EC, July 3, 2001, No JV.55

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Community, Having regard to the Agreement on the European Economic Area, and in particular Article 57 thereof, Having regard to Council Regulation (EEC) No 4064-89 of 21 December 1989 on the control of concentrations between undertakings (1), as last amended by Regulation (EC) No 1310-97 (2), and in particular Article 8(2) thereof, Having regard to the Commission's decision of 1 March 2001 to initiate proceedings in this case, Having given the undertakings concerned the opportunity to make known their views on the objections raised by the Commission, Having regard to the opinion of the Advisory Committee on Concentrations (3), Whereas:

I INTRODUCTION

1. On 31 January 2001, the Commission received a notification pursuant to Article 4 of Regulation (EEC) No 4064-89 ("the Merger Regulation") by which the undertakings Hutchison Ports Netherlands BV (Hutchison) and Rotterdam Container Participatie Maatschappij BV (RCPM) acquire within the meaning of Article 3(1)(b) of the Merger Regulation joint control of the undertaking Europe Combined Terminals BV.

2. On 1 March 2001, the Commission decided in accordance with Article 6(1)(c) of the Merger Regulation and Article 57 of the EEA Agreement to initiate proceedings in this case.

3. The present operation was preceded by a similar operation notified to the Commission pursuant to Article 4 of the Merger Regulation in March 1999 (Case COMP/M.1412). That operation involved the proposed joint acquisition by Hutchison Port Holdings Ltd (HPH) and Rotterdam Municipal Port Management (RMPM) of ECT through a single-purpose holding company so that the parties would have joint control of ECT.

4. In June 1999, the Commission issued a statement of objections, considering that the combination of HPH.s container terminals in the United Kingdom and ECT's container terminals in Rotterdam would lead to the creation of a dominant position on the market for the provision of stevedoring services to deep-sea container ships in Northern Europe.

5. In July 1999, HPH and RMPM decided to abandon the operation.

6. The current operation was originally notified on 3 November 1999 by Hutchison Atlantic Ltd, Rotterdam Container Participatie Maatschappij BV and ABN AMRO Effecten Compagnie BV (ABN) as an agreement falling within the scope of Council Regulation No 17 of 6 February 1962, First Regulation implementing Articles 85 and 86 of the Treaty (4), as last amended by Regulation (EC) 1216-1999 (5) (Case COMP/37.688/D2 - RCPM + Hutchison + ABN AMRO).

7. On 24 October 2000, the Commission issued a statement of objections pursuant to Article 18 of the Merger Regulation (Case COMP/JV.52 - ECT - Port of Rotterdam), finding that Hutchison Altantic Ltd (through its subsidiary Hutchison Ports Netherlands BV (Hutchison)) and RMPM (through its subsidiary RCPM) had acquired de facto joint control of ECT and that the notified transaction therefore constituted a concentration within the meaning of Article 3(1)(b) of the Merger Regulation. The Commission concluded that by failing to notify the transaction under the Merger Regulation and by putting the concentration into effect before such notification, Hutchison and RMPM had infringed the provisions of Articles 4(1) and 7(1) of the Merger Regulation.

8. The Commission took the view that Hutchison Altantic Ltd and RMPM had acquired joint control of ECT on the basis of a number of elements, which, taken collectively, indicated that the two parties would act together when exercising their voting rights in ECT (through ECT Beheer).

9. Hutchison Atlantic and RMPM replied to the statement of objections on 20 December 2000, rejecting the Commission's view of the operation. By letter of 19 January 2001, the parties withdrew their request for an oral hearing, while indicating that they might submit further written statements.

II THE PARTIES

10. The main acquiring parties are:

(a) Hutchison Netherlands BV (Hutchison), a subsidiary of Hutchison Atlantic Ltd, which in turn is an indirect subsidiary of Hutchison International Port Holdings Ltd (HIPH). All of these undertakings are members of the Hutchison Whampoa group, Hong Kong. The latter is a conglomerate with a diverse range of activities. The main activity of HIPH and its subsidiaries is the provision of stevedoring services in ports world-wide. In Europe, HIPH controls, inter alia, the deep-sea container terminals of Felixstowe and Thamesport (United Kingdom);

(b) Rotterdam Container Participatie Maatschappij BV (RCPM), a holding company that is wholly owned by Rotterdam Municipal Port Management (RMPM). The latter is a department of the Municipality of Rotterdam. RMPM is responsible for the development and management of the port of Rotterdam;

(c) ABN AMRO Effecten Compagnie BV (ABN), a wholly-owned subsidiary of ABN AMRO Bank NV. ABN AMRO Bank is engaged in the purchasing, selling, management and administration of securities, property and other assets; the granting of loans; asset management and acting as a banker; it also participates in other companies.

11. The acquired undertaking, Europe Combined Terminals BV (ECT), is a company engaged in the provision of stevedoring services at certain European ports, foremost Rotterdam.

12. The notifying parties are Hutchison and RCPM. The notification on Form CO has been submitted "without prejudice" to the parties' view that the operation does not constitute a concentration.

III THE OPERATION

13. The acquiring parties have created a new holding company, ECT Beheer BV (ECT Beheer) which holds 100% of the share capital of ECT. Hutchison and RMPM/RCPM (6) each have 35 % of the shares in ECT Beheer, with ABN holding 28 % and an employees' trust (Star) holding the remaining 2%.

IV CONCENTRATION

14. The present operation leads to Hutchison and RMPM acquiring joint control of ECT. In this respect, paragraph 32 of the Commission Notice on the concept of concentration under Council Regulation (EEC) No 4064-89 on the control of concentrations between undertakings (7) states that "[...] collective action can occur on a de facto basis where strong common interests exist between the minority shareholders to the effect that they would not act against each other in exercising their rights in relation to the joint venture".

