for example, allowed to be much longer and wider affects suspension design, as does the much higher permitted total weight. According to information provided by competitors, US heavy goods vehicles have on average longer bellows and shorter pistons than European heavy goods vehicles (28). Similarly, the different cabin design (29) in European heavy goods vehicles has some impact on the type of spring used. Whereas in the United States standardised trailers with standardised loading dock dimensions are largely used, European customers as a rule require trailers to be equipped with lifting axles which allow the loading height of the trailer to be adjusted, since standardised trailers and loading docks are not used (30). Furthermore, in contrast to Europe, the compressed air supply is allowed to continue to run while idling, which also has to be taken into account in air spring design. All in all, because of the smaller number of axles, European customers use fewer, but technically more sophisticated air springs (31).

24. Other differences regarding air springs stem from the differing driving habits of European and non-European drivers. According to the Commission's information, European truck drivers prefer more comfort than, for example, drivers in the United States. European customers prefer and demand technically much more sophisticated designs than drivers on other continents (32). The lesser importance attached to comfort in the United States is also confirmed by a look at the statistics provided by Continental on the proportion of heavy goods vehicles equipped with air springs in Europe and in the NAFTA countries (33). Moreover, European customers prefer air springs with "natural" frequencies of 1 Hz or lower, and these require hollow pistons which typically are made from steel. In the United States, an air spring frequency of 1.2 Hz is usual, and this can be achieved without reservoir pistons. Consequently, composite pistons are used much more frequently in the NAFTA countries (34).

25. A further important difference between the springs typically used in Europe and in the United States has to do with the material used for the rubber bellows: whereas in the NAFTA countries springs made of natural rubber are customary, most European customers require their manufacturers to be able to supply rubber bellows made of synthetic rubber/neoprene. However, according to the Commission's information, US factories are not as a rule designed to produce neoprene bellows, but produce natural rubber bellows.

26. The fact that the differences described here constitute a not inconsiderable market entry barrier for non-European suppliers without their own European production is also evident - contrary to the view put forward by Continental - from the apparently small differences in thread measurements of fixing screws and thread in covers and pistons. Since most customers in Europe prefer metric measurements, which are not customary in the United States, spring customers are forced either to purchase parts with the right threads in the United States - generally at higher prices - or to purchase them in Europe, transport them to the United States, assemble them there and then transport the entire spring back to Europe (35).

27. In its reply to the statement of objections, Continental maintained that air spring customers were manufacturers operating worldwide who were seeking on the world market the best conditions for their globally distributed products (36). The market survey showed, however, that these customers have different suppliers for different regions. This does not seem very surprising, although - as it has been possible to show - the products which firms operating worldwide, such as DaimlerChrysler or Volvo/Renault, sell in the United States clearly differ from those supplied in Europe.

28. The view taken by the Commission that market conditions for the purchase of air springs in the United States differ from those in Europe (37) is shared by all the air spring manufacturers surveyed (38).

Since success in the air spring business is crucially dependent on engineering during product development, there must be close cooperation between manufacturer and customer, and this can be achieved only with difficulty from outside Europe

29. Where the parties contend that know-how plays a subordinate role in the development of an air spring for a utility vehicle, since it is "generally available" (39), this was not confirmed by the results of the market survey. A utility vehicle air spring is nowadays produced individually for each new type of vehicle axle or trailer. It is adapted, through close cooperation between development engineers on both sides, to the relevant design dimensions, the suspension angles, the weight and a number of other parameters. Frequently, the actual order for a specific spring model is preceded by an order via a development partnership, in which a spring manufacturer undertakes to develop a new air spring type to the point where it is ready for serial production. Such development partnerships, which usually cover a period of 6 to 24 months, usually involve very close cooperation with the customer at all stages of product design, from the manufacturer and testing of prototypes to the manufacture and testing of series-produced models.

30. In a development partnership, the engineering performance of the development partner and its ability to accommodate the wishes and individual requirements of the customer are clearly key. Customers expect the air spring manufacturer to have a precise knowledge of their individual requirements. To be chosen for a development partnership, it is essential to have specific customer and product know-how, which is regularly acquired only through many years' experience of business with the utility vehicle customers of a particular region. But even where an order for an air spring is not preceded by any separate contractually agreed development partnership, the actual purchase is always preceded by a planning and consultation procedure, which means that the seller must be geographically and intellectually close to the customer.

31. The sale of an air spring thus does not consist primarily of the sale of an already available product, but ultimately also involves first and foremost a consultation procedure in which the ideal spring design for the relevant characteristics of the vehicle are developed jointly with the customer.

32. Customers attach the highest priority to avoiding any mistakes in air spring development, since the air spring is an extremely sensitive component in any heavy goods vehicle, whose failure can quickly jeopardise human life as well as the vehicle. In order to ensure absolute reliability of springs, communication between the developer and the customer must operate smoothly. Being completely at home in the national language is therefore a major pre-condition for access to customers. Foreign competitors have reported here that German customers in particular have reservations vis-à-vis foreign engineers that make it significantly more difficult for them to gain all-important customer confidence (40).

33. Although some customers - at least from the group of utility vehicle manufacturers - are undertakings that operate on a worldwide basis, it is still the case at present that specific vehicle models that differ significantly from one another in technical terms (on the basis of the numerous differences described) are supplied for each market (e.g. NAFTA, Community).

34. Accordingly, most of the customers surveyed stated that only European manufacturers have sufficiently precise knowledge of the individual requirements of European customers and can meet their technical and qualitative demands. In their view, the close cooperation required with spring manufacturers also means that the manufacturer must always be easily and rapidly accessible (41). Conversely, domestic and foreign manufacturers identify the need for development partnerships as a clear import barrier for new manufacturers (42). It is all the more serious as the development partner is usually the only or the main supplier for the whole lifetime of the individual model (43).

The trend towards just-in-time manufacturing considerably reinforces the disadvantages of non-European producers

35. Since manufacturing penetration is steadily decreasing in the utility vehicles area as elsewhere and customers wish to reduce their own stocks as far as possible so as to cut costs, the ability to supply the required springs rapidly is a prime consideration for customers. Customers therefore fear that supply bottlenecks could arise in the case of non-European manufacturers (44).

Imports of air springs are also made more difficult by customs duties and transport costs

36. Continental's opinion that longer transport operations do not involve any significant cost disadvantages for foreign manufacturers is not confirmed by such manufacturers. In point of fact, the Community charges an import duty of 2.5% on air springs (45). Transport costs must be added to this import duty, with the result that, according to the estimate of Firestone, the only importer of utility vehicle air springs into Europe, the cost disadvantage amounts to some [5-10%]* (46). A manufacturer wishing to import springs from outside Europe must, however, bear not only customs and transport costs, but also the costs of keeping the necessary central stocks. Thus, Firestone has to maintain a large, cost-intensive central stock in the Netherlands in order to operate in Europe at all. Without the necessary central stocks, Firestone would not be able to supply its customers rapidly enough. However, the long delivery times and the storage of springs again generate financing or interest costs, which European competitors do not incur (47).

37. All in all, the additional costs involved represent a quite considerable import barrier for competitors. The only two non-European competitors, Firestone and Goodyear, stated that the cost disadvantages associated with imports were crucial in deciding on the acquisition or establishment of a production plant in Europe (48).

(2) The existence of significant market entry barriers is confirmed by a look at the generally small quantity of imported air springs - furthermore, such imports may be largely transient

38. In its reply to the decision to initiate proceedings, Continental takes the view that the proportion of air springs (OEM/OES) imported into Europe amounts to up to [...%]*. However, the Commission's detailed market survey provides a different picture of the scale of imports into Europe.

39. At present, all the major manufacturers except one produce their air springs for the European market at a European production plant. The sole exception at present is the US manufacturer Firestone. Firestone's decision to involve itself more actively in the European market was in turn a consequence of the decision taken by one of its main customers, the US truck manufacturer Paccar, as part of a global purchasing strategy, to switch to the US manufacturer Firestone. Shortly after receiving the order for equipping DAF trucks, Firestone decided to set up its own production plant in Europe, since the problems associated with distant production were felt to put it at a competitive disadvantage. Firestone here is following the example set by the only other US firm operating in Europe, Goodyear, which decided from the outset to acquire European production facilities.

40. Although this means that at present there are still imports into the European market amounting to EUR [10-20]* million, corresponding to a market share of [5-15%]*, even if one accepts that imports into Poland will not lapse completely with the start of production at the new Polish factory in 2005, it is to be expected that imports will decline considerably. Overall, therefore, the currently still existing imports of air springs by Firestone from outside Europe are not such as to persuade the Commission that there is a world market. On the contrary, the fact that all the foreign manufacturers now have or are setting up their own plants in Europe is clear evidence - in association with the other import barriers referred to - of a European-wide market.

(3) The differing conditions of competition between Europe and other regions are also reflected in the differing market shares in the various regions

41. The Commission's view that there is a Europe-wide market is further borne out by a look at the market position of the individual competitors in the different regions. While, according to Continental's figures, the US market is clearly dominated by Firestone (72%) and Goodyear (24%), European manufacturers have almost no presence in the United States (Continental itself, which is the European market leader, puts its turnover in the NAFTA area at [<5%]*). In Europe, the two US firms have market shares of only [5-15%]* each.

42. The Commission therefore takes the view that there is a European market for air springs for utility vehicles (OEM/OES).

(b) AIR SPRINGS FOR UTILITY VEHICLES (IAM)

(aa) Relevant product market

43. Continental proposes that a separate market should be defined for air springs which are sold in the free spare parts market to spares dealers or final customers (IAM market). As already explained, the market conditions in the IAM market do in fact differ so substantially from those in the OEM/OES market with regard to the respective customers, the extent of necessary cooperation and the price level that the product market for air springs for utility vehicles which are sold in the free spare parts market has to be assessed separately from the competition viewpoint (49).

(bb) Geographic market

44. Continental proposes that the market for IAM spare parts should be defined globally, the reason being that there are no links between manufacturers and suppliers. The market survey carried out by the Commission did not support this view, however. Because of the large differences in the air springs used in the different regions (e.g. North America, Europe), imports do not play a significant part in the IAM sector either. Many market participants accordingly indicated that knowledge of local conditions and requirements was very important for independent dealers in particular and that the spare parts business was strongly marked by mature relations. The local proximity of the spring supplier is also very important in the IAM market because the dealer must be able to procure and deliver spare parts at very short notice.

45. The Commission therefore assumes that, as with the OEM/OES market, the market for IAM air springs is a Europe-wide market (50).

(c) AIR SPRINGS FOR CARS

(aa) Relevant product market: air springs for cars

46. The market survey did not confirm the view taken by Continental that all spring systems for cars (e.g. coil springs, hydro-pneumatic springs and air springs) belong to a single product market. Rather, the view must be taken that air spring systems differ substantially from other types of suspension, both from the demand and from the supply point of view. Although the differences may be less distinctly perceptible as regards the ride comfort achieved than in the case of utility vehicles, for example, both systems nevertheless have distinctly different suspension properties; furthermore, air springs allow an adjustment in level that with a steel spring system can be achieved only in conjunction with hydraulic or hydro-pneumatic shock absorption. Thus the technical properties of steel springs vary not only in "standard" spring systems but also in comfort suspension systems (e.g. the Active Body Control offered by DaimlerChrysler).

47. Nor does the fact that some car models are supplied with the option of air or hydraulic springs affect the separate competitive assessment of the air spring market. Even if such different equipment variants may be possible, this does not alter the fact that the two types of spring have different spring characteristics. For car manufacturers, air spring systems cannot moreover simply be replaced by steel spring systems, not only because of the different suspension properties but also for image reasons. For most manufacturers who use air springs these springs are an important "prestige" product offered primarily in cars in the luxury segment, so as to differentiate from a traditional steel spring system and create an additional possibility for price differentiation.

48. For this very reason, it should not be expected that air spring customers would simply switch to steel springs in the event of price increases. Air spring system and car manufacturers have confirmed to the Commission moreover that a change from a car equipped with air suspension to steel springs cannot only scarcely be entertained on account of the different product characteristics, but would also not be possible without comparatively large technical modifications on account of different installation measurements and a different spring geometry, which again is associated with corresponding costs (51).

49. It is also an undisputed fact that the manufacturing process differs distinctly for air springs and steel springs and requires completely different know-how. This is reflected in the fact that none of the producers of steel springs active in Europe produce or could produce air springs as well. In particular, the production of the air spring rubber bellows requires specific know-how in this respect, which the large component and automobile manufacturers do not have. Consequently, in view of the lack of flexibility in switching production on the part of steel spring manufacturers, it is not appropriate to include the steel spring market in the investigation as a possible relevant market from the competition point of view. The Commission therefore considers that air springs and steel springs have to be assessed separately.

50. The merger results here in overlaps in the sub-market for car air springs. In the case of car air springs, the unit consists of rubber bellows, which is manufactured using the sleeve design process by spraying onto a metal mandrel (52). The rubber bellows are usually mounted on a piston at the lower end and are secured at the top by a metal or composite plate (as a rule by clamping rings) (53).

51. Car air springs form part of a complex air suspension system. Air suspension systems generally consist of two to four air springs, one or more air compressors, compressed air accumulators, vibration dampers, sensors, control units and other parts needed for operation. Of the merging parties, however, only Continental, and not Phoenix/Vibracoustic, has so far sold entire air spring systems, so that there are no market share additions in the air suspension system area. The market to be investigated more closely here, for the air springs used in air suspension systems, is upstream from the market for air suspension systems. Air springs are purchased by system suppliers or car producers which use the air springs to manufacture their own air suspension systems.

52. There is no need here to distinguish between an OEM and OES air spring market on the one hand and an IAM market on the other, since an independent aftermarket has not hitherto developed, primarily because of the long life of air springs, which as a rule exceeds the life of the car.

53. A further subdivision into different product markets for air springs for "conventional" cars, air springs for sports utility vehicles (SUVs) and air springs for vans is likewise not necessary. The argument against any such subdivision is not only the fact that the vehicle types are technically at present very similar, so that a distinction is difficult (for example, the borderline between SUVs and estate cars/off-road vehicles is often fluid). Above all, however, the production process for all three types of spring is essentially the same (the sleeve design process). All the suppliers can, therefore, manufacture air springs for all vehicle types. On the customer side too, market conditions for all three vehicle types do not differ so significantly as to suggest a separate competition assessment, since almost all customers supply systems for all three vehicle types.

(bb) Relevant geographic market: Europe

54. Continental takes the view that the market for car air springs is worldwide. In so doing, Continental argues in particular that there are already at present worldwide supply relationships. The Commission's market survey showed, however, that European vehicles continue to use European air springs almost exclusively. According to the Commission's information, of the fifteen vehicle models with air springs manufactured in Europe in 2003, [...]* were equipped with Phoenix/Vibracoustic or Continental air springs (54). The only competitor, who operates on a small scale in Europe, is the US manufacturer Gates. Although in 2003 [5-15%]* were still imported, Gates has in the meantime set up its own plant for air springs production in Aachen, from which it serves the European market. The air springs which Gates supplies for the BMW "X5" are made in the United States, but the X5 is produced in the United States and not in Europe. In Europe, even the manufacturer Jaguar which belongs to the US Ford group works with the German air spring manufacturer Vibracoustic and not with a US air spring manufacturer.

55. The reason why European manufacturers show such a clear preference is probably the complex technology involved in air suspension systems, whose design requires close cooperation between spring manufacturer, car maker and suppliers of other system components. Since as a rule the other component suppliers with which the air spring customers cooperate closely mostly for a fairly long period are also located in Europe, most customers prefer to cooperate with European air spring manufacturers (55). European manufacturers' preference may also stem from the current predominance of just-in-time manufacturing nowadays. Because of the lesser degree of manufacturing penetration, the companies endeavour to minimise the danger of losses of production due to difficulties in the supply of individual components. From the manufacturer's point of view as well, the importation of air springs into Europe involves disadvantages not only because of transport costs, but also because of the import duties charged by the Community (56).

56. Lastly, a look at the market shares of the main competitors in Europe and the United States also shows that market conditions differ substantially between the two regions. According to the data provided by the notifying party, over 50% of the market in the NAFTA countries is accounted for by two manufacturers (Delphi 36%, Firestone 18%), which did not operate in Europe in 2003, while Continental and Phoenix/Vibracoustic rank among the smaller suppliers there.