15. The Commission bases its conclusions regarding the nature of the present operation on the following elements, which, taken collectively, show in its view that there are strong common interests between Hutchison and RMPM such that the parties when voting will not act against each other:

(a) a commonality of understanding between the parties ("the strategic investors") established through their joint involvement in the previous operation (Case COMP/M.1412, described in paragraph 3) and the absence of any such obvious commonality as between either of these two parties and the third main shareholder, ABN ("the financial investor"). The involvement in the previous operation gave each strategic investor the opportunity to explore the other's objectives in investing in ECT and come to an understanding as to how these objectives might be reconciled. To a large extent these objectives were and are common objectives. (8) By contrast, the information available to the Commission (9) shows that ABN's role in the operation is that of project financier to the two strategic investors and that its decision to take up a substantial equity stake was the action of a "white knight" seeking to assist two important actual or potential clients. This information shows further that ABN intends to reduce the size of its stake substantially (that is to say, by 50% or more) as soon as it has the opportunity to do so. This short-term view must be contrasted with the two strategic investors' long-term commitment, manifested, inter alia, through an undertaking to maintain a substantial shareholding in ECT for a period of at least five years;

(b) the structure of the shareholdings and voting rules, which seem to have been tailored in such a way as to allow the two parties to exercise joint control over ECT. Hutchison and RMPM together have 70% of the shares and voting rights, with a 60% majority being required for most decisions; (10) they also have four out of five seats on the Board of Supervisory Directors. (11) Although ABN would theoretically be able to form a voting majority with either of the strategic investors, the evidence shows that this was not the intention when the voting rights and rules were decided; (12)

(c) a high degree of mutual dependency as between the two strategic investors regarding the success of their respective investments in ECT. Hutchison thus counts on RMPM to ensure the approval of future ECT development plans, RMPM in turn expects Hutchison to provide expertise and capital. In the former respect, RMPM has stated that it has "chosen" ECT to be the one and only stevedore in the western part of the port of Rotterdam (Maasvlakte) to receive and handle large vessels - Hutchison's expectation that RMPM will give preference to ECT when considering applications for further expansion in Rotterdam must accordingly be regarded as wellfounded. Expansion is vital if ECT is to maintain its position as the leading container terminal operator in Northern Europe. (13) Equally essential is access to expertise and capital. Hutchison, as the world's leading container terminal operator and financially well endowed, is indisputably well placed to assist; (14)

(d) a common guarantee provided by the strategic investors for the benefit of the financial investor. The common guarantee takes the form of a clause (Clause 11(5)) in the shareholders. agreement by which the two strategic investors undertake to maintain a minimum shareholding of at least 24% each in ECT for a period of five years following Closing. The evidence shows that this clause serves a dual purpose: on the one hand it provides each strategic investor with reassurance as to the seriousness of the other's commitment to ECT, on the other hand it provides an assurance to the financier, ABN, that its exposure on loans to ECT Beheer is limited by the backing of the Hutchison Whampoa Group and the Port of Rotterdam. Clause 11(5) must be read in conjunction with provisions in the credit agreement (the "Term Loan Agreement") that give ABN the right to call in the loans granted to ECT Beheer should the strategic investors' shareholding fall below the agreed minimum level.

16. The above elements tie Hutchison and RMPM closely together and make it unlikely that either would choose to act against the other when voting.

17. The parties have argued, inter alia, that RMPM, by indicating its willingness to make land available to Maersk for the latter to develop its own terminal at the Delta site in Rotterdam (which subsequently led to the establishment of the Maersk Delta joint venture between Maersk and ECT), has acted contrary to the interests of ECT (and of Hutchison). Further evidence of diverging interests is provided, in the parties' view, by RMPM.s decision to grant rights to the Euromax terminal (a joint venture between ECT and P&O Nedlloyd). The parties claim that the establishment of these terminals will result in a significant loss of revenue and profit for ECT.

18. The Commission notes that the parties themselves have contended that the alternative to allowing Maersk and P&O Nedlloyd to obtain their "own" facilities was to see these lines depart from Rotterdam altogether. Whether or not there was ever a significant risk of this occurring, (15) RMPM could legitimately consider that it was in the best interests both of ECT and of RMPM to tie these two shipping lines to Rotterdam through joint venture agreements. It should be noted that there is every indication that Maersk and P&O Nedlloyd would have preferred to have complete ownership and control of "their" respective terminals. The fact that ECT was awarded a substantial stake in both joint ventures (33.3% in Maersk Delta; 50% in Euromax) can be directly attributed to RMPM using its influence and leverage as port manager (allocating sites and leases) to obtain the best possible outcome for ECT.

19. The parties have also referred to various declarations by the chairman of RMPM to the effect that it is impossible for RMPM, as port manager, to give ECT preferential treatment. In this respect, it is sufficient to note that these declarations are contradicted by various statements that RMPM has made to the ECT management and workers' council. (16)

20. Finally, the parties have stated that ABN.s participation in ECT is neither transitory nor restricted to the role of project financier. Again, this statement is contradicted by various statements of the party in question. (17) ABN AMRO has made no secret of its desire to reduce its exposure to ECT at the earliest opportunity.

21. For all of the above reasons, it is concluded that the current operation constitutes a concentration falling within the scope of the Merger Regulation, as Hutchison and RMPM acquire joint control over ECT within the meaning of Article 3(1)(b) of that Regulation.

V COMMUNITY DIMENSION

22. The undertakings concerned have a combined aggregate world-wide turnover (18) of EUR 7 386 million (Hutchison EUR 6 694 million; RMPM EUR 350 million; ECT EUR 342 million). They each have an aggregate Community-wide turnover in excess of EUR 250 million (Hutchison EUR [...]* million; RMPM EUR [...]* million; ECT EUR [...]* million), and none of the undertakings concerned achieves more than two-thirds of its aggregate Community-wide turnover within one and the same Member State. The notified operation therefore has a Community dimension within the meaning of Article 1(2) of the Merger Regulation.

VI ASSESSMENT UNDER ARTICLE 2 OF THE MERGER REGULATION

A Relevant Market

1. Product markets

23. Hutchison and ECT are both container terminal operators providing stevedoring services to deep-sea container vessels in Northern Europe. There is no, or only very marginal, overlap between the parties to the concentration with regard to other business activities.

24. The parties themselves propose (19) that the relevant product market should be that for the provision of stevedoring services to deep-sea container ships. This market definition can be accepted, subject to the following remarks and modifications.

25. Vessels carrying non-containerised cargoes and short-sea vessels have other requirements with regard to services and facilities than deep-sea container vessels. Similarly, while it is customary to distinguish between different container traffic flows, the stevedoring service provided in respect of these different traffic flows is essentially the same.

26. Container terminal throughput can be broken down by traffic flow as follows:

(a) hinterland traffic (also known as .direct deep-sea.), that is, containers transported directly onto/from a deep-sea container vessel to/from the hinterland (via barge, truck or train), and

(b) transhipment traffic, that is, containers destined for onward transportation to other ports. Transhipment traffic can in turn be sub-divided into relay traffic and feeder traffic. "Relay" refers to containers moved between deep-sea vessel and deep-sea vessel (for example, containers transferred from a Far East liner shipping service to a transatlantic service). "Feeder" concerns the movement to or from deep-sea vessels from or to shortsea vessels (feeder vessels).

27. In its decision of 1 March 2001, the Commission considered that the available evidence at that stage of the procedure did not support the conclusion that one or more of hinterland, feeder or relay traffic should be regarded as distinct markets.

28. Following further investigation, that conclusion should be revised.

29. The parties have argued that hinterland and transhipment traffic flows are inseparable in that transhipment traffic is intrinsically linked and secondary to hinterland traffic. With regard to a possible further breakdown of transhipment traffic into feeder and relay flows, the parties maintain that feeder and relay services are negotiated under one terminal contract and that terminal operators neither can, nor do in practice, price discriminate between these two flows. It therefore follows, in the parties. view, that neither hinterland nor transhipment, and certainly not feeder and relay, constitute discrete relevant markets. Hinterland and transhipment

30. Ocean Shipping Consultants Ltd (OSC), in its report "North European Containerisation", (20) appears to identify a transhipment hub port market which is distinct from the various regional hinterland port markets. (21) Admittedly, hinterland and transhipment flows are linked inasmuch as most deep-sea (22) container vessels carry both hinterland and transhipment containers and in that the choice of port of call is dictated primarily by hinterland considerations. However, the range of ports and terminals that may be potential substitutes for transhipment traffic is not necessarily identical to the range of ports and terminals that may be substitutes for specific hinterland traffic.

31. A case in point is the identification by OSC of a United Kingdom/Ireland market, on which the main United Kingdom deep-sea ports compete primarily with each other for United Kingdom/Ireland hinterland traffic, but a broader "hub port market" on which the United Kingdom ports compete not only with each other but also with North Continental terminals for transhipment traffic. (23) The latter traffic is, as the notifying parties have repeatedly stressed, "footloose"; that is to say, transhipment can, in principle, take place at any North European port at which the vessel in question calls to load or unload hinterland cargo. The range of possible substitute ports for transhipment volumes is therefore greater than that for volumes destined for or originating in any specific hinterland. This is reflected in the pricing policy of the North European terminal operators: handling charges per move for transhipment containers are, in general, lower (24) than the rates per move for hinterland containers.

32. The arguments and submissions made by the parties support the notion of (stevedoring services in respect of) hinterland traffic and transhipment traffic as discrete relevant markets. The parties thus contend that there is no significant competition between the United Kingdom and North Continental ports for each other's hinterland and that real competition can only exist between the main ports within each separate hinterland region (for example, the Benelux region [Belgium, the Netherlands and Luxembourg] and the United Kingdom region). (25) The parties further point to evidence of carriers switching volumes between ports. This evidence (26) shows that switches of entire liner shipping services or of individual strings occur almost exclusively between ports in the same hinterland region (for example, between Rotterdam and Antwerp, Felixstowe and Thamesport, Hamburg and Bremerhaven), whereas switches of transhipment volumes occur between ports located in different hinterland regions (for example, Rotterdam and Bremerhaven, Felixstowe and Antwerp).

33. Replies from third parties, while confirming that the basic stevedoring service provided in respect of hinterland and transhipment traffic is essentially the same, generally support the view that the range of possible substitute terminals and ports for hinterland traffic is different from the range of substitutes for transhipment traffic.

Feeder and relay

34. OSC distinguishes between the .hub and spoke' (that is to say, feeder) and relay sectors, but acknowledges that the two areas are blurred and that statistics are not fully comprehensive. (27) This is confirmed by the Commission's own investigation, which has shown that individual operators (shipping lines and terminal operators) have differing interpretations of the terms "feeder" and "relay". (28) The investigation also shows that most terminal operators currently have difficulty in identifying and quantifying feeder and relay flows so that it could reasonably be argued that they would have difficulty in price discriminating between the two, should they wish to do so.

Conclusion on the product market

35. It is concluded from the above that (the provision of stevedoring services in respect of) hinterland traffic and transhipment traffic each constitutes a discrete relevant product market, but that there is insufficient evidence to justify a further breakdown of the transhipment market into discrete feeder and relay markets.

36. The notifying parties agree with this conclusion. (29)

2. Geographic dimension

Hinterland

37. The available market studies and other evidence suggest that there are a number of discrete regional hinterland markets within Northern Europe.

38. OSC has thus identified three main hinterland regions: (30)

(a) "North Continent West" comprises all terminals in the range between Rouen and Amsterdam, with competitive pressures primarily focused between Antwerp and Rotterdam, with additional capacity at Le Havre;

(b) "North Continent East" currently consists primarily of the ports of Bremerhaven and Hamburg;

(c) The "United Kingdom/Ireland market" encompasses the British Isles and includes the major United Kingdom gateway terminals (31).

39. A fourth region consists of the "Scandinavia/Baltic market"; the volumes involved do not however, as a rule, justify a direct call (an exception being the port of Gothenburg). (32) This market is primarily served by feeder links from terminals located in one or more of the three main regions (for example, from the North Sea Terminal in Bremerhaven).

40. OSC's conclusions regarding Northern European regional hinterland markets are confirmed, inter alia, by the service patterns of the main shipping lines and alliances, which generally provide for at least one direct call in each of the main hinterland regions identified above.

41. Competition from Southern European ports is increasing. However, as far as hinterland markets are concerned, third party comments indicate that this competition is currently limited to reclaiming the "natural" local and regional catchment areas of the Southern European ports and contesting specific central European markets such as Switzerland and Austria. (33) There is little evidence of competition from Southern European ports for the "natural" (that is to say, local and regional) catchment areas of the Northern European ports, although the reverse may be true. As regards the third party answers quoted by the parties in their Statement of Defence (see paragraph 41 and section V.3 of the Statement of defence) the following should be noted:

(a) Eurogate, after having explained the situation at Contship Italia (34), states that the Mediterranean ports are competitors only to a limited extent and could therefore hardly be seen as substitutes.

(b) As regards the port of Algeciras, operated by an A.P. Møller subsidiary, it should be noted that this port is (geographically speaking) at the crossroads between the North-South and East-West services and therefore not easily comparable with the ports in the North European range. The answer quoted by the parties (35) moreover does not indicate from where volumes have been transferred.

(c) It should further be noted that HHLA states clearly that it does not see Mediterranean ports as a substitute for deep-sea ports in Northern Europe36.

(d) APL37 sees benefits to serving "[...] this hinterland (including the Rhur [sic] area) [...]" because of "[...] reduced transit times from Asia". The shipping line stated however that the possible overall cost savings could only be achieved "[...] if a change of schedule is implemented which reduces the number of vessels currently on the route". It should further be noted that none of APL's traffic related to Gioia Tauro or Algeciras "[...] has previously moved over a North Europe port". As regards the potential attractiveness of Southern European ports in general APL noted that "[a]s hinterland connections improve in Southern Europe, Med ports will see their deep-sea traffic increase".