57. For these reasons, the market for car air springs must be defined as Europe-wide and not worldwide.

(d) AIR SPRINGS FOR RAIL VEHICLES

(aa) Relevant product market: air springs for rail vehicles

58. Continental takes the view that all the spring systems used in rail vehicles belong to a single product market, irrespective of whether the products are steel springs, air springs, hydro springs, rubber-metal parts, rubber parts, bushes, bearings, buffers or other spring parts used in rail vehicles. The results of the Commission's market survey indicate, however, that it is necessary to define a separate market for (secondary) air springs57 for rail vehicles.

59. As the "secondary" spring element (58) the air spring is supposed to dampen the transmission of vibrations between the bogie and the vehicle body. An air spring for rail vehicles generally consists not only of the rubber bellows/clamping plate/piston unit, but of other metal and rubber-metal parts attached to it ("bearer spring"). The air spring helps to dampen vibration and sound and to adjust the level at the interface between the bogie and the vehicle body.

Primary suspension systems and other vibration damping parts should not be included in the product market for secondary air springs

60. Continental justifies its view that all types of vibration damping parts in rail vehicles belong to one market essentially by arguing that all three types of component are present at least in all passenger trains (59). This argument is not as such convincing, since the only decisive criterion can be whether the products are regarded as substitutable by the relevant customers and whether possibly the suppliers of other vibration damping products would be able to switch to the production of air springs. Neither is the case here.

61. It is undisputed that air springs in a rail vehicle cannot be substituted by other vibration damping elements. This applies not only to the numerous rubber and rubber-metal parts used in vehicles, but also to primary or secondary steel springs. Neither of these elements can even remotely take over the function of an air spring in the bogie. This has to do not only with the completely different design dimensions, geometries and product characteristics of other spring systems (e.g. steel springs or rubber-metal parts). A changeover is also ruled out because of the importance of air springs for the safety of the entire train. In rail vehicles, the air spring is located in a key position, namely the point of contact between the bogie and the coach body. Each individual air spring is designed precisely to meet the individual requirements of the particular type of rail vehicle. Switching to, for example, a steel spring, which has completely different suspension characteristics and is nowadays used mostly only as a primary spring, would be out of the question on safety and comfort grounds alone.

62. This assessment was shared by all the parties surveyed, who stated that air springs were not substitutable by steel springs (60). This tallies with the distribution of spring types in European trains which have a secondary spring: less than 10% have a steel spring, more than 90% an air spring (61), which indicates that the steel spring is declining further and in high-speed trains has been completely discontinued.

63. Nor is it correct that customers purchase all their vibration damping parts for their rail vehicles almost exclusively as a set. Although it may be economically advantageous to purchase several vibration damping elements from one and the same manufacturer, the key factor is that air springs are not purchased by most customers as part of a complete system together with other primary and miscellaneous suspension elements (62). Almost all customers stated that they purchase primary suspension elements separately from air springs. This may also be partly due to the fact that a number of manufacturers specialise more in the rubber-metal parts area (e.g. Trelleborg), whereas Continental and Phoenix are regarded as the leaders in air springs. Furthermore, of the two market leaders in air springs, Continental and Phoenix, at most Phoenix would be in a position to cover virtually the complete range of suspension elements. Continental's product range in the rubber-metal parts area is, according to competitors, rather limited (63).

64. On the production side too, air springs are not interchangeable with other steel, rubber-metal or rubber parts used for vibration damping in rail vehicles. Although on the production side some production switching flexibility may exist between different types of rubber-metal parts, this does not apply to the production of air springs themselves, which requires quite specific know-how. This applies in particular to the actual air bellows, which in Europe are produced by only four firms (Continental, Phoenix, CF Gomma and Paulstra/Hutchinson). Because of safety considerations, their manufacture requires specific know-how in rubber bellows production, which other competitors have to spend time and effort developing (64).

There is no need for a separate competition assessment of a market for rubber bellows for rail air springs

65. A number of factors seem to argue in favour of the definition of an even narrower market for the rubber bellows used in rail air springs. In Europe, there are only four manufacturers of such air bellows. It is also undisputed that the manufacture of rubber bellows requires a particularly large degree of know-how which manufacturers not otherwise involved in the production of rubber bellows must first acquire. Furthermore, entry into air bellows production involves high investment costs (65). On the other hand, the rubber bellows represent an added value proportion of only some 30% in the total air spring. All air spring manufacturers purchase a sometimes considerable proportion of the parts required for air springs from other manufacturers (either metal parts, rubber-metal parts or rubber bellows, depending on their own production focus). Consequently, from the competition policy point of view, it appears appropriate to dispense with any further demarcation of a market for rubber bellows.

Nor is there any need to distinguish between the OEM/OES market and the independent aftermarket

66. In contrast to the utility vehicle air springs sector, there is in the case of air springs for rail vehicles no need to distinguish between two separate markets for sales to OEM/OES customers and sales to IAM customers. The need for any such distinction is obviated both by the long life of rail air springs (10 to 20 years) and by their individual design for the equipment in which they are to be used. Because of the small numbers of rail air springs produced and the high investment costs for plant and equipment (66) and qualification, the OES is, apart from a few large orders, the only spring supplier for a specific model and also supplies replacement springs. A number of customers buy replacement parts for the entire life of the vehicle along with the initial model (67). An independent aftermarket has accordingly not been able to develop for rail vehicles.

67. The Commission therefore starts from the assumption of a separate market for air springs for rail vehicles.

(bb) Relevant geographic market: Europe

68. Continental takes the view that the market for rail air springs should be defined as worldwide and not Europe-wide. Certain findings of the Commission's market survey indicate, however, that the market has a European bias.

69. Thus, while it is true that, among the customer groups, rail vehicle manufacturers (68) operate worldwide, many customers (for example, the national rail companies which also play a key role in the decision to procure parts) operate only nationally. The question of whether a customer operates worldwide or not cannot be the decisive criterion for the definition of the geographic market. The question is rather whether market conditions in the various regions differ significantly from one another and - above all - whether foreign (actual or potential) competitors are able to exert significant competitive pressure on the European air spring manufacturers.

70. A look at current supply relationships in Europe shows that foreign competitors are currently still active in Europe on a small scale. Thus, even in the case of the internationally active air spring customers in Europe, their worldwide presence is not reflected in their purchasing behaviour. Leading manufacturers of rail vehicles such as Siemens and Alstom purchase their air springs exclusively in Europe. The two largest rail companies in Europe, SNCF and Deutsche Bahn, traditionally work with their respective national air spring suppliers. This national or at least European preference of rail companies is significant in that the rail companies as final customers of the trains equipped with the air springs have a substantial influence on the choice of the air spring manufacturer - not least because of the significance of the air spring for safety considerations (69). Thus, a supplier of air springs must not only qualify with the vehicle manufacturer itself as being a reliable supplier before it can get an order. In order to be authorised to operate with the relevant spring model by the large national railway companies (e.g. Deutsche Bahn, SNCF), an additional approval procedure must be completed. The air springs of a given vehicle are thus ultimately selected in agreement with the rail company, which evidently restricts the options open to the vehicle manufacturers (70).

71. However, non-European suppliers too (e.g. Sumitomo and Toyo from Japan) have already won contracts for the supply of rail air springs in Europe. Toyo and Sumitomo, for instance, supply the Shinkansen high-speed train and are thus basically able to offer more sophisticated solutions as well. In addition, Toyo in particular, through its close cooperation with Gadelius Europe AB, has in the meantime secured a competent sales partner in Europe which has many years' experience in the European business. It should be expected therefore that the share of imports will increase significantly in the coming years.

72. All things considered, the precise definition of the geographic market can ultimately remain open here, since even on the narrowest possible market definition the merging parties would not have a dominant position which would significantly impede competition in the Common market.

(e) ANTI-VIBRATION SYSTEMS FOR MOTOR VEHICLES

(aa) Relevant product market

73. The activity of the parties also overlaps in the field of anti-vibration systems. These are rubber-metal parts (71) which - unlike the springs - are not used to dampen the horizontal oscillations between wheel and axle, but to dampen the higher-frequency vibrations at very different places in the vehicle. Anti-vibration systems are used for example for engine mounting and for supporting the running gear suspension (e.g. at the connections between suspension arms and chassis). They are intended to capture the oscillations and vibrations occurring in the vehicle and prevent their transmission to the vehicle body or the coachwork.

74. As in its earlier decisions (72), the Commission can leave open in this case as well whether the market for anti-vibration systems should be divided into further submarkets (e.g. for running gear bearings, engine bearings and torsional vibration dampers or for sales in the OEM/OES and in the IAM segment). The market survey showed that, even if the markets were further subdivided, no dominant position would arise on one of them which would significantly impede competition in the Common market (73).

(bb) Geographic market

75. Continental considers that the market for anti-vibration systems should be defined as worldwide. In its earlier decisions the Commission defined the geographic market for motor vehicle anti-vibration system parts as "at least EEA-wide". The market survey, however, gave clear indications that the market should be defined as Europe-wide. Thus most European vehicle manufacturers work with European suppliers, and the volume of imports and exports is low.

76. Here too, though, it can ultimately remain open whether the market should be defined as Europe-wide or worldwide, since in neither case would the merger lead to the creation of a dominant position through which competition in the Common market would be significantly impeded (74)

(f) EXPANSION JOINTS/TUBE CONNECTIONS

(aa) Relevant product market

77. Expansion joints are mostly connections made from rubber between two pipes (tube connections) which serve to reduce stresses and vibrations which occur in rotating machinery, in temperature differences, pressure fluctuations or as a result of other influences in piping systems. As flexible tube connections they contribute to preventing wear, leaks or even fatigue failure in the joint between two pipes. Continental and Phoenix are active almost exclusively in the production of unrefined expansion joints, i.e. expansion joints without the necessary metal flange-mounted parts ("refined expansion joints")

78. Whether two separate markets for unrefined and refined expansion joints are to be differentiated, as Continental suggests, can be left open here, since the merger, even assuming a submarket for unrefined expansion joints (where the parties' main activity lies) would not lead to additions of market share that would raise competition concerns.

(bb) Geographic market

79. Whether the market for expansion joints is to be defined as worldwide - as Continental suggests - or as Europe-wide can also be left open, for, even with a Europe-wide definition, a dominant position through which competition in the Common market would be significantly impeded would not be created.

(g) MOTOR VEHICLE HOSES AND HOSE LINES

(aa) Relevant product market

80. Continental and Phoenix produce hoses and hose lines for a variety of different uses. The hoses are made in particular from rubber or synthetic rubber and also from plastics. They are often combined with metal or plastic parts (e.g. with screws or tubes). The parts combined with the hose can, in more complex hose products (e.g. coolant hoses), account for most of the value added. In such cases the concept of "hose line" is used in the industry.

81. Continental proposes that a separate product market should be defined for motor vehicle hoses and hose lines. The Commission agrees with Continental that basic differences between products, customers and suppliers justify leaving hoses for applications other than motor vehicles (e.g. industrial applications) out of the competition analysis.

82. Continental does not take the view, however, that the market for motor vehicle hoses and hose lines should be further subdivided. Although the products do differ, hose manufacturers are able to produce various hoses. On the other hand, however, the parties concede that the know-how for producing many different kinds of hose is only transferable to a certain extent between the different types of hose and hose line. In the BTR/Pirelli case, the Commission likewise defined a submarket within motor vehicle hoses, namely coolant, heating and turbo hoses (75). Other segments where both parties to the merger are active are fuel hoses and hose lines and refrigerant hoses and hose lines. Although basic product and production differences suggest that separate markets should be defined, the question whether the market for automotive hoses and hose lines should be subdivided accordingly can ultimately be left open, since in no case is there a threat of a dominant position in one of the submarkets being created or strengthened which would significantly impede competition in the Common market (76).

(bb) Geographic market

83. Continental considers that the market for all types of automotive hose and hose system should be defined as worldwide, or at least EEA-wide. Even if - as in the rest of the automotive supplied parts sector - there is much in favour of defining a Europe-wide market, the precise geographic definition of the market can be left open here, since the merger, even with a Europe-wide market definition, would not lead to a dominant position on one of the markets through which competition in the Common market would be significantly impeded (77).

(h) HEAVY STEEL CORD CONVEYOR BELTS

(aa) Relevant product market: Heavy steel cord conveyor belts

84. Both Continental and Phoenix achieve a substantial part of their turnover with the sale of conveyor belts. Conveyor belts are belts made mostly of rubber or plastic (if necessary with textile or steel reinforcements) which are used to transport single items or bulk materials. They are used in a wide range of industries in various sizes and to various specifications. The differences relate, for example, to the material used (rubber, various reinforcing materials), dimensions, resistance to tear and abrasion and other special properties (grip, permeability, care of materials, etc.).

85. Continental proposes to distinguish between the product markets for light conveyor belts (PVC belts for moving light goods, e.g. at check-out counters), heavy conveyor belts (rubber belts for transporting heavy, bulk materials) and specialty belts (special-purpose heavy conveyor belts) (for the latter, go straight to (i)).

(1) Light and heavy conveyor belts are not to be assigned to the same market

86. The Commission shares Continental's view that fundamental differences between light and heavy conveyor belts justify leaving light conveyor belts out of account for competition assessment purposes. Light conveyor belts differ not only in terms of their function (78) but also in terms of their characteristics (composition, structure, dimensions) so markedly from heavy conveyor belts that from the demand side point of view they are not interchangeable. Similarly, not only do the customers for heavy and light conveyor belts differ; owing to the fundamentally different production process in the case of heavy conveyor belts, the manufacturers of both types of belt also differ markedly from one another. Whereas the covering layer of a light conveyor belt is usually made of PVC, in the case of heavy conveyor belts rubber is used. Suppliers of light conveyor belts are therefore unable, without a mastery of rubber technology and a secure supply of the necessary primary products, to manufacture heavy conveyor belts. The distinction between light and heavy conveyor belts corresponds, moreover, to the terminology used in the market, as understood by customers and as used by all manufacturers. Neither Continental nor Phoenix is active in the area of the production of light conveyor belts.

(2) Heavy textile conveyor belts are not to be assigned to the same product market as heavy steel cord conveyor belts

87. The two most important categories of heavy conveyor belt are steel cord conveyor belts (79), on the one hand, and textile or textile/fabric-reinforced conveyor belts ("textile conveyor belts"), on the other. In the case of steel cord conveyor belts the rubber coating on the belt is reinforced by the insertion of steel cords. The steel cords are intended first and foremost to prevent the belt, which is made mostly of rubber, from stretching in a lengthwise direction when laden with heavy materials or in the case of a long belt length. Steel cord reinforced belts can carry extremely heavy loads over distances of several kilometres. In the case of textile conveyor belts, on the other hand, it is not steel cords that are used as tensioners but synthetic fibres (nylon, polyester), placed - sometimes in several strata'between the layers of rubber.

88. Continental expresses the view in its observations on the Commission's decision to initiate proceedings that there is no need to distinguish between a separate market for textile and steel cord conveyor belts, but instead that textile belts and steel cord belts belong to a Common market for "heavy conveyor belts". To justify this viewpoint, it is stated in the notification documents (p. 111) that: "Regardless of their composition (fabric/steel cord), [both types of belt] are usable for all applications and are interchangeable". This viewpoint is nevertheless clearly contradicted by the market survey.

89. The findings of the market survey support the Commission's view that textile conveyor belts are not to be included in the competition assessment of the market for steel cord conveyor belts.

90. First of all, Continental itself concedes that, from the customer's standpoint, the question of the interchangeability of steel cord and textile belts regularly arises only in the case of new installations (80), because textile belts cannot be used in existing steel cord conveyor installations without substantial technical modification. As a rule, existing steel cord belt installations are equipped with steel cord belts, and existing textile belt installations with textile belts (81). However, only a small proportion (82) of business transactions involving heavy conveyor belts is accounted for by the equipping of new installations.

Steel and textile belts have, on account of their different characteristics, separate areas of application

91. Even in the case of the residual business involving the equipping of new installations with conveyor belts, the type of conveyor belt is as a rule determined by the conveyor belt use, textile belts being fundamentally different from steel cord belts as far as their basic characteristics are concerned.