(e) Hanjin, also quoted in the Statement of Defence as arguing in favour of competition from Southern European ports should however be quoted completely because it states, for example, (38) that "[... Algeciras and Gioia Tauro] cannot be seen as substitutes for North European ports". It adds (39) that it does "[...] not believe that the efforts of Mediterranean ports [to develop their market position] will be left unanswered by the Northeuropean [sic] ports. A lot of cargo traditionally goes via Northeuropean ports, and these ports definitely will safeguard their interests".

42. It is concluded from the market studies (40) submitted by the parties that there is a significant degree of overlap between the various North Continent hinterland markets, but that there is only a limited degree of overlap between the North Continent hinterland markets and the United Kingdom/Ireland hinterland market. This is a view supported by the notifying parties and several third parties. The precise delineation of the various hinterland markets can therefore be left open, as it is sufficient for the purpose of this case to note that the United Kingdom/Ireland market (on which Hutchison is present) and the Continental markets (on which ECT is present) constitute discrete relevant markets. In the light of the limited degree of overlap between the North Continent and the United Kingdom/Ireland hinterland markets the concentration will neither create nor strengthen a dominant position on any hinterland market. There is therefore no need to further assess the impact of the concentration on these markets.

Transhipment

43. The available evidence suggests that the potential substitutes for transhipment traffic in Northern Europe include all main deep-sea ports in the Gothenburg - Le Havre range; (41) that is to say, all ports having sufficient hinterland volumes to justify a direct call by a container liner shipping service. (42)

44. The Commission has considered whether Algeciras should be included in the range of potential substitutes. It has concluded that this question can be left open in the instant case, as [almost all]* of the throughput of the main Algeciras container terminal, (43) which is controlled by the A.P. Møller group, is generated by Maersk Sealand, an A.P. Møller subsidiary. (44) These volumes represent captive production, which, for the reasons given below at paragraph 71, should not be included in the market when determining market shares. (45) The inclusion of Algeciras in the relevant market would therefore have little or no impact on the structure of that market.

45. The parties have argued that the range of substitutes should be extended to include other Southern European ports (see section V.3 of the Statement of Defence). They point, inter alia, to the fact that the transhipment throughput of the Southern European ports has grown dramatically in recent years, both in absolute terms and as a proportion of overall Northern and Southern European transhipment traffic.

46. The growth of transhipment traffic in Southern Europe is not indicative of substantial competition for this traffic as between the Northern European and Southern European port ranges. While some Southern European ports may indeed serve relay traffic on Equatorial trades, they are not an alternative for Northern European ports with regard to feeder traffic. (46) In this context it should be noted that relay traffic accounts for less than a quarter of overall Northern European transhipment volumes. (47) It should further be noted that the increase of transhipment traffic in Southern Europe is largely due to the growth in importance of regional feeder hubs, such as Gioia Tauro. These hubs act as distribution centres for traffic destined for those Mediterranean ports that generate insufficient hinterland volumes to justify a direct call by a major liner shipping service. These regional Southern European hubs provide only marginal competition to the main Northern European ports, the prime function of which, with regard to transhipment, is to act as regional distribution centres for traffic to and from the less important (in terms of hinterland volumes) Northern European ports. Note has also been taken of the fact that most shipping lines, when asked if they would contemplate switching feeder or relay volumes to Southern European ports, have replied in the negative (see paragraph 41).

Conclusion on the relevant market

47. It is concluded that the relevant market to be considered in this case is the market for the provision in Northern Europe of stevedoring services for transhipment traffic carried by deep-sea container vessels.

B Competitive Assessment

1. Current market structure

Throughput

48. 1999 is the most recent year for which complete transhipment throughput data are available. OSC estimates that Hutchison and ECT had a combined share of [around 50%]* of the Northern European transhipment market in that year: (48)

<table>

49. It is apparent from Table 1 that each of the closest competitors had less than half the market share of Hutchison and ECT (HHLA: [between 15 and 25%]*; Eurogate: [between 15 and 25%]*). The Commission's investigation has moreover shown that ECT throughput was understated by OSC. (49) Once this correction has been made, the market share of Hutchison/ECT increases [slightly]* (50).

50. It is clear that the post-merger market is highly concentrated: the market would have a Herfindahl-Hirschman Index (HHI) of 3079. (51) The operation has led to an increase in HHI of 1129. (52) The combined Hutchison and ECT share of the transhipment market has remained stable in recent years: OSC estimates that the parties. combined market share was [below 50%]* in 1997. The fragmentation of competition (the closest competitor to Hutchison/ECT in 1997 had a transhipment share of approximately [under 20%]* was also a feature of the market in 1997.

51. OSC has also provided a breakdown of transhipment throughput by major port in 1999 (53) and 1997: (54)

Table 2

TEU Transhipment 1999

Felixstowe [>800.000]*

Thamesport [<80.000]*

Southampton [<80.000]*

Le Havre [<250.000]*

Zeebrugge [<50.000]*

Antwerp [<600.000]*

Rotterdam - ECT [>1.500.000]*

Rotterdam - Other [<80.000]*

Bremen/Bremerhaven [>500.000]*

Hamburg [>1.000.000]*

Gothenburg [<30.000]*

Others [>50.000]*

Total 5.436.000

1997

Felixstowe [<800.000]*

Thamesport [<50.000]*

Southampton [>60.000]*

Le Havre [>100.000]*

Zeebrugge [<50.000]*

Antwerp [>200.000]*

Rotterdam [>1.000.000]*

Bremen/Bremerhaven [>200.000]*

Hamburg [>1.000.000]*

Gothenburg [<30.000]*

Others [>50.000]*

Total 3.878.100

52. Table 2 provides an indication of the strong position of Rotterdam and Felixstowe as preferred transhipment ports. The only real challenge to that position in 1999 came from Hamburg, the transhipment throughput of which amounted to approximately half that of Rotterdam and Felixstowe combined.

53. Moreover, Table 2 provides an additional illustration of the degree of concentration prevailing in the transhipment market: five ports (Rotterdam, Hamburg, Felixstowe, Antwerp and Bremerhaven) accounted for 91% of all transhipment throughput in Northern Europe in 1999. (55)

54. The parties have argued that the importance of the transhipment market is inflated, in absolute terms and relative to hinterland traffic, by the fact that each transhipment container is .double-counted. in port and terminal statistics. Third parties reject this view: each transhipment container generates two quayside moves and each move generates revenue, albeit - as a rule - at a discounted rate. Further, each move "consumes" terminal capacity. The "double-counting" therefore provides an indication of the importance of transhipment traffic to terminal operators in respect both of revenue and of capacity. The Commission has verified in the course of its indepth investigation that the statistics provided by terminal operators count each transhipment container as two moves and each hinterland container as one move.

55. The parties have also argued that the 1999 throughput figures should be adjusted to take account of the "loss" by ECT of two major customers: Maersk Sealand and P&O Nedlloyd. Maersk Benelux BV and ECT have entered into a joint venture (Maersk Delta BV) to provide a dedicated terminal for the Maersk Sealand shipping line. The terminal started operations in October 2000, but will not reach full capacity until 2005. (56) P&O Nedlloyd and ECT have signed a letter of intent to enter into a similar joint venture (Euromax). Assuming that the letter of intent will be followed by a formal joint venture agreement, that joint venture is expected to start operations in 2004.

56. In the Commission's view, any attempt to adjust 1999 throughput figures to reflect events that may take place after 1999 would be fundamentally flawed inasmuch as it would take no account of other developments on the market between 1999 and the date of the events in question. (57)

57. An illustration of the difficulty of predicting the impact on market share of the loss of a customer or customers is provided by the parties themselves. In the Statement of Defence, paragraph 182, the parties note that in the year 2000 Genoa experienced an increase in transhipment traffic of 21.6% over 1999 figures, despite having lost the traffic of the Mediterranean Shipping Company (MSC) to La Spezia in 1999. In this context, it should be noted that Northern European transhipment traffic is predicted to grow by some 135 % in the period up to 2012. (58) It is thus reasonably certain that the loss of the traffic of a major customer will be offset by demand from other customers.

58. In the Statement of Defence (pages 29 et seq.), the parties present various statistics for the year 2000 which purport to show that the combined market share of Hutchison/ECT has decreased since 1999. The Commission has the following comments on those statistics:

(a) The table on page 29 of the Statement of Defence shows overall deep-sea container port demand in Northern Europe in 1999 and 2000. As the relevant market has been defined by reference to transhipment traffic, this table is of little value. In any event the parties do not explain how the year 2000 figures have been obtained. (59)

(b) The above comments also apply to the table on page 30. According to the parties, this table shows that "the combined market share of total deepsea container traffic of Hutchison/ECT has fallen from [35-40%]* in 1999 to [25-30%]* by the end of 2000, having excluded traffic transferred to Maersk Delta" (page 31, first paragraph). This, the parties claim "represents a significant reduction in the combined market share of Hutchison and ECT during the period of only one year, which the parties firmly believe that the Commission may not ignore". In the table, [slightly more than 1 000 000]* TEU are shown as having been transferred from ECT to Maersk Delta in the year 2000. The available evidence (replies from Maersk and ECT) shows that only [less than 75 000]* TEU were handled by Maersk Delta in that year. The table is therefore, at best, extremely misleading and does not in any way support the conclusion that "the combined market share of total deep-sea container traffic of Hutchison/ECT has fallen from [35-40%]* in 1999 to [25- 30%]* by the end of 2000". (60)

(c) Finally, in tables on pages 32 and 33 of the Statement of Defence, the parties attempt to show that the combined transhipment market share of Hutchison/ECT has declined from [above 50%]* in 1999 to [30-40%]* in the year 2000. First, it should be noted that the transhipment figures for 2000 - with the exception of the figures for Felixstowe, Thamesport and ECT Rotterdam - do not represent actual transhipment throughput in the year 2000. (61). Secondly, the Commission notes that these figures indicate that the parties' combined market share in the year 2000, before any adjustment, was [just under 50%]*. The parties have argued that their combined 1999 market share was [even further below 50%]* (see page 33 of the Statement of Defence, paragraph 110). On that basis, the parties' combined market share has increased by almost [1,5-3%]* since 1999. Thirdly, the [30- 40%]*combined market share mentioned above is obtained by deducting [125 000-175 000]* transhipment TEU switched from ECT to Bremerhaven in June 2000 and [more than 500.000]* transhipment TEU switched from ECT to Maersk Delta in 2000. In this respect, the Commission notes that the parties have stated that the ECT transhipment figure before adjustment [>1 500 000]* TEU (62) represents the actual throughput of that terminal in 2000. That being the case, there should be no need for any further adjustment to take account of switches of traffic from or to ECT in that same year (2000). The Commission also notes that Maersk Delta did not handle [more than 500 000]* transhipment TEU in the year 2000 - the correct figure is [less than 10 000]* TEU. (63)

59. Notwithstanding the above comments, the fact that two major customers have taken, or will take, a participation in a joint venture terminal in Rotterdam is an element that may be relevant for an assessment of the concentration's future position on the market and is therefore examined below (paragraphs 68 et seq; 86 et seq.).

Share of port calls

60. The above figures and tables (see Tables 1 and 2) reflect the market position of the concentration in terms of throughput in 1999. Another indication of the parties. market strength is provided by their high share of port calls by the major liner services on the Northern Europe - Far East and Transatlantic trades. These trades together are estimated to account for almost three quarters of total Northern European container trade. It should also be noted that services on the Europe - Far East trades also deploy the newest and largest vessels (see further below, paragraphs 97 et seq., regarding the impact of larger vessels). (64)

61. According to Drewry, (65) Hutchison/ECT (Felixstowe, Thamesport, ECT Rotterdam) accounted for [less than 40%]* of total scheduled weekly calls by the major carriers on the North Europe - Far East trade in 2000. Hamburg accounted for [less than 20%]*, Le Havre for [close to 10%]* and Antwerp for [less than 10%]*.

62. The parties' share in the same year (2000) of total scheduled weekly calls in Northern Europe by the major transatlantic services was [around 30%]*. Bremerhaven and Le Havre accounted for [less than 20%]* each, and Antwerp for [around 15%].

Capacity

63. Terminal and port capacity statistics do not distinguish between "hinterland" capacity and "transshipment" capacity. These statistics can therefore provide - at best - only a very imperfect indication of the parties' current position and possible future position on the transhipment market.

64. As mentioned in the Decision of 1 March 2001 (paragraph 64), Drewry and OSC have each made separate, and differing, estimates of the amount and structure of current Northern European deep-sea capacity. Drewry estimates that Hutchison/ECT had a combined share of [around 30%]* of overall North European deep-sea port capacity in 1999. OSC estimates that the parties accounted for [less than 30%]* of overall North European deep-sea port capacity in the year 2000. (66)

65. There are significant differences between the two estimates, which are not fully explained by the fact that they cover different years. Estimating capacity is a complex exercise, inasmuch as it involves the weighing of several factors, such as berth lengths, number of cranes, stacking areas, workforce productivity, in order to arrive at an overall conclusion as to effective capacity. This in turn must be distinguished from latent capacity, which is an estimate of potential capacity, assuming a specific, not yet attained, level of productivity.