92. Whereas Continental maintains on p. 112 of the notification documents that both types of belt possess the "same tear resistance", the market survey has shown that, although textile belts have become tougher in recent years, even the strongest textile belt does not have even half the tensile strength of which steel cord belts are capable (83). According to the results of the Commission's market survey, nearly half of the belts used by the major European lignite producers have duty ratings higher than the strength possessed by textile belts (84).

93. The higher tensile strength of steel cord belts is, however, not the only determining factor when it comes to choosing a steel belt for certain applications. The lower elasticity of steel cord belts is at least as important. According to competitors (85), textile belts stretch under load by up to 2% of their length, whereas steel cord belts stretch only 0.1%. In the case of conveyor installations which can be many kilometres long this difference is of great importance when it comes to choosing the conveyor belt. For all relatively long installations, a steel cord conveyor belt, and not a textile conveyor belt, is therefore the equipment of first choice.

94. It may be technically and theoretically possible to span the same distance using many tensioning rollers and by linking together end-to-end a large number of shorter conveyor installations with textile belts. Such a solution does not, however, make economic sense as it not only requires a far larger technical expenditure (as a result of the necessary additional tensioning and redirecting rollers as well as the necessary motors and electricity supply (86)). It would also, by creating additional linkages and junctions for the transport of material between the individual conveyor installations, substantially increase the wear and tear on the belt. The more axles, motors and tensioning devices an installation requires, however, the more intensive its maintenance will be. Such a cost intensive and technically questionable solution is therefore not a realistic alternative to the use of a long steel cord conveyor belt (87).

95. Just as unconvincing is the argument advanced by Continental to the effect that in the case of long transport distances customers can switch to transport by lorry or rail. European customers have - despite being aware of this fall-back possibility - almost without exception opted for transport by conveyor belt. Moreover, such a transport possibility (which is quite common outside of Europe - e.g. in South America) would, while dispensing with the use of steel cord conveyor belts, nevertheless not justify making textile belts a substitute in this field for steel cord belts.

96. A further difference between steel cord and textile conveyor belts, lastly, is that steel cord belts are able to withstand higher material temperatures than textile belts (88).

97. Given these differences, conveyor belt customers have a clear preference for one or other type of belt, depending on the area of use (89). Whereas textile belts are more suited for lighter goods and for transport over shorter distances, steel cord belts are invariably used where heavy materials have to be transported over particularly long distances. The clearest example of a "traditional" steel cord application is opencast lignite mining. Here, steel cord conveyor belts are used in 95% of cases. A switch to textile conveyor belts is out of the question for such conveyor installations (90). This area of use accounts, however, for more than half of the entire European market for steel cord conveyor belts. For the bulk of the remaining business, the type of belt to be used is determined either by an existing conveyor installation (for steel or textile belts) or by the conveyor application. This finding is not invalidated by the fact that, in the case of a few applications (light goods, short distances), the idea may be entertained of building a new steel or textile conveyor belt. This area of overlap does not, however, play a significant role in the overall steel cord conveyor belt market, which is essentially characterised by the business of equipping existing installations.

98. Market participants have also confirmed that the areas of application of textile and steel reinforced belts differ fundamentally. Manufacturers (91) and customers (92) have almost unanimously answered in the negative the question whether they might substitute a textile conveyor belt for a steel cord conveyor belt for the same application (93).

As a result of there being little product substitutability, textile belt manufacturers do not exert any appreciable competitive pressure

99. From the supply side point of view also, there is a clear distinction between steel cord and textile conveyor belts. Continental's assertion that all heavy conveyor belts are interchangeable from the supply side point of view because they are manufactured on the same machinery (94) is strongly contradicted by competitors. Whilst it is true that many parts of the textile and steel cord belt production process are similar, the crucial question is whether the manufacturers of textile belts can at short notice and without major investment switch to manufacturing steel cord belts and thereby bring competitive pressure to bear on the manufacturers of steel cord belts.

100. In response to Continental's assertion that, in order to switch production, the "only" thing that that needs to be done is to change the substrate, it must be pointed out that switching to steel cords in a textile belt production line requires costly modifications. Not only must stands for the drums of the up to a hundred or more vulcanised steel cords in the conveyor belt be bought and installed, but the steel cords must be inserted under tension and vulcanised in between the rubber membranes, which means having to acquire and install an additional tensioning apparatus. The building of a new production line for steel cord conveyor belts using an existing textile belt installation therefore requires considerable financial investment, which Continental itself puts at EUR 2 million. The manufacture of steel cord conveyor belts is therefore not only technologically more demanding and requires more space than the production of textile belts because it is necessary for production to be in a continuous plant. It also calls for specific production know-how (95) - with regard, for instance, to the manufacture of a rubber mixture which is permanently bonded to the steel cords. According to competitors, a switch within the same production line from the production of steel cord belts to that of textile belts makes sense therefore. On the other hand, it makes little sense economically to switch on a production line hitherto designed for textile belts to steel cord belt production (96).

101. Lastly, the circle of customers in the textile and in the steel cord conveyor belt sector differs considerably: the steel cord sector is generally concerned more with applications whose customers place higher demands on the qualities of the belt than in the textile belt sector. The know-how needed to market both types of belt also differs (97).

102. Since the market position of Continental and Phoenix in textile belts would be unobjectionable from a competition law point of view even after a merger, the considerations below are confined to heavy conveyor belts in the area of steel cord conveyors (98).

(bb) The geographic market for heavy steel cord conveyor belts is at most Europewide

103. In the merger notification Continental expressed the view that the market for all types of heavy conveyor belt is "to be defined as worldwide, or at least as Europe-wide". Since then, the notifying party has, however, denied the possibility of a Europe-wide market definition; it now takes the view that the market for heavy conveyor belts is to be defined in all cases as worldwide.

104. The findings of the in-depth market survey militate, however, clearly against a worldwide market. Although certain factors militate even in favour of the existence of national markets, the Commission considers the steel cord belt market on balance to be Europe-wide.

(1) Imports from outside Europe play no part in the market for heavy steel cord conveyor belts

105. European customers have a clear preference for European, and often even for national, manufacturers of steel cord conveyor belts (99). While, because of the leading role played by the two big German steel cord belt manufacturers in belt technology, conveyor belts may be exported to other regions of the world (possibly even on the basis of invitations to tender (100)), this does not alter the fact that no appreciable competitive pressure is exerted on European conveyor belt manufacturers from outside Europe.

106. All leading manufacturers of conveyor belts thus have a production plant in Europe from which they supply European customers. Imports, which stand at [0-10%]*, consist primarily of small volumes from a large number of individual suppliers who play hardly any discernable role as competitors to Continental or Phoenix.

(2) There are high and manifestly effective barriers to access to the European market from outside Europe

107. Not least because of their high weight, the transport of steel cord belts over long distances gives rise to increasing difficulties. With rolls of steel cord conveyor belts weighing up to 50 tonnes, carrying them by road by ordinary lorry is out of the question. The problem is even greater when a belt - from, say, Asia - has to be carried literally half way round the world. To the pure transport costs, which are estimated by competitors in the case of, say, export from Asia, at about [5-20%]* (101), must be added the customs duties payable on import into Europe. The import duty on steel cord conveyor belts currently amounts to 6.5% of the goods' value (102). To this must also be added interest costs during the several-week-long journey and any warehousing costs incurred in storing a minimum quantity of certain types of belt in Europe (103). All in all, it can be seen that imported belts suffer from a cost disadvantage compared with domestic belts.

108. Customers state further that they avoid non-European belts because of their poor quality (104), which is partly due to their different composition (105). Foreign belts, they say, sometimes do not possess the necessary ISO certification (106). The long delivery periods quoted by non-European competitors and the danger of possible delivery bottlenecks also put off customers from purchasing from outside Europe (107).

109. A particular market foreclosure effect vis-à-vis exports is produced in the steel cord conveyor belt field by the qualification procedure preceding any belt procurement. Steel cord conveyor belts are tailor made to the individual requirements of each conveyor. Since customers demand absolutely hitch-free operation, all large customers procure a belt (e.g. in the lignite sector) only if it has first been thoroughly tested in the laboratory and in the field. During the test phase, as well as during actual operation, customers and manufacturers work closely together. Language barriers, coming on top of the sheer distance and the unreachability of contacts, may constitute a decisive drawback when it comes to accessing the European market (108).

(3) The market shares of the main competitors in the various regions differ significantly, which indicates that no mutual competitive pressure is being exercised

110. It is also clear from the different positions of competitors in the various regions that competitive conditions in Europe differ significantly from those in other regions (such as North and South America or Asia). Goodyear and Bridgestone have thus not succeeded in transferring their strong market position in the United States to Europe (market share in Europe: [<5%]*). Conversely, Continental earns only a small proportion of its sales in the North American market (EUR [5-15]* million with steel cord and textile belts in 2003 (109)).

111. That there is actually no competitive pressure on European manufacturers from outside Europe can be seen from the fact that, despite well-publicised reports of there being huge overcapacity in the Chinese market (there is talk of excess capacity of 250 000 tonnes per year, or approximately the size of the European market), imports from China play no perceptible role. None of the customers surveyed by the Commission mentioned buying belts from Chinese manufacturers.

112. The Commission accordingly considers that the market for steel cord conveyor belts is at most Europe-wide.

(i) HEAVY FILTER BELTS

(aa) Relevant product market

113. Continental proposes that a separate product market should be defined for specialty belts. Specialty belts accordingly are heavy conveyor belts which have been further developed for specific areas of application and which are characterised by increased production expenditure and special, application-specific know-how. The only specialty belts which Continental and Phoenix produce are filter belts and pipe or roller belts. Only in the case of filter belts are there considerable additions of market share.

114. Heavy filter belts are perforated belts which serve in particular to remove the moisture from wet materials during transport. Low pressure is sometimes used (in vacuum filter plants) to draw off the liquid. There are some signs that filter belts form a separate market within heavy filter belts and should not be included in a comprehensive overall market for specialty belts. Not only is the area of application completely different to that of pipe belts and other heavy conveyor belts, but the manufacture of filter belts requires different know-how and also tools, which are necessarily different to those for the manufacture of other specialty belts. Ultimately, however, the question whether filter belts are an independent market can remain open, since even with the definition of an independent filter belt market the creation of a dominant position which would significantly impede competition in the Common market need not be feared.

(bb) Geographic market

115. The fact that filter belts are made to the individual specifications of each user seems, as in the market for steel cord conveyor belts, to indicate a Europe-wide market. On the other hand, filter belts are a niche product for which there are anyway only a few customers and manufacturers worldwide.

116. Since the competition analysis, even assuming a Europe-wide (110) market, does not lead to the creation of a dominant position as a result of which competition in the Common market would be significantly impeded, the question of the geographic market definition in this case can be left open.

2. EFFECTS OF THE MERGER

(a) AIR SPRINGS FOR UTILITY VEHICLES (OEM/OES)

(aa) After the merger, Continental/Phoenix would be by far the strongest supplier with very high market shares

117. According to the market shares communicated by Continental, Continental and Phoenix accounted for [60-70%]* of the turnover achieved in the EEA in 2003 with sales of air springs (OEM/OES) (Continental: [40-50%]*; Phoenix: [10-15%]*). The in-depth investigation largely confirmed the estimates submitted by Continental (111). To the Commission's knowledge, Phoenix is, with a market share of [15-20%]*, the second largest competitor after Continental. Other major competitors are CF Gomma and Firestone ([10-15%]* each) and Goodyear [5-10%]*.

<emplacement tableau>

118. The figures for all competitors for the years 2001 and 2002 do not differ significantly. In both years, Continental/Phoenix had a market share of [60-65%]*. An upward trend is apparent, however, in the case of Phoenix, which managed to increase its market share by three percentage points, while that of Continental [...]*.

119. The lead over the three leading competitors, CF Gomma from Italy and Firestone and Goodyear from the USA (112), would increase considerably as a result of the transaction. The merged undertaking Continental/Phoenix would be four times bigger than the next-largest competitor and would have by far the broadest customer base of all competitors.

120. A market share of over 60% already suggests that the planned takeover would give Continental a dominant position in the market. The Herfindahl-Hirschman index (HHI) would be [3 500-4 000]*, and the change [1 250-1 750]*. These figures indicate, moreover, that the merger would give the undertaking competitive room for manœuvre which the remaining competitors would be unable to limit effectively.

(bb) The merger would bring together the two leading competitors in the European market

121. Not only are Continental and Phoenix the leading and the second-largest manufacturers of air springs in Europe by market share, but they also follow each other closely in terms of essential competition parameters. This has been confirmed by a survey of customers, who are key to manufacturers' success in this respect (113). The Commission thus asked customers and competitors to rank manufacturers present in the European market in six categories ("technical competence/innovation", "quality/reliability", "customer service/support", "past experience", "price" and "spare capacity").

122. The view expressed by Continental that Phoenix is only a fairly minor competitor in the market for utility vehicle air springs is not supported by the Commission's market survey. From customers' point of view, either Continental or Phoenix ranks first or second according to most of these characteristics (114). As a result of the merger, therefore, a firm which is regarded by customers as Continental's strongest competitor would disappear from the market.

123. A similar picture emerges from the answers given by competitors to the same question. Here, Continental ranks as top firm in the four categories "innovation", "quality", "customer service" and "past experience". Phoenix numbers among the cheapest firms. Only the category "spare capacity" goes to Goodyear, followed by Continental.

(cc) The relevant markets are not traditional bidding markets

124. Continental has argued that the market for utility vehicle air springs is a bidding market and that therefore market shares such as those in the present case do not signify market dominance since, even after a takeover, there would still be at least three credible, strong competing suppliers. To back up this argument, Continental submitted a report on competition in bidding markets.

125. In the report, Continental, or Lexecon, the firm commissioned to draw it up, describes customers' purchasing procedure as a multi-stage procedure which is always set in train by a call for tenders. The competition to develop a particular model is followed by competition for the initial order then by further competition for the follow-up contract. In the Commission's opinion, the modelling assumptions made by Continental in its comments and those made in the submitted report may hold true in other bidding markets - e.g., in the motor vehicle sector. Nevertheless, the Commission's market survey has shown that, in the area of utility vehicle air springs at any rate, actual market relationships do not correspond to those portrayed by Continental.

126. In particular, the survey of market participants to enquire into their purchasing procedures (115) indicated that Continental greatly overestimates the importance of calls for tenders to the air springs market as a whole. The Commission's market survey has shown that the market for utility vehicle air springs can be described at most in parts as a bidding market. It is indisputable, however, that there is no "open" bidding market in which a transparent procedure is guaranteed whereby all bids are published or rendered accessible to every competitor and whereby every supplier has at least the same chance of taking part (116).

127. In fact, of the 18 customers whose answers the Commission analysed, nine indicated that they carry out no calls for tenders. This applies not only to smaller purchasers, where the number of items itself rules out a call for tenders, but also to larger customers. Even among those large customers who occasionally make calls for tenders, there are cases where contracts are awarded without any call for tenders. Such "calls for tenders" as there are, however, often limited in advance to a few selected market participants. Phoenix describes this state of affairs as follows:

"Larger contracts from OEM/OES usually start with a "request for quotation", however, only known partners receive such RFQ. Without close contact to the OEM/OES it does not come to attention to suppliers that a new demand is negotiated. Additionally, OEM/OES do not reveal all their decision parameters and necessary information in RFQs. (...) the whole process does not represent the typical characteristics of tenders (open communication of requirements, availability of full documentation, transparent process steps and a minimum equality of chances for all applicant, final decisions). It is hard for a new supplier to find out even basic project facts and whereabouts (117)".

128. One of the reasons for the absence of open tendering procedures also in the case of large manufacturers (118) is clearly the highly sensitive nature of the data that would be necessary for a tendering procedure. According to competitors, manufacturers are reluctant to make data on the technical details of the planned vehicle openly available (119). The Commission estimates the volume of contracts that are awarded in at least a partly tender-like form at less than 50%. In the remaining cases, customers simply contact individual suppliers and enter into more protracted negotiations which then lead to the conclusion of a contract.