66. At the Commission's request, the parties have attempted a reconciliation between the Drewry and OSC estimates. On the basis of this reconciliation, supplemented by information obtained directly from port authorities and terminal operators, the Commission has made its own estimate of capacity and capacity shares in the year 2000.

These are set out in Table 3.

<table>

67. As Table 3 shows, Hutchison/ECT had a combined capacity share of [30-35%]* in the year 2000, roughly equivalent to its share of capacity in the year 1997 ([30-35%]*). Hamburg and Antwerp had shares of [less than 20% each]* and [10-15%]* respectively while Le Havre and Bremerhaven accounted for [less than 10% each]*.

68. Allowance has been made in Table 3 for the specific circumstances of the Maersk Delta BV terminal (Rotterdam) and North Sea Terminal (Bremerhaven).

69. Maersk Delta BV is a joint venture between Maersk (66.66%) and ECT (33.33%), the main purpose of which is to provide a dedicated deep-sea container terminal in Rotterdam for the Maersk Sealand shipping line. Despite having only a minority stake in Maersk Delta, ECT has certain veto rights in respect of, inter alia, the joint venture.s pricing policy. It also has equal representation on the Board of Directors, although the Maersk-appointed chairman has the casting vote in event of deadlock. The terminal became operational in October 2000 and will, during a phasing in period expected to continue until 2005, gradually absorb Maersk Sealand volumes currently handled by ECT. The joint venture will not be able to accommodate third parties before 2007. (67)

70. The North Sea Terminal is a joint venture between Maersk (50%) and Eurogate (50%), the main purpose of which is to provide a dedicated terminal in Bremerhaven for Maersk Sealand. Unlike the Maersk Delta terminal, the North Sea Terminal also provides services to third parties (these volumes accounted for [above 10%]* of the overall throughput of the terminal in the year 2000).

71. In general, where integration exists between supplier and purchaser (intra-group supplies), the production in question ("captive production") should not be included in the relevant market for the purpose of assessing market shares. If captive production can be switched to non-captive production within the short-to-medium term, it may constitute potential competition. In the instant case, it is necessary to examine the actual and potential use of each terminal solely or jointly owned or controlled by individual shipping lines in order to determine the proportion of terminal production and capacity which should be regarded as captive.

72. In Table 3, capacity in the year 2000 relating to the Maersk Delta terminal (approximately [below 1 000 000]* TEU) has been apportioned between the owners, Maersk and ECT, in relation to their respective shareholdings (Maersk: 66.6%; ECT: 33.3%). This allocation reflects the fact that although ECT has a minority shareholding it has joint control68 through veto rights in respect, inter alia, of pricing and a non-compete clause which ensures that Maersk Delta cannot compete with ECT for third part business before the year 2007. The Maersk portion ([...]* TEU) should be regarded as captive capacity, on the basis that the terminal will be dedicated exclusively to Maersk Sealand until 2007. It should therefore be excluded in Table 3 both from the ECT market share and from total throughput. The ECT portion ([...]* TEU), which does not constitute captive capacity for ECT, has been added to ECT capacity.

73. The year 2000 capacity of the North Sea Terminal in Bremerhaven ([1 200 000-1 600 000]* TEU), which is jointly owned by Maersk (50%) and Eurogate (50%), has been apportioned between the joint venture partners in accordance with the breakdown of throughput in 1999.

74. Of the throughput handled by the North Sea Terminal in 1999 ([more than 1 000 000]* TEU), [less than 90%]* was accounted for by Maersk Sealand volumes, while the remainder [...]* represented third party volumes. It follows that [slightly less than half the throughput]* represented captive Maersk production which should not be included in the market. The remainder, [...]*, is third-party production (for Maersk and Eurogate) which should be included in the market.

75. Accordingly, [less than half]*, of the capacity of the North Sea Terminal has been excluded in Table 3. The [remainder is]* included in the overall capacity ([of more than 2 300 000]* TEU of Bremerhaven.

76. The parties have argued that in the instant case captive production and capacity should be included in the market because it (a) provides indirect competition through its benchmarking effect on prices and (b) represents potential competition.

77. With regard to (a) the Commission considers it unlikely that the prices set for captive production (internal sales) could have a benchmarking effect, inasmuch as they are unlikely to be transparent and given that in-house pricing follows other principles than market pricing, that is to say, generally lower prices are set for inhouse sales. (69) It follows that even if competitors were in the unlikely position of being able to obtain information about the in-house prices they could not readily determine the market prices the captive producer would have to ask. With regard to terminals with captive production, the benchmarking effect, if any, would be extremely limited, especially in a case of a captive producer which does not have sufficient capacity available to cater for third party needs.

78. In respect of (b) the Commission takes the view that in the case of the North Sea Terminal conversion of captive production to third-party production in response to a price increase at terminals controlled by Hutchison/ECT is unlikely to occur to any significant extent. Bremerhaven is one of Maersk Sealand.s main hub ports. To displace Maersk Sealand volumes from that port to other ports - assuming that this would be feasible - in order to make available further capacity at the North Sea Terminal for third parties would cause Maersk serious inconvenience and cost to its liner shipping operations. This inconvenience and cost is likely to outweigh any benefit Maersk might derive from increased revenue from the provision of services to third parties. Nor is there any evidence that the capacity of the North Sea Terminal could be expanded at short notice in order to meet increased demand from third parties.

79. The Maersk Delta terminal is of course precluded from providing services to third parties within the timeframe under consideration (period up to and including 2005), (70) so the issue of potential competition does not arise. Further, ECT has joint control over the terminal through its representation on the Board and its veto rights. It is therefore clear, in the Commission's view, that the Maersk Delta terminal could not under any circumstances be regarded as a competitor of ECT, even if the terminal's production were to be regarded as non-captive. The Euromax terminal, to which the parties refer, will not become operational before 2004 and will in any event be jointly controlled by ECT and can for that reason not be regarded as a (potential) competitor of ECT.

80. The notifying parties have interpreted comments from third parties as offering support for the proposition that all shipping line owned or controlled capacity and production should be included in the market. In this respect, a distinction should be made between facilities dedicated to a particular shipping line but owned and operated by a terminal operator and dedicated facilities partly or wholly owned by a shipping line. In the former case there is no vertical integration (captive production), in the latter case there is. Third parties generally confuse the two types of facility, referring to both as .dedicated facilities.. The former should of course be included in the market (output on the market), the latter should not (internal output).

81. The parties have argued that in the alternative all capacity relating to terminals jointly owned by shipping lines should be excluded from the market. Accordingly, Table 4 includes neither capacity relating to the Maersk Delta terminal ([less than 1 000 000]* TEU) nor [less than 90%]* (the throughput accounted for by the Maersk Sealand shipping line) of the North Sea Terminal capacity:

<table>

82. As can be seen from Table 4, the exclusion from the market (ports of Rotterdam and Bremerhaven) of the capacity of the terminals in which Maersk has a participation and presence has no significant effect on the concentration's share of capacity (increase [of less than 1%]*).

Conclusion

83. It can be concluded from the above analysis of the current market structure that the concentration will lead to the creation of a dominant position on the relevant market. In reaching this conclusion, particular importance has been attached to the following factors:

(a) the current high combined market share of Hutchison and ECT on the relevant market, in conjunction with the relatively low market share of the closest competitor;

(b) the fact that the high combined market share of Hutchison/ECT has remained stable over time;

(c) the concentration of transhipment traffic to only five ports in Northern Europe, two of which are controlled by the parties;

(d) the parties' high combined share of port calls by the major liner services on the Northern Europe - Far East and Transatlantic trades.

2. Future developments

Capacity developments

84. The parties argue that the Hutchison/ECT share of overall North European deep-sea port capacity will fall drastically in the period up to 2005 as a result of several factors.

85. The first of these factors has been mentioned above (see paragraph 69) the Maersk Delta terminal will progressively take over land areas and infrastructure which are either currently used or were intended to be used by ECT. ECT's scope for physical expansion of capacity within the current limits of the Port of Rotterdam will be restricted as a consequence. The terminal will be dedicated exclusively to Maersk Sealand until the beginning of 2007.

86. Further, ECT and P&O Nedlloyd have signed a letter of intent to establish a 50/50 joint venture terminal in Rotterdam, Euromax. Unlike the Maersk Delta terminal, the Euromax terminal will be established on "virgin" land, and thus represents a capacity addition. (73) The terminal is primarily intended to serve P&O Nedlloyd and its partners, but will also provide services to third parties as and when spare capacity arises. The terminal is expected to start operations in 2004, with P&O Nedlloyd volumes being gradually transferred from other terminals.

87. The Maersk Delta terminal represents third party business for ECT and captive production for Maersk. A non-compete clause prevents it from competing with ECT. A decision to depart from the jointly agreed pricing guidelines for the provision of stevedoring services to Maersk Sealand vessels requires the approval of both parties to the joint venture. It follows that existing capacity and capacity additions at that terminal must be split in such a way as to reflect that situation. The same applies to the Euromax terminal. With regard to the latter terminal, the situation is complicated by the fact that the terminal will, in principle, be open to third parties. However, bearing in mind that the terminal will start operations in 2004 and will require a phasing-in period, during which the transfer of P&O Nedlloyd volumes will have precedence, the Commission considers it unlikely that significant third party volumes will be handled by Euromax before the end of the year 2005. (74) Moreover, even if significant third-party volumes were to be handled by Euromax during the period under consideration, that terminal - over which ECT would have joint control - could not in any sense be regarded as a competitor of Hutchison/ECT.

88. The second of the factors liable, in the parties' view (see paragraph 80 et seq. of the Statement of Defence), to have an impact on the Hutchison/ECT share of capacity in the period leading up to 2005, is the capacity additions being planned both at other existing deep-sea terminals and at new locations. OSC expects these developments to lead to an increase in overall North European deep-sea terminal capacity from 32.25 million TEU in the year 2000 to 46.75 million TEU in 2005. The Hutchison/ECT share of that capacity (2005) is projected to be [above 20%]*, or [more than 10]* million TEU. This estimate excludes from the parties' share not only capacity relating to Maersk Delta but also capacity attributable to Euromax.

89. Table 5 shows the estimate of capacity and capacity shares in the year 2005, on the basis of the OSC data, updated by information obtained directly from the notifying parties and third parties. In the second column, the captive production of the Maersk Delta, Euromax, North Sea and MSC (Antwerp) terminals has been excluded from the market:

<table>

90. Table 5 suggests that the Hutchison/ECT share of overall capacity may be expected to decline from the present level of [above 30%]* to [below 30%]* in 2005. It should be noted that this figure is likely to be significantly understated, as both Southampton (P&O Ports) and Aarhus (Maersk Ports) are likely to have significant captive throughput. (83) The exclusion of this captive throughput from the market would have the effect of increasing the capacity share of Hutchison/ECT.

91. It should also be noted that this projection appears to assume that the levels of productivity at the ECT terminals will not increase significantly over the period in question. Given the levels of productivity already achieved at other Hutchison terminals, (84) this assumption is open to challenge. The parties' capacity share must moreover be compared with that of the closest competitors: In the year 2000 the closest competitor had a capacity share of [10-15%]*; in 2005 that share is projected to be [still within that range]*. (85)

92. Moreover, capacity alone is a poor indicator of transhipment market share (see paragraph 63) and it cannot be assumed that all new capacity will attract commensurate throughput. The example of Zeebrugge indicates that even for a port possessing substantial competitive assets such as low handling charges, good water depth and a good location in relation to the shipping lanes, it may be difficult to attract volumes when faced with competition from established major ports. (86) Most new facilities will also require a phasing in period, which may cover several years. (87)

93. Table 6 indicates what the situation would be in 2005 if, as the parties have argued in the alternative, all capacity relating to terminals jointly owned by shipping lines were to be excluded from the market:

<table>

94. This exercise leads to an insignificant increase in the parties' share of future capacity [of under 1%]*.

95. The parties have argued that new developments, such as Flushing (Vlissingen), (92) can be expected to offer strong price competition to established facilities, as the new facilities will have an incentive to price at marginal cost in order rapidly to grow the volumes needed to achieve breakeven (see inter alia paragraphs 84 and 95 of the Statement of Defence). However, the strategy of pricing below average total cost is not one that can be sustained indefinitely. Furthermore, such a strategy invites a matching response from the incumbent operators. The incumbent operators will generally be in better position to endure a protracted price war than the new operators, having had a longer period in which to amortise investments and pay back loans. In the specific case of Hutchison/ECT, it should also be noted that the concentration has the backing of the financially strong Hutchison Whampoa Group.

Switch of traffic

96. At paragraphs 112 et seq. of the Statement of Defence, the parties claim that MSC will switch all of its current transhipment traffic at Felixstowe ([...]* TEU) away from that port. This has not been confirmed by MSC, which states that it has switched [about one fifth]* TEU from Felixstowe, but has left about [four fifths]* TEU overall at that port. (93)

The impact of large vessels

97. The largest container vessel delivered in 1988 had a capacity of 4 340 TEU. In 1999, the largest vessel delivered had an estimated capacity of 7 500 TEU's. (94) OSC estimated that the typical vessel deployed on the Far East trades in 2005 will have a capacity of 5 500-6 500 TEU's. The largest vessel on those trades will have a capacity of 8 500 TEU's. (95) A typical vessel on the Transatlantic in 2005 will be able to carry 4 000 TEU, while the largest vessel will be able to carry 4 800 TEU's.