129. In such a procedure the existing supplier is always at an advantage as a change of manufacturer or even just the involvement of a further supplier is always costly for the buyer (120). It is not surprising, therefore, that it is rare in this market for manufacturers to change their main supplier.

130. To the Commission's knowledge, it is accordingly not true that the majority of manufacturers re-advertise every three years or so in a kind of "regular bidder competition" their order volume for a particular model or re-place the order by benchmarking. In this respect, the basic assumption underlying the arguments advanced by Continental to characterise the market as a bidder market is false. Instead, it is success at the initial contract stage that determines as a rule the supply relationship throughout a model's production run. In answer to the question whether they worked regularly with one and the same main supplier or whether they had changed main supplier in the last three years, 72% of those surveyed stated that they worked with one and the same main supplier (121). These 13 undertakings include all the major vehicle manufacturers.

131. This finding is not surprising when it is considered that the products are predominantly developed in cooperation with the customer on a tailor-made basis. In the case of the major vehicle manufacturers, so-called development partnerships are, moreover, as indicated, the norm, whereby the personal "chemistry" between teams of engineers is paramount.

132. Continental states that an order is essentially executed in various stages: the development stage, often taking the form of a development partnership; the initial order to produce; and the placing of one or more further production orders during the course of the vehicle model's production run. Also according to Continental, however, there is a 90% chance that the supplier who secured the development order will also secure the subsequent initial production order.

133. In fact, to the Commission's knowledge, it is extremely rare for an order for even part of a production to be placed with another air spring supplier. A 70% majority of the customers surveyed indicated that the initial supplier supplies throughout the production run and that no other supplier gets a look-in (122). Only one third of manufacturers take on board a new supplier as a rule after the first three years following the introduction of a spring, but mostly only with a smaller percentage. It is therefore crucial to the success of a supplier in the market that he already succeeds in securing the development order, and, contrary to the view taken by Continental, the appointment of a development partner is also very much tantamount to a preliminary decision regarding subsequent supply contracts.

(dd) With the disappearance of Phoenix, the competitive pressure on Continental will be considerably reduced

134. The takeover of Phoenix by Continental would not only lead to an addition of market shares and production know-how, but it would also considerably reduce the competitive pressure to which Continental used to be subject. Phoenix was ranked by customers in the assessment of its strengths and weaknesses in the area of technical competence/innovation as the leading competitor of Continental and is cheaper than Continental. This scenario is also reflected in the behaviour of customers when orders are being placed, Phoenix being Continental's most common direct competitor.

Phoenix was reckoned to be a particularly innovative competitor

135. The Commission asked customers which of the six characteristics of manufacturers were most important to them. First came quality and reliability, followed by technical competence/innovation and price. From the customer's point of view, Phoenix is considered to be the closest competitor of Continental in the second of these three areas. In the area of technical competence, Continental is regarded as No 1 and Phoenix as No 2 (123). Continental and Phoenix are regarded by most of the market participants surveyed as the technologically leading, most innovative manufacturers of air springs in the OEM/OES sector. Post-merger there would be no other competitors with as broad a product range as the merging parties. For this reason, many customers have expressed concern that, as a result of the considerable market power of a merged Continental/Phoenix, price increases might take place.

Phoenix is the closest competitor if several competitors compete for a tender

136. This competitive proximity of the two merging parties is also reflected in the results of customer enquiries to air spring manufacturers and their ranking of the quotations given. The Commission has analysed the supply relationships with 18 customers (124). Nine customers placed their orders direct without calls for tenders or requests for quotations addressed to more than one manufacturer. In these cases the contract went in [...]* instances to one of the two merging parties (in [...]* cases to Continental and [...]* to Phoenix). In three cases more than one supplier was considered. In the case of the remaining nine customers, who asked more than one supplier to provide a quotation or who organised calls for tenders, a total of 15 contracts were to be awarded. In total, there were therefore 18 requests for quotations where more than one supplier was involved. In the case of nine of these 18 requests for a quotation, both Continental and Phoenix participated or were asked for a quotation. In [...]* cases, either Phoenix or Continental won and the other came second. In one case Continental came second and Phoenix third. Hence in [70-80%]* of all contracts for which both merging parties competed or in [30- 40%]* of all open calls for tenders, the merging parties ranked as closest competitors. Firestone was asked for a quotation alongside Continental only five times and was the closest competitor of Continental in only two of these cases, accounting for 11% of all requests for quotations.

No fundamental change in the competitive assessment in the event of a separate consideration of the two segments truck/bus and axle/trailer business

137. Continental suggests in its reply to the Commission's statement of objections that the market for utility vehicle air springs should be subdivided into two segments: truck and bus manufacturers, on the one hand, and axle and trailer manufacturers, on the other. As indicated above, the Commission is of the opinion that such a subdivision of the relevant market is unnecessary. If, however, this hypothesis were to be considered correct, a combined Continental/Phoenix is still, at [50-60%]*, by far the leading supplier in the axles and trailers segment. The basic competitive assessment of the planned takeover would remain unchanged.

138. Likewise in the trucks and buses segment, the competition problems would be no less significant. Contrary to what Continental states in its reply to the statement of objections, Phoenix is no "unknown quantity" when it comes to the more development-intensive business with truck and bus manufacturers. In terms of technical competence, customers rank Phoenix among the leading manufacturers. It must be taken into account in this connection that, since it started cooperating with Freudenberg within the framework of Vibracoustic, Phoenix has been contractually prevented from becoming active in this market. Phoenix has hived off its OEM/OES business in the utility vehicle air spring area to the Vibracoustic joint venture operated in conjunction with Freudenberg, retaining only the OEM/OES business with axle and trailer manufacturers. Vibracoustic, however, has focused strategically more on the truck air springs business and has remained largely inactive in the utility vehicle market. Nevertheless, Phoenix has retained from the time before the joint venture a very good reputation in relation to development competence, even if its market position has since worsened (125). There are therefore grounds for believing that market shares, according to which Phoenix occupies a weaker position here than in the case of trailers/axles, do not reflect Phoenix's true potential. Since Phoenix is receiving more and more enquiries from major manufacturers, two agreements were recently entered into with Vibracoustic to the effect that Phoenix might take on such business (126). With respect to the first of these two enquiries, Phoenix was in second place behind Continental, whereas the second procedure is still under way. It is therefore to be expected that Phoenix's market position in the years ahead would improve still further without the merger.

139. This positive trend in Phoenix's market position is also reflected in the market shares. Phoenix/Vibracoustic increased their sales from EUR [10-20]* million in 2001 to EUR [15-25]* million in 2003, i.e. by [5-10%]*, and their market share by three percentage points. Continental, on the other hand, [...]*.

Phoenix possesses a favourable cost structure on account of its production in Hungary

140. Owing to its cost structure, Phoenix is able to compete effectively with Continental. Continental has two plants producing air springs in Europe: one in Hanover and another in Turkey, a low-wage country. In 1996 Phoenix brought on stream a plant in Nyiregyhaza in Hungary, which does all of Phoenix's assembly work and provides Phoenix with a low-cost production base. The market survey has shown that Phoenix's production costs are lower than Continental's. Phoenix is able as a result to offer customers lower prices. Phoenix's cost structure is slightly more favourable than Continental's, the Turkish plant's capacity amounting to only about [...%]* of the Hanover plant's, that is to say, over [...%]* of total capacity is located in Germany, a high-wage country. Phoenix, ranking as the second-best supplier from the customer's point of view after Continental, has thus so far been best placed to exert competitive pressure on Continental, being perceived, in the light of its similar innovative strength and quality, as a low-priced alternative to Continental (127). Following Phoenix's takeover, this alternative would no longer exist.

(ee) Customers have insufficient alternatives available so as to be able to counteract the increasing market power

141. In the event of a price increase by the merged Continental/Phoenix it would in practice be impossible for customers to place an order for a fairly large quantity at short notice with competitors. First of all, it takes time for a new supplier to come on stream, and secondly competitors' available spare capacity is running too low. What is more, patents act as an entry barrier.

Restriction of choice due to the need for a qualification process

142. In the case of all major new developments, so-called development partnerships are agreed whereby engineering teams of the air spring manufacturer and of the customer jointly develop and test the air springs. Customers estimate the time needed to develop and test a new air spring at up to 24 months (128). A supplier who has not so far supplied these customers and might thus be able to shorten this qualification process must first of all go through the process. This would give Continental/Phoenix, as the market leader with the largest customer base, an enormous advantage.

143. Nor would it be possible for many manufacturers to implicate a second supplier in already existing supply contracts without further ado. In many instances, batch sizes are too small to make it economically attractive for a competitor to invest in tools and dies which could not be amortised over the remaining period. What is more, the first supplier as a rule possesses patents and/or trade mark rights which he would have to transfer to the second supplier. Continental has argued that air spring technology has matured and that patents therefore do not play a major part in the awarding of contracts. The only advantage a manufacturer derives from patents is, it claims, the resulting reputation as an innovator. The Commission's market survey has shown, however, that patents can in fact play a market-partitioning role in the awarding of contracts. They are thus perceived by customers, in the event of the awarding of a production contract to a second supplier in an ongoing project, as an obstacle (129). A competitor has given the Commission a specific example of a case where Continental tried to prevent that competitor from jointly soliciting an order from a Swedish truck manufacturer (5-series Scania trucks) on the ground of an existing property right (130).

Absence of competitors who are technically as competent and as capable of delivering

144. Even if economic or legal reasons did not stand in the way of the award of a contract to a competitor, competitors would be unable to accept a major order for reasons of capacity.

145. Firestone is an American firm which occupies an especially strong position in its US home market. It is market leader there with a 70% market share (131). Since 1998 Firestone has been present in the European market. Air springs are imported from the US and stored in a large warehouse in Holland so as to be able to meet the delivery reliability standards demanded by customers. Firestone is in the process of building a new factory in Wolsztyn, Poland, which will come on stream in 2005. The factory's capacity should, however, be absorbed to a considerable extent by existing contracts. Firestone has indicated, moreover, that its four US factories are all being used to capacity.

146. Goodyear has the smallest market share of the three main competitors. An American corporation, Goodyear owns two utility vehicle air spring plants in Europe: one in Kranj, Slovenia, and the other in Montluçon, France. Goodyear's French plant is, however intended first and foremost for the production of air springs for industrial applications. In recent years Goodyear has incurred heavy losses and is heavily indebted. The sale, announced by Goodyear's CEO, of niche businesses, including parts of its engineering division to which the air spring business belongs, might help (132).

147. The Italian firm CF Gomma has its manufacturing base in Milan and is strong in the Italian market alone, being the traditional main supplier of the Iveco group. CF Gomma is perceived by customers as a low-cost supplier with only limited technical knowhow (133). It states that its annual capacity of 600 000 items is 90% used up. Unlike Continental, Firestone and Goodyear, there is no major group behind CF Gomma. CF Gomma achieved a turnover in 2003 of EUR 431 million. The new entity Continental/Phoenix would accordingly be more than 20 times bigger than CF Gomma.

There are no new market entrants waiting in the wings

148. In the last five years there has been no new entry into the air springs market. Continental has stated that the next market entry will probably be by a manufacturer who is already active in the air springs market but who operates exclusively or primarily in the IAM business, such as the Turkish firm Aktas. So far, however, Aktas has not won any development contracts or any initial contracts in the west European market. A majority of customers have replied134 that Aktas has no, or too little, development know-how and therefore will not be considered as an OEM/OES supplier in the foreseeable future.

149. It must therefore be assumed that competitors will not be able, either individually or together, to limit the merged undertaking's freedom of manœuvre to such an extent that it does not acquire a dominant position.

The other side of the market does not have sufficient countervailing power

150. Nor, lastly, is the possibility that competition problems might arise ruled out by the fact that the merging parties' customers include several large vehicle manufacturers. The air spring manufacturers' customer base in the utility vehicle sector does not consist by any means exclusively of large, internationally active groups. In the area of axles and trailers, a whole series of small or medium-sized manufacturers whose countervailing buyer power cannot be compared to that of the large automobile groups are active.

151. Even though the openly expressed concerns of the large multinational truck and bus manufacturers are thus less serious than those of the smaller competitors, there are in addition to large utility vehicle manufacturers (135) in particular a number of smaller axle and trailer manufacturers in Europe who fear the negative impact of the planned takeover.

152. The situation is made worse by the fact that the relationship between vehicle and trailer manufacturers and suppliers has changed appreciably in recent years. The reason for this is not just to be found in manufacturers' increased dependence on their suppliers due to the widespread practice of just-in-time production. Above all, the steadily decreasing manufacturing penetration in vehicle manufacture means that vehicle manufacturers are becoming more reliant on their suppliers, to whom increasingly large parts of the production process are devolved and on whose conduct the security of production essentially depends. As a consequence thereof, the demand-side power of the manufacturers has also diminished.

153. The Commission therefore considers that the takeover of Phoenix by Continental in the OEM/OES market for utility vehicle air springs would create or strengthen a dominant position as a result of which effective competition in the Common market or in a substantial part thereof would be significantly impeded.

(b) AIR SPRINGS FOR UTILITY VEHICLES (IAM)

The market shares in the IAM area turn out to be appreciably smaller than in the OEM/OES area

154. In the independent aftermarket for air springs for utility vehicles the market position both of Continental and of Phoenix/Vibracoustic is substantially weaker than in the OEM/OES market. According to data supplied by Continental, the combined market share in terms of sales of air springs for utility vehicles in the IAM area in the EEA came to [30-40%]* (Continental: [15-20%]*; Phoenix: [20-25%]* of the sales volume in 2003). To the Commission's knowledge, the parties' market share in Europe (136) is even smaller. There the combined market share comes to only [25-30%]* (Continental: [15-20%]*; Phoenix: [5- 10%]*). Continental is therefore currently more or less on a par with Firestone [15-20%]*, Goodyear [15-20%]* and Aktas [15-20%]*.

<emplacement tableau>

155. After the merger, Continental and Phoenix would thus be the largest supplier in the market. However, in view of the moderate market shares, the creation of a dominant position seems unlikely.

There are a sufficient number of strong competitors to prevent the creation of a dominant position

156. The gap between the merged entity of Continental/Phoenix and the next three competitors would be only [5-10%]*. This (small) gap is unobjectionable from a competition law perspective because the decisive factors for success in the IAM air spring business are fundamentally different from those in the OEM/OES area. Unlike there, what matters in the IAM business is not so much technical competence in developing a spring for a new model as being able to make a product as cheaply as possible. This explains why non-European manufacturers with little experience of the European market can be much more successful than in the OEM/OES business.

157. The reproduction of a replacement spring for a particular type of truck is technologically less demanding than the joint development of a completely new spring for a new vehicle type. For that reason, there is significantly more competition in the IAM market than in the OEM/OES area, the more so as in the IAM area Continental and Phoenix do not - unlike in the OEM/OES area - enjoy a substantial lead over their main competitors in the development or manufacture of IAM springs.

158. From the customer's standpoint also, it is easier to switch supplier than in the OEM/OES market. Customers are mostly large replacement part dealers who maintain contact with a large number of spring manufacturers from whom they obtain springs. Since development partnerships play no part in the IAM business, customers in the IAM business are more willing to switch supplier during the lifetime of a model than in the OEM/OES business.

159. On the whole, there is therefore no likelihood of the creation of a dominant position in the IAM market for utility vehicle springs.

(c) AIR SPRINGS FOR CARS

(aa) After the merger the new undertaking would have a strong position in the production of air springs for cars in Europe

160. Continental and Phoenix are currently by far the largest suppliers of air springs for cars in the European market, and there is only one other supplier who sells air springs for European vehicles. Continental itself estimates the combined market share of the merging parties in car air springs in the EEA in 2003 at [>90%]* (Continental: [15-20%]*; Phoenix: [70- 80%]*). The results of the Commission's survey of market shares, which are presented in the following table, basically confirm the estimate of Continental and Phoenix (albeit with slightly higher shares for Continental).

<emplacement tableau>

(bb) The merger would remove the only current competitor of Continental and Phoenix/Vibracoustic

161. The range of choice available to air spring customers would deteriorate dramatically after a merger: whereas previously they still had a choice between two large air spring suppliers who were roughly equally strong in terms of technology (137), after the merger there would be only one supplier active in the European market, who in addition would be relatively small.

162. For actual and potential competitors, the already large gap between them and Continental/Phoenix/Vibracoustic in terms of product portfolio, customer base and technological know-how would dramatically increase. The merged undertaking would not only have easily the largest customer base in Europe; the combination of Continental's competencies in system integration (which Continental has acquired as a result of its activities in air springs and braking systems) and Phoenix/Vibracoustic's competencies in production technology and product design would create an extremely strong manufacturer who would have no equivalent competitor in Europe - perhaps even worldwide. By acquiring Phoenix/Vibracoustic's air spring production, Continental would succeed in eliminating the strongest competitor in Europe.

163. The only competitor present in the European market in 2003, the US manufacturer Gates, whose market share is only [5-15%]*, is unlikely to be in a position to offset the decline in competitive pressure on Continental occasioned by the disappearance of Phoenix/Vibracoustic. Not only may Gates's brand new production plant in Aachen still be too small for the company to build up a comparable market position to that of Phoenix/Vibracoustic, but its customer base, hitherto limited to two customers in Europe138, may scarcely be big enough for it to compete in Europe on an equal footing with Continental/Phoenix, the dominant unit after the merger. Even if, in future, Gates is increasingly considered as a supplier as a result of Phoenix/Vibracoustic's disappearance, it would take years, on account of the lengthy testing and qualification procedures, for it to build up a comparable market position.

(cc) It is doubtful whether new market entries by non-European competitors would be able to limit sufficiently the increased market strength of the merging parties

164. An entry into the car air springs market by European manufacturers which have so far not been active in the production of car air springs is in the Commission's opinion unlikely. Truck air spring manufacturers - such as CF Gomma - do not have the necessary production know-how for success in the car air springs market to become a serious competitor for the new unit in the near future. The production know-how for commercial vehicle air springs differs not inconsiderably from that which is necessary for the successful production and sale of car air springs (e.g. production using the sleeve design, qualification with key customers in the car sector, competence as regards system integration, etc.).

165. The possibility cannot be ruled out, however, that the disappearance of Vibracoustic might induce car makers to turn increasingly to foreign manufacturers for their supplies. There are in fact signs that, for example, the US manufacturer Firestone might in future become more active in the European market (139). Nevertheless, the likelihood of new market entries is reduced somewhat by the market growth - characterised by the parties as sluggish - coupled with a relatively small market volume (about EUR 30 million in 2003 (140)).

(dd) It is not certain that the other side of the market has sufficient countervailing power to effectively contain Continental's growing room for manœuvre

166. Continental rightly states that some air spring customers are large automobile firms which in many areas have sufficient countervailing buyer power vis-à-vis their suppliers. However, this applies to only that part of production which is sold direct to the car manufacturers and not to the sometimes smaller system manufacturers. It is by no means certain, moreover, that the car manufacturers can in fact effectively contain the room for manœuvre enjoyed by the merging parties as a result of their near monopoly in the European market. For example, the production of rubber bellows requires quite specific know-how which car manufacturers do not themselves have. This eliminates the possibility which exists in the case of many other products of being able to switch to in house production. The present case differs in this respect from other cases in the automotive supply industry. Even relatively large automobile manufacturers have accordingly clearly expressed their concern about the merger (141).

(ee) No need for a decision on account of the commitment offered by Continental to divest the participation in Vibracoustic

167. Lastly, the question whether the merger creates a dominant position in the market for car air springs too can be left open because of the commitment offered by Continental to divest the participation in Vibracoustic to Freudenberg. The Commission's competition concerns in the car air springs area are removed - contrary to the view expressed by Continental - not because Freudenberg will in the absence of such a commitment exercise its call option and in the event of an acquisition of Phoenix by Continental purchase its holding in Vibracoustic. Even if such a development may seem likely, it should be pointed out that the exercise of the call option by Freudenberg after the merger is by no means certain. Inasmuch as it is perfectly possible that Freudenberg might not exercise its call option and that the car business might remain under the joint control of Phoenix/Continental, the competition analysis cannot simply assume a cessation of the car air springs business. The Commission must rather, in the interests of safeguarding competition, examine whether the merger raises no competition problems also in the event of Freudenberg not exercising its call option.

168. The commitment given by Continental to divest the participation held by Phoenix in Vibracoustic to Freudenberg, with which the divestiture of Phoenix's businesses in the car area is also linked, completely removes any competition problems in the car air springs area, with the result that there is no longer any need to examine the question whether the merger might have led here, too, to a dominant position. The divestment of the Vibracoustic participation to Freudenberg is a precondition for the Commission's exemption decision in the utility vehicle air springs (OEM/OES) area (142).

169. Through the transfer to Freudenberg of the 50% shareholding held by Phoenix in Vibracoustic, all potential competition problems in the car air springs area are removed. Phoenix does not have any separate car air spring production, for which before Vibracoustic was formed Freudenberg, then Vibracoustic alone, was responsible. Through the transfer of the participation in the joint venture, Phoenix loses all activities in the car air springs area, since not only production but also research and development as well as marketing activities were centred on Vibracoustic. As a result, there would be no addition of market shares of relevance to competition in the car air springs area.

(d) AIR SPRINGS FOR RAILWAY VEHICLES

(aa) The merger of Continental and Phoenix would lead to relatively large market shares in the rail air springs sector

170. Continental estimates the combined market share of Continental and Phoenix in the railway vehicle air springs sector, assuming the market is worldwide, at [20-25%]*, and assuming the market is Europe-wide, at [35-45%]* (Phoenix: [20-25%]*; Continental [15-20%]*). In its market investigation, however, the Commission arrived at higher market shares for the Europe-wide market of approximately [55-65%]*. The results of the market survey are shown in the following table, which is based on the (confidential) information supplied by the main competitors.

<emplacement tableau>

(bb) Despite currently high market shares, there is no likelihood of a dominant position being created in the railway sector

171. The high market shares shown in the table are not sufficient, in the Commission's view, to create a dominant position in the market for air springs for railway vehicles. For one thing, the manufacturers of secondary rail vehicle air springs are much more dependant on cross supplies than in the case of other products. And for another, railway customers have sufficient opportunities, not least because of the long product cycles, to cultivate other suppliers of rail vehicle air springs.

Major importance of cross supplies in the case of rail vehicle air springs

172. The market for air springs is characterised by the fact that none of the air spring manufacturers participating in the air springs business is capable of manufacturing on its own all the metal, rubber-metal and rubber parts needed for these extraordinarily complex air springs and of thus acting independently. The Commission accordingly agrees with the argument put forward by Continental to the effect that the railway vehicle air springs sector is characterised by a mutual dependence of manufacturers on supplies of individual components from other manufacturers and that one manufacturer's product portfolio complements that of another. The reason for this may lie in the fact that the air spring manufacturers have moved into the air springs sector from various component sectors, some from the rubber-metal sector and others from the rubber bellows sector.

173. The companies currently active in the production of rubber bellows (Continental, Phoenix, CF Gomma and Paulstra/Hutchinson) have to purchase a large proportion of the rubber-metal parts used in the production of air springs from other manufacturers - mostly other competitors in the air springs sector such as Trelleborg. Other competitors, such as Freudenberg/Schwab or Trelleborg, are traditionally strong in the rubber-metal area and purchase rubber bellows from other competitors. The principle of cross supplies is therefore fundamental to the functioning of this industry.

174. Despite the comparatively strong market position of Continental in the rubber bellows sector, to the Commission's knowledge there is no danger of strong manufacturers of a component being able to exploit their position with a view to preventing competitors from supplying products. Continental thus points out that it recently concluded a contract for the supply of rubber bellows to the British firm Ferrabyne, thereby enabling the latter to become active in the railway vehicle air springs sector and enter into competition with Continental. Similar examples can be cited for Phoenix.

175. The slightness of the risk that Continental and Phoenix might be able to exploit their comparatively strong position in the rubber bellows sector is especially apparent from the relative value of the rubber bellows in a rail vehicle air spring. Between 50% and 70% of the components of an air spring are metal and rubber-metal parts, whereas the rubber bellows account for only one third of the value. An air spring costs on average about EUR 1 200. Of this, about EUR 400 is accounted for by the rubber bellows. Firms such as Continental therefore have to purchase considerable quantities of rubber-metal parts in order to be able to supply an air spring. The risk of conduct independent of customers and competitors is thereby reduced.

Even after the merger, customers would retain the possibility of switching to other suppliers

176. To the Commission's knowledge, even after the merger railway customers would have at their disposal sufficiently experienced competitors to whom they could switch in the event of a price increase.

177. It must be borne in mind in particular here that rail vehicles have a very long life span. Wagons, in particular, can be used without difficulty for up to 30 years. Since railway companies carry out maintenance and repairs to their vehicles as a rule themselves, they purchase the necessary replacement springs when they place the initial equipment order. Consequently, there is no separate IAM market. The long product cycles mean that railway companies place new orders at substantially longer intervals than, say, purchasers of utility vehicle air springs. The lead times for the development of a new air spring for a new rail vehicle are also correspondingly much longer.

178. In view of these long lead times, it is - as the past has shown - even possible for a planned model to be assigned to a new supplier which has not yet been active in this market. The competitor Paulstra/Hutchinson, for example, was expressly induced by the French railway company SNCF to enter the market for air spring bellows. Although Paulstra/Hutchinson was not previously active in rubber bellows production, the long development period enabled it to acquire the necessary production know-how and build a corresponding assembly line. Paulstra/Hutchinson now supplies the air springs for the high-speed TGV train (including the corresponding rubber bellows). It must be borne in mind that the production of air springs for high-speed trains is far more demanding than the manufacture of the more simple springs for underground trains or trams. The severalyear- long lead time up to qualification by the SNCF as supplier for the high-speed train may therefore serve more as an extreme example.

179. Although Paulstra/Hutchinson has been active in the market for rail vehicle air springs for a comparatively short time and has so far operated predominantly in France, it has already secured a few orders from customers outside France and, with the TGV contract under its belt, can count on securing further orders both in France and elsewhere.

180. Another alternative for customers of Continental and Phoenix is to turn to the suppliers Toyo and Sumitomo, which have been active in this product market for a long time. Both of these manufacturers are suppliers for the Japanese high-speed train, the Shinkansen. Toyo is the Japanese market leader for air springs and is able to supply the entire air spring including the rubber bellows. It cooperates with a company called Gadelius which markets Toyo air springs in Europe. The main customer is Bombardier, to which Gadelius supplies, among other things, air springs for the X2000 tilting train. In Europe, Toya therefore not only supplies technically less demanding solutions (e.g. in the underground railway sector), but it is also known for technically more costly projects. In addition, Toyo also already supplies rubber bellows to at least one competitor in the context of the above-mentioned cross supplies.

181. The Japanese manufacturer Sumimoto has likewise won several contracts in the underground railway sector, both in Spain and in the United Kingdom. It also supplies the air springs for a passenger train in Ireland. To the Commission's knowledge, other manufacturers are also in a position to become active in Europe within a relatively short space of time.

182. A contribution to the further opening-up of the European market has been made in recent years by the increased setting of uniform European standards in the rail vehicle air springs sector. As a result of the standardisation of rail vehicle air spring requirements within Europe, the previously less transparent regional differences between individual national systems are being ironed out, thereby increasing the transparency and hence the attractiveness of the European market. It is therefore highly likely that Japanese air spring manufacturers in particular will be able to expand their market positions in Europe in the near future.

Conclusion

183. For the reasons given, it is unlikely that through the merger a dominant position of the parties in the market for rail vehicle air springs would be created or strengthened, as a result of which effective competition in the Common market or a substantial part thereof would be significantly impeded.

(e) ANTI-VIBRATION SYSTEMS FOR MOTOR VEHICLES

184. In an overall market, defined as being Europe-wide (143), for anti-vibration systems, Continental and Phoenix would, according to data supplied by the merging parties, have a combined market share of not more than [15-20%]* (Continental: [10-15%]*; Phoenix/Vibracoustic: [0-5%]*) in terms of sales volume in 2003. The two merging parties would therefore still lag behind competitors ZF Boge (19.6%) and Trelleborg (16.5%).

185. Even if further submarkets were to be defined within the Europe-wide market for antivibration systems for motor vehicles, the parties would have higher market shares only in the engine mounting area (combined share in terms of sales volume in 2003: [15-20%]*; Continental: [10-15%]*; Phoenix/Vibracoustic: [5-10%]*; the share of the largest manufacturer, Trelleborg: 22.1%). These market shares are not likely to create a dominant position on the part of the merging parties.

(f) EXPANSION JOINTS

186. Even if the market for expansion joints were to be defined as Europe-wide (144), the market share of Continental and Phoenix would, according to data supplied by the merging parties, come to [25-30%]* (sales volume in 2003: Continental: [15-20%]*; Phoenix: [10-15%]*; other competitors: Woco: 13%; Trelleborg: 11%; CF Gomma: 9%). Despite the gap with competitors, this market position is in the Commission's opinion, in view of the existence of a sufficient number of worthy competitors, unlikely to lead to the creation of a dominant position in the expansion joint market.

(g) MOTOR VEHICLE HOSES AND HOSE LINES

187. In the overall European market (145) for motor vehicle hoses and hose lines, the combined market share of Continental and Phoenix in 2003 came to [15-20%]* (Continental: [10- 15%]*; Phoenix: [<5%]*; turnover shares for 2003). Even if separate submarkets were to be defined, the merger would not lead to any market position on the part of the merging parties in any of the submarkets that might be objectionable from a competition law perspective. Even if a separate market for coolant, heating and turbo hoses were to be defined, Continental and Phoenix would have a market share of under 20% (Continental: [5-10%]*; Phoenix: [10-15%]*). The merger would therefore not lead to the creation of a dominant position either in the market for motor vehicle hoses and hose lines or in any possible submarkets.

(h) HEAVY STEEL CORD CONVEYOR BELTS

(aa) The merger of Continental and Phoenix would lead to extraordinarily large market shares

188. The Commission's research among competitors and customers into the actual volume of steel cord production in Europe showed that the total volume of the market was substantially less than assumed by Continental. In particular, Continental seems significantly to overestimate the sales figures of substantial competitors and imports into Europe from the Far East. The market shares determined by the Commission can be seen from the following table:

<emplacement tableau>

189. It is clear from the table that the merger of Continental and Phoenix would bring together by far the largest competitors in Europe. It would lead to a combined market share of over 70%. Even if one included the Metalcord/Metaltrans belt types produced by Sempertrans among steel cord belts (which however, in view of the clearly different product characteristics, is rejected by the manufacturer itself (146)), the parties' combined market share would still be [60-70%]*; Sempertrans in that case would still account for only [20-30%]*. In any event, therefore, a commandingly strong supplier would be created in the European market.

(bb) The market structure would clearly deteriorate, since the gap with the nearest competitors would drastically increase

190. The market share table also makes clear that the already considerable gap between Continental and Phoenix and their nearest competitors (e.g. Sempertrans or Bridgestone) would increase considerably. With the disappearance of the strongest competitor, the degree of concentration (HHI) in the steel cord conveyor belt market would increase by more than [2 000-3 000]* points. Many of the remaining competitors, moreover, are relatively small manufacturers which only operate regionally and can only supply fairly small quantities.

191. The gap with the nearest competitors is thus so considerable that a mere glance at the very unbalanced market structure suggests that the merging parties will have room to manœuvre independently. This assumption is confirmed by closer examination of the market structure.

cc) After the merger, customers will have no or only a few fall-back possibilities

192. The range of choice for conveyor belt customers will be considerably restricted by the merger. As with other technical rubber products, steel cord conveyor belts are not goods which one can order "from the catalogue" as it were, but products whose production and sale require considerable know-how and close cooperation with the customer. For various reasons, customers cannot simply switch to other suppliers.

Qualification procedures as entry barriers for new suppliers

193. Conveyor belts are regularly ordered from a manufacturer only if the product has proved to be of high quality and technically reliable. A new supplier can get a look-in only if it has passed the appropriate qualification procedure of the manufacturer concerned. Such a qualification procedure is needed because the reliability of a conveyor belt, particularly in the steel cord sector, is extremely important. Although, in such cases, heavy and occasionally sharp goods sometimes have to be transported over several kilometres, there must be as little damage as possible throughout the working life of steel cord conveyor belts (up to 25 years). Even relatively small tears make it necessary to halt the entire production plant, which in lignite mining for example leads to considerable costs. In particular, customers want to avoid relatively large longitudinal tears, which would make it necessary to replace the whole belt.

194. The requirements for a belt vary according to the type of application and the area of use. Competitors describe the qualification procedure as "very complicated (147)". A manufacturer must therefore first prove to each customer that its belt meets the specific requirements for the customer's application. In the lignite sector, which accounts for more than half the turnover, the large manufacturers' qualification procedure comprises several stages: first, the manufacturer must ensure that the customer will allow test belts to be run on his plant at all. Even this step is a considerable hurdle for many manufacturers, since some large lignite customers accept such manufacturers as test belt suppliers only if they have already passed a previous laboratory test (148). Here the belt's characteristics (resistance to impinging materials, environmental influences and corrosion) are first tested in the laboratory. Only when this has been conducted, which can take more than a year, do manufacturers then order, where appropriate, test belts for the subsequent second stage of the qualification procedure, the trial phase.

195. Customers are not free, however, to order any quantities of test belts they like from different manufacturers. [...]*(149). It is to be feared, consequently, that the parties would use their future dominant position to conclude more such exclusive contracts (150).

196. In any event, the exclusive contracts prevent other competitors from gaining access with test belts to customers. If the manufacturers nevertheless manage to be accepted at a customer test plant, one or more test belts are then tested under practical conditions. This trial phase can take several years (151). Only then is a supplier considered at all as a supplier of relatively large quantities of steel cord conveyor belts.

197. How difficult it is for new competitors to pass the qualification procedure is shown by the fact that even today not all large competitors of Continental and Phoenix have qualified with the most important customers and have, without exception, only supplied test belts. RWE explains: "Even relatively large suppliers of steel cord belts are sometimes not able to meet the necessary requirements as regards quality and reliability at the first attempt. Thus [one manufacturer]* has so far failed to qualify with RWE, because the embedded steel cords did not pass the corrosion test at the first attempt." (Auch größere Anbieter von Stahlseilgurten sind z.T. nicht in der Lage, die notwendigen Anforderungen hinsichtlich Qualität und Zuverlässigkeit im ersten Anlauf zu erfüllen. So ist z.B. [ein Hersteller]* mit der Qualifikation bei RWE bisher gescheitert, da sie eingebetteten Stahlseile den Korrosionstest im ersten Anlauf nicht bestanden haben (152).)

Many competitors do not satisfy the customers' quality requirements for steel cord conveyor belts

198. The above quote shows that Continental and Phoenix's competitors, which have only a small market share in Europe, are not competing with the merging parties on an equal footing, at least in the lignite mining customer segment. Customers give as the reason for the particularly strong position of the two remaining German conveyor belt firms that steel cord conveyor belt technology was developed in Germany. The technological lead of German manufacturers can probably also be attributed to the fact that in mining outside Europe other transportation techniques (in particular, conveyance of materials by lorry) are mostly used. Continental and Phoenix are actually working together with, for instance, the University of Hanover, which has developed special expertise in this field (153). Knowledge of the right material composition and other parameters of product and production technology are decisive, however, in being able to supply the customer with a high-quality product. Even competitors confirm that Continental and Phoenix have outstanding product and production know-how in the field of steel cord conveyor belt technology (154).

199. Accordingly, the Commission found in its market survey that many of the small and even large competitors cannot easily meet the quality requirements (in regard for instance to tolerances, load-bearing capacity, corrosion resistance, etc.). Customers (155) and competitors (156) in equal measure have reported problems with qualification in respect of belts which do not originate with Continental or Phoenix.

200. Since at least in the case of steel cord conveyor belts for lignite customers commodity products are not involved, but instead the product can be sold only after successful qualification, the fact that all manufacturers possess spare production capacity plays only a minor role as far as the competition assessment is concerned. It is not spare capacity, but the ability to satisfy lignite customers' stringent technical requirements, that is the key to success against the competition here. As indicated, Continental and Phoenix enjoy a clear advantage over their competitors in this respect.

201. All main competitors have, moreover, qualified - if at all - with the leading large customers only with individual, specific test belts. Only Continental and Phoenix have gone through the qualification process for all the important types of belt. Especially in the case of wide belts, both companies possess a de facto monopoly in Europe.

De facto monopoly in Europe in belt widths over 2.40 m

202. Changing to other suppliers is made more complicated above all by the fact that Continental and Phoenix are the only manufacturers who are able to make the entire spectrum of required belt widths in Europe. [...]*. The other competitors only own presses for belt widths of up to 2.40 m in Europe. To succeed with conveyor belt customers in the lignite sector, however, it is particularly important to be able to make conveyor belts over 2.40 m wide (157). Thus, according to information from the largest steel cord conveyor belt customer, RWE, two thirds of all the belts it uses come within the ST 4500 (2.80 m) category (158). At the second largest customer, Vattenfall Europe, the figure is still 60% (159).

203. The next competitor too, Sempertrans, is very much at a disadvantage when competing with the merging parties: it lacks access to large customers because its belts do not qualify and because it has no widebelt capacity. To build one's own production line (160) for wide conveyor belts of between 2.40 m and 3.20 m, however, requires, the Commission understands, large investment of up to EUR 18 million for a completely new production line (161). These investment costs are all the more prohibitive for competitors as the European market (unlike those in other regions) is no longer growing but stagnating or shrinking (162).

204. As a result of the lead in product quality and product range, the number of competitors who are in the same league as Continental/Phoenix as regards competitiveness is thus clearly declining, which will further increase Continental and Phoenix's room for manœuvre after the merger.

New market entries in the wide conveyor belt sector which could limit the parties' room for manœuvre should not be expected

205. Because of the barriers described, no new market entries should be expected which could appreciably restrict the room for manœuvre of Continental and Phoenix. The considerable costs and logistical problems (163) involved in setting up conveyor belt production in Europe are all the more prohibitive as, again according to Continental, no market growth can be expected in Europe in the coming years. According to competitors, given the considerable investment costs of setting up one's own production in Europe, "it cannot be expected that competitors will commit more strongly to Europe after the merger" (nicht damit zu rechnen, dass Wettbewerber sich nach dem Zusammenschluss verstärkt in Europa engagieren (164)).

206. In the view of RWE Power, the largest customer in the steel cord conveyor belt sector, it is "pretty well impossible for a new supplier to qualify with RWE just like that without many years' experience in the steel cord/mining industry, if it [has] not previously worked with RWE" (nahezu ausgeschlossen, dass ein neuer Anbieter ohne langjährige Erfahrung im Stahlseil-/Tagebaugeschäft sich bei RWE aus dem Stand heraus qualifiziere, wenn er bisher nicht mit RWE zusammengearbeitet [hat]) (165).

207. The new market participants given by Continental in the notification documents are accordingly regional suppliers, or suppliers specialising in non-European regions, of belts of ordinary quality, which in Europe are achieving sales only in the de minimis sector (166).

208. Even taking into account that a large part of deliveries in the lignite segment are anyhow fixed by long-term contracts, it seems in view of the commanding market position of the merging parties especially in the lignite segment unlikely that sufficiently strong competitors will establish themselves by the time new orders are placed.

209. It must be borne in mind that the merging parties' dominant position in the lignite segment gives them clear advantages also in the segment of business with so-called project customers. [...]*.

Given the commanding market position of the merging parties, the customers' buying power is not enough to restrict the room for manœuvre resulting from the merger sufficiently

210. Continental rightly asserts that many of the large steel cord belt customers are large mining or energy companies which have large purchasing volumes. It is in fact correct that [70-80%]* of Continental's turnover is accounted for by the three customers in the lignite sector. Phoenix, similarly, generates [60-70%]* of its turnover from only five large customers.

211. [...]* However, in the Commission's view, these contracts are not a sign of lively competition in the conveyor belt market. [...]*. The security of production thus achieved for the belt factories and the cessation of price competition from the merging parties' competitors during this period justify the price reductions from an operational standpoint. The degressive prices are the operational penalty for maintaining several years' production security and for not being exposed to any pricing pressure from competitors during that period.

212. It is by no means clear, moreover, whether the price reductions achieved in current competition between Continental and Phoenix will also be available to customers in the next contract negotiations. In any event, after a merger, the only substantial competitor able to exert price pressure on the other one in each case will have disappeared.

213. Lastly, it should be noted that the apparent buying power of the large customers will be considerably restricted by the fact that there are currently not enough suppliers of equivalent quality for the products they each need, which limits their bargaining room. Furthermore, both Continental and Phoenix supply a not inconsiderable part of their production to less important customers as well, who only buy relatively small quantities. Since, likewise, there are often not enough equivalent-quality competitors for these customers' products, their buying power too is unable to restrict the room for manœuvre that Continental will have won from the merger.

(dd) The merger creates or strengthens a dominant position for Continental and Phoenix in the market for steel cord conveyor belts

214. Given the negative change described in the decisive parameters for the competitive structure, it should be assumed that Continental and Phoenix gain room for manœuvre through the merger, which they would not have if they remained in competition with each other. Numerous market participants have pointed out to the Commission that the merger leads to a situation where Continental/Phoenix is actually the sole supplier for many customers - especially in the lignite segment - since the disappearance of the main competitor would significantly strengthen the competitive position of the new undertaking, especially in price negotiations (167). The Commission assumes therefore that the merger creates or strengthens a dominant position of the merging parties in the market for heavy steel cord belts, as a result of which effective competition is significantly impeded in the Common market or a substantial part thereof.

(i) HEAVY FILTER BELTS

The merger would lead to nominally high market shares in the "niche market" for filter belts

215. In the market for filter belts, the merger would, according to data supplied by Continental, lead to nominally very high market shares for the parties ([80-90%]* in the EEA, of which [60-70%]* accounted for by Continental and [20-25%]* by Phoenix; turnover volume shares for 2003 with a total turnover volume for Europe of EUR [...]*). Worldwide the parties would still have a share of [40-50%]* of turnover volume for 2003. The Commission takes the view, however, that the currently strong position in the filter belt market would not lead to the creation of a dominant position on the part of the merging parties.

To the Commission's knowledge, a dominant position is not to be feared owing to the special situation in the filter belt market

216. It should be pointed out first of all that the filter belt market is an extremely small market in which - not least because of the small market volume - only a small number of suppliers currently operate. Thus in 2003 only one other supplier was active in the filter belt market in Europe. Of decisive importance to the Commission's competition assessment in such a case is therefore the question whether it is possible for other manufacturers of heavy steel cord belts to enter the filter belt market without considerable investment costs. The Commission is convinced that that is the case here.

217. In order to manufacture a filter belt, all that is in fact normally needed is one or more mouldings which are placed in the vulcanising press, with the help of which holes are "vulcanised into" the - otherwise conventional - heavy steel cord belt. Continental points out that the necessary know-how constitutes no major obstacle to new market entrants as it is generally available to all manufacturers with experience in the manufacture of heavy steel cord conveyor belts. Competitors have in fact confirmed that they are in principle able to manufacture specialty belts such as filter belts without major difficulties or investment costs (168).

218. Consequently, even assuming that the market is Europe-wide, there is, in the Commission's opinion, at least sufficient potential competition to prevent the creation of a dominant position in the filter belt sector.

VI. COMMITMENTS ENTERED INTO BY THE NOTIFYING PARTIES

219. To remove the Commission's competition concerns Continental made the following commitments, the full text of which is given in the Annex.

220. Continental basically undertakes:

. (a) on the basis of a preliminary agreement, which was concluded on 1 October 2004, to sell the 50% share in the joint venture Vibracoustic, which is currently held by Phoenix, to Freudenberg;

. (b) to sell the entire OEM/OES air spring production (both air springs for utility vehicle and bus manufacturers and for axle and trailer producers) in the Hungarian plant in Nyíregyháza to Freudenberg/Vibracoustic. This business comprises essential parts of Phoenix's plant in Hungary (e.g. personnel and machinery) and the entry into existing contracts with customers and a supply agreement ensuring that Freudenberg will continue to be supplied with the necessary semi-finished products for production.

. (c) on the basis of a preliminary agreement concluded on 8 October 2004, to sell to the competitor Semperit/Sempertrans a production line for steel cord belts up to 3.20 m wide, including a vulcanisation press for steel cord conveyor belts up to 3.20 m wide, which is jointly owned by Continental and Phoenix.

VII. ASSESSMENT OF THE NOTIFIED OPERATION UNDER ARTICLE 2 OF THE MERGER REGULATION, TAKING THE COMMITMENTS INTO ACCOUNT

1. Air springs for utility vehicles (OEM/OES)

221. The commitment by which the competition concerns about air springs for utility vehicles will be removed consists of two parts: firstly, Continental undertakes to transfer Phoenix's 50% share in Vibracoustic to the other partner, Freudenberg. Secondly, it promises to transfer Phoenix's entire OEM/OES air spring production for utility vehicles in Hungary to Freudenberg. In the Commission's view, these commitments are likely to dispel all the concerns about utility vehicle air springs.

222. With the takeover of the 50% share of Vibracoustic, not only will the corresponding research and development activities of Vibracoustic in the utility vehicle OEM/OES sector go to Freudenberg. The restriction of Vibracoustic to OEM/OES business with truck and bus customers hitherto effected by the joint venture agreement will also be removed. Vibracoustic, which from now on is controlled only by Freudenberg, is free to operate in the axle and trailer sectors and the IAM as well. Since Vibracoustic already has many years' sales experience in the OEM/OES sector, it is to be expected that the ability to offer the whole air spring range will put Vibracoustic in a position to exist as a long-term viable competitor in the market.

223. The second part of the commitment, the transfer of Phoenix's utility vehicle air spring production (OEM/OES) in Hungary to Freudenberg/Vibracoustic, simultaneously removes the previous weakness of Vibracoustic, which hitherto had no air spring production of its own. By the commitment to transfer to Freudenberg/Vibracoustic that part of the Hungarian plant where the utility vehicle air spring production is located, Freudenberg/Vibracoustic is receiving not only a modern air spring production of its own. Production is also located in a cheap area, which will probably be of great significance for the future success of the owner in the air spring market.

224. Continental has already concluded a preliminary agreement recorded by a notary with Freudenberg/Vibracoustic, in which Continental undertakes to transfer the ownership of all the necessary means of production (machinery, including a calender). Also agreed is the transfer of all existing supply contracts with customers to Freudenberg/Vibracoustic, leaving the necessary personnel for production and development, making available all the necessary know-how for operation and the transfer of all supply contracts concerning necessary semi-finished products to Freudenberg/Vibracoustic. The contractual obligations described are sufficient in the Commission's view to ensure the long-term presence of Freudenberg/Vibracoustic in the market.

225. Crucial to this assessment is, in particular, the fact that the commitment will probably completely remove the increase in market share resulting from the merger in the market for utility vehicle air springs (OEM/OES). The entire production of Phoenix, including existing supply contracts in this sector, is transferred to Vibracoustic. As a result, Freudenberg/Vibracoustic will be able to extend its previous area of activity beyond its present scope to the axle and trailer segment and to the independent aftermarket. The commitment is thus likely to remove the concerns expressed by the Commission regarding the creation of a dominant position on the market for utility vehicle air springs.

2. Steel cord conveyor belts

226. By the commitment, lastly, to sell to a competitor a production line for steel cord conveyor belts up to 3.20 m wide, including a suitable vulcanisation press, Continental is quite significantly facilitating market entry for the buyer, Semperit/Sempertrans, in this segment which is so important for success in the steel cord belt market. As already explained in the competition analysis, the Commission's concerns in the steel cord belt market relate in particular to the lignite conveyor belts segment. Customers here have unanimously stated that the ability to manufacture wide belts as well is very important for their decision to deal with a particular belt supplier. After the takeover of Phoenix by Continental, the merged undertaking, as has been shown, would have had a monopoly on this type of belt in Europe. Because of the high investment costs of a new production line for belts of up to 3.20 m in width when at the same time the market volume is stagnant, a new entry into this market should not in all probability be expected. A particular bottleneck which competitors have described here is the need to acquire a special vulcanisation press.

227. The buyer of the plant, the main competitor in the steel cord belt sector Semperit/Sempertrans, has confidently stated that by buying such a production line it can succeed in making up the technological difference vis-à-vis the merging parties and manage successfully to supply wide steel cord conveyor belts to customers from lignite mining. The ability to offer wide belts could act as a "lever", so that the firm is increasingly considered as a supplier of other belt widths as well. The Commission is persuaded that the commitment would enable the competitor Semperit/Sempertrans, which has already made considerable progress in the qualifying process with most customers, to compete on a permanent basis with Continental/Phoenix. The market survey carried out by the Commission in the light of the commitment offered has confirmed this view. All of the customers who responded are of the opinion that the commitment offered by Continental to divest a production line for wide belts is likely to eliminate the competition problems in the area of steel cord conveyor belts.

228. By making sure there is an opening in the most important segment of the steel cord conveyor belt market - lignite belts - the commitment is likely to dispel altogether the Commission's concerns that the merger could lead to a dominant position in the market for heavy steel cord conveyor belts.

229. The commitments presented by the notifying party thus ensure that the notified merger does not lead to the creation of a dominant position in the markets for air springs for cars, rail vehicles and utility vehicles (OEM/OES) and in the market for heavy steel cord conveyor belts as a result of which competition in the Common market would be significantly impeded.

VIII. CONDITIONS AND OBLIGATIONS

230. Under first sentence of the second subparagraph of Article 8(2) of the Merger Regulation, the Commission may attach to its decision conditions and obligations intended to ensure that the undertakings concerned comply with the commitments they have entered into vis-à-vis the Commission with a view to rendering the concentration compatible with the Common market.

231. Measures that change the structure of the market are to be made the subject of conditions, but the implementing steps which are necessary to achieve this result are to be made the subject of obligations on the parties. Where a condition is not fulfilled, the decision by which the Commission declared the merger to be compatible with the Common market no longer stands. Where the parties commit a breach of an obligation, the Commission may revoke the clearance decision under Article 8(5)(b) of the Merger Regulation; the parties may also be subject to fines and periodic penalty payments as provided in Article 14(2)(a) and 15(2)(a) of the Merger Regulation (169).

232. In accordance with this basic difference, the Commission makes its Decision conditional on the complete fulfilment by the parties of the commitments to:

• (a) sell the shares held by Phoenix in Vibracoustic to the partner Freudenberg (Section B 1 of the Annex);

• (b) sell the air springs business in the Hungary plant (utility vehicles OEM/OES) to Freudenberg/Vibracoustic (Section B 2 of the Annex);

• (c) sell to Semperit a steel cord conveyor belt production line with a vulcanisation press for the production of 3.20 m wide belts belonging to Simpelkamp (Section B 3 of the Annex).

233. These commitments alter the structure of the market. The remaining commitments are to be made the subject of obligations, which will basically safeguard the competition effects of the above-mentioned conditions or ensure that they are implemented.

IX. CONCLUSION

234. For these reasons it can be assumed that, subject to full compliance by the notifying parties with the obligations entered into, the proposed merger does not create or strengthen a dominant position as a result of which effective competition would be significantly impeded in the Common market or a substantial part thereof. The merger should therefore be declared compatible with the Common market and the EEA Agreement subject to full compliance with the obligations contained in the Annex under Article 2(2) and Article 8(2) of the Merger Regulation and Article 57 of the EEA Agreement,

Has adopted this decision:

Article 1

The notified merger, whereby Continental acquires sole control over Phoenix, shall be declared compatible with the Common market and the EEA Agreement.

Article 2

Article 1 shall apply provided Continental completely fulfils the commitments made in Sections B.1, B.2 and B.3 of Annex A.

Article 3

This Decision is issued subject to the complete fulfilment of the other commitments made by Continental mentioned in Annexes A and B and Schedules B.1 and B.3.

Article 4

This Decision is addressed to: Continental AG

Vahrenwalder Straße 9

D-30165 Hanover

Germany

Notes

1 OJ L 24, 29.1.2004, p. 1.

2 OJ L 395, 30.12.1989, p. 1; corrigendum: OJ L 257, 21.9.1990, p. 13.

3 OJ L 180, 9.7.1997, p. 1.

4 OJ C ., ..2004, p. ..

5 OJ C ., ....2004, p. ....

6 Turnover calculated in accordance with Article 5(1) of the Merger Regulation and the Commission notice on the calculation of turnover (OJ C 66, 2.3.1998, p.25). Any turnover achieved before 1 January 1999 was calculated using average ECU exchange rates and then converted in EUR at the rate of 1:1.

* Parts of the original decision are omitted in the public version to ensure that confidential information is not disclosed; those parts are enclosed in square brackets and marked with an asterisk.

7 Phoenix distributes air springs for cars and some air springs for utility vehicles (springs for tractor cabs and buses in the OEM/OES area) through a joint venture ("Vibracoustic") operated with the firm Freudenberg, Weinheim, in which Phoenix has a 50% holding and exercises joint control. The relevant market shares are accordingly allocated to Phoenix here. Vibracoustic has its own production for car air springs, but not for heavy goods vehicles or for rail springs, which are supplied from Phoenix's plant in Hungary.

8 In view of the substantial differences between suspension systems for heavy goods vehicles, cars and rail vehicles, these areas must be assessed separately in each case. Nor did Continental suggest a Common market for these various vehicle types (cars, rail vehicles, utility vehicles).

9 Manufacturers supply not only complete air springs (bellows, clamping plate, piston, known as "full assembly"), but also sell individual rubber bellows or rubber bellows with piston only ("service assembly"). However, it does not appear necessary to examine different markets for the different types of design separately, since rubber bellows associated with metal parts are mostly sold and especially since all leading manufacturers are able to supply all the various types and the respective customer groups do not differ substantially from one another. In its notification, Continental uses the term "air spring system" to cover all three elements. Since rubber bellows are not discussed in detail here, no differentiation is made in this respect and the term "air spring" is used to designate all the variants.

10 See, for example, answers 3 and 5 by Goodyear to the first questionnaire sent to competitors (utility vehicle air springs).

11 In Europe, drivers nowadays attach particular importance to ride comfort, so that air springs are much more widespread (71% as against 51% in the NAFTA countries, see Continental's notification documents, p. 61). See also paragraphs 18 to 42.

12 See, for example, answers 3 to 5 to the first questionnaire to competitors (utility vehicle air springs).

13 Interview with Firestone on 19 August 2004.

14 See questions 3 to 5 in the first questionnaire to competitors (utility vehicle air springs). 15 See questions 3 to 5 in the second questionnaire to customers (utility vehicle air springs). The small number of divergent replies stems mostly from US firms (e.g. Hendrickson) or firms that work closely with US firms. In the United States, as already mentioned, steel springs are still frequently used nowadays.

16 OEM = Original equipment manufacturer.

17 OES = Original equipment supplier.

18 IAM = Independent aftermarket.

19 See answer 22 in the first questionnaire to competitors (utility vehicle air springs).

20 See below, paragraphs 154 et seq.

21 See question 7 in the first questionnaire for manufacturers of utility vehicle air springs, and question 2 in the first questionnaire for customers of utility vehicle air springs.

22 It seems possible at most to distinguish a market for driver's seats and cab suspensions. However, since their production does not involve any production-specific know-how that is not also available to other competitors, the Commission is not identifying any sub-market here, particularly as there would be no appreciable differences as regards the competition assessment.

23 The term "Europe" is used here to designate the EEA countries including Switzerland, the accession candidates Bulgaria, Romania and Turkey, and the states comprising the former Yugoslavia.

24 According to the information provided by the market participants surveyed, this is due not only to the long, straight roads, but also to the more frequent use of cruise control.

25 See, for example, Goodyear's answer to question 6 in the second questionnaire to competitors (utility vehicle air springs).

26 See Firestone's answer to question 6 in the second questionnaire to competitors (utility vehicle air springs).

27 See, for example, Article 34(4)(2)(d) of the German Vehicle Registration and Approvals Act (StVZO), which provides that the permitted double axle load may amount to 19t only if (inter alia) air springs are used.

28 See, for example, Phoenix's answers to question 7 in the second questionnaire to competitors (utility vehicle air springs).

29 The total permitted length of US heavy goods vehicles is longer than in Europe, with the result that the construction (and hence also the suspension) of the vehicle is different from that of European models; for example, in European vehicles, in contrast to US vehicles, the engine is mostly situated above the front axle and not in front of it.

30 See, for example, Goodyear's answers to question 7 in the second questionnaire to competitors (utility vehicle air springs).

31 See, for example, Vibracoustic's answers to question 7 in the second questionnaire to competitors (utility vehicle air springs).

32 Market participants attribute this, amongst other things, to the fact that US drivers mostly own their own vehicles and attach more importance to price than European importers, whose drivers are employees and evidently demand more comfort; on the question of the higher comfort requirement, see, for example, the answers to question 8 given by Firestone or CF Gomma in the second questionnaire to competitors (utility vehicle air springs).

33 Whereas only 51% of trucks were equipped with air springs in the United States in 2003, the corresponding figure in Europe was 71%, see Continental's notification documents, p. 61.

34 See Firestone's answer to question 6 in the second questionnaire to competitors (utility vehicle air springs). Other product differences stem, for example, from the requirement of European bus customers that they be able to change the air springs themselves, see Goodyear's answer to question 7 in the second questionnaire to competitors (utility vehicle air springs).

35 See interview with Firestone on 19 August 2004.

36 Reply to the statement of objections, p. 9.

37 That market conditions even within Europe are not completely uniform does not call this result into question - contrary to what the parties maintain in their reply to the statement of objections - since market conditions at any rate in the most significant countries of the EEA for the air springs business are essentially uniform.

38 See the answers to question 4 in the second questionnaire to competitors (utility vehicle air springs).

39 See the reply to the statement of objections, p. 11.

40 On the relevance of language, see, for example, the interview with Firestone on 19 August 2004.

41 See, for example, the answers to question 25 in the first questionnaire to customers (utility vehicle air springs).

42 See, for example, the answers given by Phoenix and Firestone to question 34 in the first questionnaire to competitors (utility vehicle air springs).

43 See below, paragraphs 130 et seq.

44 See, for example, the answers to question 25 in the first questionnaire to customers (utility vehicle air springs).

45 Council Regulation (EEC) No 2658-87 of 23 July 1987 on the tariff and statistical nomenclature and on the Common Customs Tariff, OJ L 256, 7.9.1987, p. 1, as amended by Commission Regulation (EC) No 1789-2003, OJ L 281, 30.10.2003, p. 1.

46 Interview with Firestone, 19 August 2004.

47 This does not prevent European manufacturers also from maintaining relatively small stocks near the place of delivery (see reply to the statement of objections, p. 17). Unlike Firestone, however, these manufacturers do not have to maintain a large transhipment centre/central stock for Europe.

48 Interview with Firestone, 19 August 2004; interview with Goodyear, 23 August 2004.

49 See above, paragraphs 14 et seq.

50 In the present case, ultimately, the definition of the geographic market is not what matters, since even if the market is defined in Europe-wide terms there is no question of a dominant position being created; see below, paragraphs 154 et seq.

51 See for instance the replies of BMW, Audi or DaimlerChrysler to question 4 in the questionnaire for car air springs customers.

52 The customary manufacturing process for car air springs thus differs from the manufacturing process for utility vehicle and rail vehicle springs, which are pressed and rolled individually.

53 As with utility vehicle air springs, manufacturers sometimes sell air springs with cover/clamping ring or sometimes just with individual rubber bellows. However, a separate consideration of the different markets for the different designs is not necessary in this case either (it would not change anything in the competition analysis).

54 According to its own information, ZF Sachs, the firm mentioned by Continental in the notification as an air spring manufacturer, does not produce air springs.

55 See, for example, the answers given by Thyssen/Krupp, ZF Sachs, DaimlerChrysler and Vibracoustic to question 12 in the questionnaire for car air spring customers.

56 An import duty of 2.5% is currently charged, see footnote 45.

57 Continental uses the term "air spring system" which is also customary in the industry to describe the secondary air spring. Here, however, for reasons of consistency, the similarly customary term .air spring. is used to describe the total unit consisting of rubber bellows/clamping plate/piston and the other metal and rubber-metal parts associated with it.

58 In the rail vehicle area, a distinction is made between primary and secondary suspension. Primary springs are elastic spring elements between the wheel set and bogie frame which as axle springs can carry out three functions at once: they react to vertical jounce and loads that arise longitudinally and laterally from the influence of the rail track on the chassis, while at the same time helping to decouple structure-borne noise. In addition, wheel wear is reduced, and this helps considerably to make modern rail vehicles more economical. Air springs are not used as primary springs. Secondary suspensions, by contrast, are pneumatic supports of the wagon box which not only dampen the vibrations of the bogie and wagon box and noise, but also allow automatic level adjustments and ride comfort that is not dependent on the load of the vehicle. See, for example, the comments in: http://www.trainweb.org/railwaytechnical/suspen.html.

59 See Continental's reply of 12 August 2004 to the Commission's decision to initiate proceedings, p. 11.

60 See answer 7 to the questionnaires to customers/competitors (rail air springs).

61 Continental's reply to the statement of objections, p. 53.

62 However, according to the information provided by customers, air springs consisting of several vibration damping parts (air bellows, rubber-metal parts etc.) are as a rule purchased as a unit or complete system, which is why the market being examined here is the market for air springs and not for individual components (e.g. rubber bellows).

63 See Question 32 in the questionnaire for competitors (rail air springs).

64 See interview with Hutchinson/Paulstra, 26 August 2004.

65 See, for example, the comments on entry barriers in the interview held on 26 August 2004 with Mr Goettmann (Freudenberg/Schwab), who puts the relevant investment costs at around EUR 25-30 million, which, however, is a maximum.

66 The costs for plant and equipment alone can amount to up to EUR 50 000, see interview with Alstom on 23 July 2004.

67 See interview with Siemens on 23 July 2004.

68 Air springs are also sold to national rail companies (e.g. SNCF, Die Bahn).

69 See, for example, the interview with Alstom on 23 July 2004.

70 See interview with Alstom on 23 July 2004.

71 The terminology in the field of anti-vibration systems is not uniform. Sometimes the term used is vibration systems or vibration damping systems (see, for instance, Commission Decision COMP/M.1778 - Freudenberg/Phoenix), sometimes the concept employed is rubber-metal parts (see, for example, Commission Decision IV/M.2603 - ZF/Mannesmann). Different firms also use their own brand names, such as "Metallgummi" (Phoenix) or "Schwingmetall" (Continental). Since, so far as the Commission is aware, the term regularly used in international usage is "anti-vibration systems" (see, for example, Commission Decision IV/M.253 - BTR/Pirelli), this is the term employed here.

72 See, for example, Commission Decisions COMP/M.2603 - ZF/Mannesmann, COMP/M.1778 - Freudenberg/Phoenix and IV/M.253 - BTR/Pirelli.

73 See below, paragraph 185.

74 See below, paragraph 185.

75 Commission Decision IV/M.253 BTR/Pirelli, paragraph 23. Commission Decision COMP/M.1869 - CVC/BTR Siebe Automotive Sealing Systems did not need to go further into the market definition in the hose sector, since there were no horizontal overlaps; see paragraph 5.

76 See below, paragraph 187.

77 See below, paragraph 187.

78 As a rule light conveyor belts are used for transporting single items, whereas heavy conveyor belts are used mainly for transporting bulk materials.

79 Also known in the industry as steel band conveyor belts.

80 See p. 111 of the notification documents; see also the interviews with Vattenfall and Semperit on 31 August 2004.

81 The Commission is aware, admittedly, of some cases where textile belts were used on an installation previously designed for steel cord belts (see the example of a 1 250 N/mm belt at [...]*, cited by Continental in the reply to the statement of objections). In its survey, however, the Commission inclined to the view that these are exceptions, which do not suggest significant competitive pressure on the part of the textile belt manufacturers. This also confirms the example of the textile belt manufacturer FennerDunlop, which has not managed with its textile belts to gain substantial market share in the steel cord belt sector. In addition, the tensile strength of the (1 250 N/mm) belt in the example of a change cited by Continental was less than that of "typical" steel cord conveyor belts.

82 According to estimates by customers and competitors, less than 10%; see the interviews with Vattenfall and Semperit on 31 August 2004.

83 According to Continental, there is nowadays the occasional textile belt with a resistance to tear of up to 3 150 N/mm. On the other hand, the load-carrying ability of the strongest steel cord conveyor belt can be as high as 7 800 N/mm.

84 The estimate put forward by Continental that only 5-10% of the worldwide (!) demand for steel cord belts was accounted for by the upper load categories' (> 2 500 N/mm) sector was not confirmed for the European market at any rate. The two largest customers thus account for over 60% of demand for broad belts in excess of 4 500 N/mm (see, for example, the interview with RWE on 26 July 2004).

85 See, for example, the interview with Mr Hanson (Bridgestone) on 26 July 2004.

86 On this point PPC refers in its answer to question 5 in the second questionnaire for steel cord belt customers.

87 See, for example, the interview with Mr Hanson (Bridgestone) on 26 July 2004; answer by PPC to question 5 in the second questionnaire for steel cord belt customers.

88 See answer by Beumer Maschinenfabrik to question 6 in the second questionnaire addressed to steel cord belt customers.

89 See answer No 6 by Kali & Salz to the second questionnaire addressed to steel cord belt customers.

90 RWE's answer to question 4 in the questionnaire for steel cord conveyor belt customers. The fact that there may be individual cases where relatively small customers also use textile or "Fleximat" steel weave belts (see reply to the statement of objections, footnote 186) does not contradict this.

91 See the answers to question 5 in the first and to question 7 in the second questionnaire for steel cord conveyor belt manufacturers.

92 See the answer to questions 4 and 6 in the second questionnaire for steel cord belt customers, in which it is expressly asked whether, instead of a steel cord belt, each customer could use a textile belt. Only one customer replied that it might "possibly" also be able to use textile belts. However, that customer said that it used only steel cord belts in the 1 100 N/mm load-carrying category.

93 The example given by Continental on p. 112 of the notification of the use of textile belts by Deutsche Steinkohle AG is not, in the Commission's view, suitable as an example of the interchangeability of both types of belt. Textile belts were, according to the user, used exclusively for fire prevention reasons. The differences that have long existed between steel cord and textile conveyor belts from the point of view of inflammability and legislation, which for a long time prohibited the use of steel cord belts under ground, militate rather against a common product market.

94 Notification documents, p. 111.

95 See the interview with Fenner Dunlop on 28 July 2004.

96 See Metso's answer to question 4 in the first questionnaire for steel cord belt manufacturers: .The production cannot be switched from textile belts to classical steel cord belts without major investments and upgrading of existing production facilities.. As regards converting production, see also question 21 in the second questionnaire for steel cord conveyor belt manufacturers.

97 See Metso's answer to question 5 in the first questionnaire for steel cord belt manufacturers.

98 A further differentiation within steel cord conveyor belts (e.g. belts for open cast lignite mining) is not made here because the market position of the parties to the merger is already dominant in the overall market for steel cord conveyor belts. In the market for specialty belts, on the other hand, there are no competition problems as far as the Commission is aware.

99 See, among others, the examples of PPC (IMAS/Conti, Greece) or RWE and Vattenfall (Phoenix/Conti, Germany).

100 It may be of decisive importance to the market definition whether non-European customers are successful in the case of worldwide invitations to tender. Large suppliers such as RWE do not, however, make such worldwide invitations to tender; see interview with RWE on 26 July 2004; see also Goodyear, answer to question 10 in the first questionnaire for competitors in the steel cord conveyor belt market: - . not very widespread. ( - nicht sehr verbreitet).

101 See Bridgestone's answer to question 30 in the second questionnaire for steel cord belt manufacturers; interview with Cobra on 22 July 2004.

102 See footnote 45.

103 Contrary to Continental's view, the storing of certain quantities of belt near customers is necessary for that very reason, so as to quickly provide replacements in the event of damage.

104 See answer by Beumer Maschinenfabrik to question 22 in the second questionnaire for steel cord belt customers.

105 Interview with Bridgestone on 26 July 2004.

106 See answer by Koch Transporttechnik to question 22 in the second questionnaire for steel cord belt customers.

107 See PPC.s answer to question 20 in the second questionnaire for steel cord belt customers and Koch Transporttechnik.s answer to question 22 in the second questionnaire for steel cord belt customers.

108 See answer by Koch Transporttechnik to question 22 in the second questionnaire for steel cord belt customers.

109 Continental's market share of the total market for heavy conveyor belts in the NAFTA countries is given as [<5%]*.

110 The market shares in a market defined as being worldwide are smaller than in a European market.

111 The turnover figures also contain the turnovers of Continental and Phoenix for driver's seat and cab suspensions. Even if it is assumed that there is a separate market for driver's seat and cab suspensions, the market shares of the parties to the merger and of the other competitors in the market for utility vehicle air springs (minus driver's seat and cab suspensions) would not be significantly different (+/- 1-2%).

112 Firestone is an American firm belonging to the Japanese Bridgestone Group.

113 In answer to Continental's objection in its reply to the statement of objections that the classifying of competitors by customers into predetermined categories has little evidential value since the answer depends very much on who supplies a particular customer, it should be pointed out that even customers who currently have no supply relationship with a given supplier can express a valid viewpoint inasmuch as they can inform themselves about the supplier's performance in particular by calling for tenders. In any case, customer attitudes play a decisive role when it comes to succeeding in a competitive market.

114 In the categories "innovation", "quality" and "spare capacity", Continental leads Phoenix, whereas from the customer standpoint Phoenix is the leader in the "customer service" sphere, followed by Continental.

115 See question 14 in the second questionnaire to customers (utility vehicle air springs).

116 See, as an example of a "traditional" case of a call for tenders, which concerned the tendering procedure with respect to the sale of buses, MAN/Auwärter (Decision of 20 June 2001, COMP M.2201).

117 See Phoenix's answer to question 3 in the second questionnaire to manufacturers of utility vehicle air springs.

118 See, for example, CF Gomma's answers concerning the purchasing procedure in the first questionnaire to manufacturers of utility vehicle air springs.

119 See Phoenix's answer to question 16 in the second questionnaire to manufacturers of utility vehicle air springs. For this reason, where they do make calls for tenders, manufacturers employ strict confidentiality statements.

120 With a few exceptions, the customers surveyed indicated that the involvement of a second supplier gives rise to difficulties in practice. The main reason given for this is the costs associated with a change of supplier (e.g., for testing and qualification procedures). Most of the customers surveyed stated, therefore, that, in view of the small volumes supplied, a change of manufacturer or the qualification of a new manufacturer can be ruled out. See answers to question 27 in the second questionnaire to customers (utility vehicle air springs).

121 Answers to question 34 in the second questionnaire to customers (utility vehicle air springs).

122 Answers to question 26 in the second questionnaire to customers (utility vehicle air springs).

123 Answers to question 41 in the second questionnaire to customers (utility vehicle air springs). Of the 13 customers who replied, including all the major truck and axle/trailer manufacturers apart from ZF Sachs, nine named Continental as the most innovative competitor. Phoenix/Vibracoustic was mentioned once as being the most innovative company and six times as the second in this area, including five times behind Continental as No 1, and therefore ranks second in absolute terms as well as relative to Continental.

124 Answers to question 18 in the second questionnaire to customers (utility vehicle air springs).

125 See interview with Phoenix on 20 August 2004. Phoenix is still performing old contracts under development partnerships predating the joint venture with Freudenberg.

126 Talks with Phoenix on 20 August 2004.

127 To the Commission's knowledge, Phoenix's average price is lower than Continental's.

128 Answer by DaimlerChrysler and MAN to question 21 in the second questionnaire to customers (utility vehicle air springs).

129 See, for example, the answer by MAN to question 27(c) in the second questionnaire to customers (utility vehicle air springs): - . the first supplier is in possession of patents/industrial property rights which it would have to transfer to the second supplier. ( - der Erstlieferant ist in Besitz von Patenten/gewerblichen Schutzrechten, die er auf den Zweitlieferanten übertragen müsste).

130 Interview with Firestone on 19 August 2004.

131 See the publication submitted by Continental in Rubber & Plastics News of 9 August 2004.

132 .In three years. time, Goodyear will be a different corporation. CEO Bob Keegan seeks to cut losses by selling assets. (Goodyear wird in drei Jahren ein anderes Unternehmen sein. Vorstandschef Bob Keegan will durch Verkäufe Schulden abbauen), Frankfurter Allgemeine Zeitung, 25 August 2004, p. 15.

133 In the customer poll carried out as part of the Commission's market survey, CF Gomma has only low rankings in the areas of technical competence/innovation and quality/reliability. CF Gomma has thus - contrary to what Continental states on p. 16 of its reply to the statement of objections - been designated as being on average the cheapest supplier.

134 73% of the answers to question 29 in the second questionnaire to customers.

135 See, for example, MAN.s answer to the first questionnaire for utility vehicle manufacturers.

136 The term .Europe. is used here to designate the EEA countries including Switzerland, the accession candidates Bulgaria, Romania and Turkey, and the states comprising the former Yugoslavia.

137 Particular regard must be had here to the fact that Continental incorporates most of its air spring production into its own air spring systems and only then sells it to manufacturers, as a result of which Continental's market shares may understate somewhat the relative strengths of Continental in the car air spring area.

138 Including Land Rover, which belongs to the US firm Ford.

139 To the Commission's knowledge, Firestone is to be the supplier for a soon-to.be-released European model of a US manufacturer.

140 Assuming an average air spring price of approximately EUR 30 - regarded as realistic by market participants.

141 See, for instance, the open answers of ThyssenKruppp, BMW or Audi to questions 21 and 22 in the questionnaire for car air spring customers.

142 See paragraphs 221 et seq.

143 The market shares in a market defined as being worldwide would be appreciably smaller than in a European market.

144 The market shares in a market defined as being worldwide would be appreciably smaller than in a European market.

145 The market shares in a market defined as being worldwide would be appreciably smaller than in a European market.

146 Sempertrans sees the Metalcord belt type as competing more with textile conveyor belts, see e-mail of 27 July 2004.

147 Interview with Cobra on 22 July 2004.

148 See interview with Vattenfall on 27 July 2004; interview with RWE on 26 July 2004.

149 See interview with RWE on 26 July 2004, and the draft supply contracts with Vattenfall and RWE, provided by Continental (Annex 51.1 and 51.2 to the reply to the Commission's questions, 13 July 2003).

150 The question of the compatibility of such contracts with Articles 81 and 82 of the EC Treaty will not be discussed further here.

151 Interview with RWE on 26 July 2004.

152 Interview with RWE on 26 July 2004.

153 Interview with RWE on 26 July 2004.

154 Interview with Fenner Dunlop on 28 July 2004.

155 Interview with RWE on 26 July 2004.

156 See, for example, interview with Bridgestone on 20 July 2004; interview with RWE on 26 July 2004.

157 Interview with Fenner Dunlop on 28 July 2004; Interview with Bridgestone on 20 July 2004.

158 Interview with RWE on 26 July 2004.

159 Interview with Vattenfall on 27 July 2004. [...]*. Yet even this information does not square with Vattenfall's statements. Besides, the view that RWE is the only customer worldwide does not really seem plausible given the numerous production plants for steel cord belts over 2.30 m wide (e.g. in Japan or Kazakhstan).

160 The theoretical possibility, put forward by Continental, that two narrower belts can be welded together along the longitudinal seam to make larger belt widths could not offset the competitive weakness of manufacturers which have no wide presses. While such a process, the Commission understands, is technically possible, it is more costly and complicated than production in a suitably wide press; according to the competitors, the customer prefers belts vulcanised in a single piece.

161 See in detail the e-mail from Fenner Dunlop, 23 August 2004.

162 See for instance interview with Cobra on 22 April 2004.

163 For instance, the necessary initial materials (e.g. rubber premixture) must be procured from a supplier or through own production, since the rubber part according to Continental's estimates makes up more than 50% of the value of a conveyor belt.

164 See interview with Bridgestone on 26 July 2004.

165 Interview with RWE on 26 July 2004.

166 In particular, the manufacturer mentioned by Continental, Cobra, (market share estimated by Continental: 4%) has no production plant in Europe but only one in China, which serves the Chinese market almost exclusively. Only minimal quantities are sold in Europe (EUR [<5] million). The sales of another manufacturer, Matador, were also severely overestimated (likewise EUR [<5] million).

167 Many customers and competitors, however, described the relevant data as confidential. For openly expressed reservations, see for example Koch Fördergurttechnik's reply to question 34 in the first questionnaire for conveyor belt customers and the interview with Goodyear on 20 July 2004; see also the replies from Fenner or Semperit to question 23 in the first questionnaire for conveyor belt manufacturers.

168 See Fenner Dunlop.s answer to question 4 in the first questionnaire on the steel cord conveyor belt sector: "Know-how is similar, switching to special applications could be done easily with no major investment involved".

169 See the Commission Notice on remedies acceptable under Council Regulation (EEC) No 4064-89 and under Commission Regulation (EC) No 447-98, paragraph 12 (OJ C 268, 2.3.2001, p. 3).

OPINION

of the ADVISORY COMMITTEE on CONCENTRATIONS given at its 129th meeting on 19 October 2004 concerning a draft decision relating to Case COMP/M.3436-Continental/Phoenix

1. The notified operation of Continental to achieve sole control of Phoenix constitutes a concentration within the meaning of Art 3 (1) b of the Merger Regulation.

2. The notified operation has a Community dimension as defined in Art 1 (2) Merger Regulation.

3. The Advisory Committee agrees with the Commission's delineation of the relevant product markets contained in the Draft Decision.

4. The Advisory Committee agrees with the Commission's definition of the relevant geographic markets contained in the Draft Decision.

5. The Advisory Committee shares the Commission's view that the concentration as initially notified would create or strengthen a dominant position on

a) the European OEM/OES-market for air springs for utility vehicles and

b) the European market for heavy steel cord conveyor belts.

6. The Advisory Committee shares the Commission's view that the question whether the concentration would create or strengthen a dominant position in the market for car air springs can be left open taking into account the commitment of divesture offered by Continental.

7. The majority of the Advisory Committee shares the Commission's view that the commitments submitted by the Parties, to sell

a) the air spring production in Hungary;

b) the 50% stake held by Phoenix in Vibracoustic and

c) a production line for wide steel cord belts, are sufficient to remove the competition concerns mentioned above (question 5.). The minority disagrees.

8. The majority of the Advisory Committee therefore agrees that the operation, subject to full compliance with the obligations entered into by the parties, should be declared compatible with the Common market and the functioning of the EEA-Agreement. The minority disagrees.

9. The Advisory Committee recommends the publication of its opinion in the Official Journal of the European Union.

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

<emplacement tableau>

FINAL REPORT OF THE HEARING OFFICER IN CASE COMP /M.3436 - Continental / Phoenix

(pursuant to Article 15 of Commission Decision 2001-462-EC, ECSC of 23 May 2001 on the terms of reference of Hearing Officers in certain competition proceedings - OJ L 162, 19.6.2001)

On 12 May 2004 the Commission received notification, pursuant to Article 3(1)(b) of the Council Regulation, of a proposed merger by which Continental AG wished to acquire sole control over Phoenix AG, both undertakings being leaders on the rubber products manufacturing market.

Having examined the information submitted by the parties to the proposed merger and conducted a market survey, the Commission concluded that the merger raised serious doubts as to compatibility with the Common market and the EEA Agreement. Market participants were consulted about the effectiveness of the commitments proposed by the parties with a view to amending the original proposal, but these were considered to be not sufficient to prevent serious doubts. On 29 June 2004, therefore, the Commission initiated the procedure provided for by Article 6(1)(c) of the Merger Regulation.

On 3 September the Commission sent the parties a statement of objections. A CD-Rom was also sent to them, giving them access to the file. They replied to the statement of objections on 17 September 2004. They waived a formal hearing.

Following the replies to the statement of objections, the Commission continued to have doubts about the following markets: air springs for cars and utility vehicles and heavy steel cord conveyor belts. It dropped the objection concerning the market for rail vehicle air springs.

On 1 October 2004 the parties proposed further commitments amending the original merger proposal, about which market participants were also consulted. In view of these commitments, and of the replies from market participants, it was considered that the merger could be authorised.

No questions have been addressed to me by the parties or third parties. The case calls for no special comment concerning the right to be heard.