98. Scheduled deliveries for the period April - December 2001 include three container vessels in the 4 000 - 4 999 TEU range, 24 in the 5 000 - 5 999 range, 19 in the 6 000 - 6 999 range, one in the 7 000 - 7 999 range, and one in the 9 000 - 10 000 TEU range. (96) In January 2001, Maersk Sealand took delivery of the 21st of a series of 25 6 600 TEU / 104 000 DWT container vessels, while P&O Nedlloyd took delivery of the first in a series of four 6 788 TEU vessels. (97) Hapag-Lloyd has four 7 200 TEU vessels on order. (98) 6 000+ TEU vessels accounted for some 27.5 % of all existing and ordered cellular capacity as of June 2000.

99. Vessels of the latest, now well established, generation of large container vessels (6 400 - 7 500 TEU) have an overall length of between 300 and 320 metres, a beam of 42.8 metres and a draught of between 14.2 and 14.5 metres. (99) The next generation, expected to be deployed between 2003 and 2008, is predicted to have a capacity of between 8 000 and 8 500 TEU, an overall length of between 330 and 340 metres, a beam of 46-47 metres and a design draught of 14.5 metres. It should be noted that the latest generation Maersk Sealand vessels, although officially rated at 6 600 TEU, are believed to have a capacity of approximately 8 000 TEU. (100)

100. Studies (101) and third party comments indicate that the advent of larger container vessels will have a significant impact on the pattern of vessel calls at North European deep-sea container ports by leading to a further concentration in port calls (102) and a concomitant increase in transhipment and relay operations. (103) In order to be competitive, the major transhipment hubs will need to provide:

. efficient container terminal services;

. direct deep-sea access; (104)

. berth depths of at least 15 metres, accompanied by depths in access channels of around 16 metres on an acceptable tidal window;

. good hinterland connections (including intermodal services).

101. With regard to the criterion of efficiency, the results of OSC's survey of productivity levels at North European ports in 1999 are set out in Table 7:

<table>

102. Table 7 shows that Felixstowe and ECT (Delta) ranked amongst the most productive terminals in Europe in 1999. It also shows that there is a considerable gap between the levels of productivity achieved by the leading terminals (Felixstowe, ECT Delta and Antwerp-Hessenatie) and those in the second and third tiers.

103. In its report on the Trinity III Extension Development (page 20), OSC notes that another Hutchison-controlled terminal, HIT, is a world leader in terms of productivity. In the same report (page 21), OSC considers that Felixstowe and Thamesport are likely to be able to increase productivity from current levels to 1 400 TEU/metre in the not too distant future. OSC benchmark projections for the year 2005, (105) foresee a benchmark of 950 TEU/metre for terminals on the North Continent and a United Kingdom benchmark of 1 150 TEU/metre. These are only slightly higher than the levels that were achieved by Felixstowe and ECT Delta in 1999. This leads to the conclusion that Hutchison/ECT.s terminals are likely to maintain their lead in terms of productivity over most of the other North European terminals for the foreseeable future. The parties have argued (Statement of Defence, paragraph 192) that they should not be penalised by the fact that they may be more efficient than other operators. This misrepresents the Commission's position: the Commission's aim is to make an objective assessment of how market shares will actually develop (prospective analysis of market structure) - the fact that estimates of future throughput may be understated as a result of insufficient allowance for efficiency gains is obviously relevant to that assessment.

104. Five Northern European ports - Rotterdam, Hamburg, Felixstowe, Antwerp and Bremerhaven - accounted for 91 % of all Northern European transhipment volumes in 1999. Of these five ports, two, Hamburg and Antwerp, are river ports. OSC has the following comment on the competitive situation of these ports: "For the port industry, the focus is primarily on vessel draught and the depth of access channels and berths. Since the early 1990s, major river ports such as Antwerp and Hamburg have responded to these developments by deepening access channels. This has provided a temporary solution to these trends. However, there is now little scope to further improve the position - except at very high and prohibitive cost - and the net effect has been increasing restrictions on the tidal access window at major container terminals. When these considerations are linked to other disadvantages of accessing these ports - principally the time factor for transiting the rivers - this has resulted in increasing difficulties. Such factors have been cited by Maersk as the major reason for leaving Hamburg in favour of the deepsea coastal terminals at Bremerhaven, and other similar developments are noted world-wide". (106)

105. The Port of Antwerp confirms that it is "prepared" to dredge access channels to a depth of 13 metres. Further dredging would be a possibility on the basis of "further insights". No timetable appears to have been set for the start of dredging operations.

106. The parties have argued that the time penalty and draught restrictions at Antwerp have not, in practice, deterred the largest vessels from calling at the port and cite as an example the call of the Maersk Sealand vessel Knud Maersk (a "K-class", 6 000 TEU,107 vessel) at the port (see paragraph 90 of the Statement of Defence). Suffice to note in this respect, that Maersk has chosen not to call at Antwerp with its largest vessels (the "S-class" vessels of 6 600+ TEU capacity - rumoured to have an effective capacity of 8 000 TEU). A comparison between the composition of the Maersk North Europe - Asia service calling at Rotterdam108 (AE1) and that calling at Antwerp (AE2) is enlightening. The Rotterdam service makes use of two S-class vessels and three K-class vessels; the AE2 service uses two K-class vessels and no S-class vessel.

107. With regard to Hamburg, that port has traditionally had a high market share of transhipment operations - despite the time penalty and draught restrictions associated with transiting the Elbe - because of its proximity to the Scandinavian/Baltic markets. The emergence of new deepwater capacity at Bremerhaven is however eroding Hamburg's market share. Maersk's decision to base one of its European hubs (the North Sea Terminal joint venture) at Bremerhaven rather than at Hamburg is significant in this respect.

108. The table in Annex I shows the level of accessibility to the main North European deep sea ports, assuming that planned dredging programmes are carried out. The table indicates that for a vessel of 14-metre draught (vessels of approximately 6 000 TEU capacity), only Bremerhaven (after dredging), ECT Delta and Le Havre offer unrestricted access. Le Havre, although offering sufficient depth, is handicapped by low productivity and poor reliability.

109. The table in Annex I also shows that Felixstowe would be one of the few ports able to offer access on an acceptable tidal window.

110. If the vessel draught is increased to 15 metres (most 6 000+ TEU vessels have a 14.5-metre draught), only ECT Delta and Le Havre are able to offer unrestricted access, with Felixstowe among those able to offer access greater than 50%.

111. With regard to hinterland connections, OSC notes that: "... the number of intermodal rail shuttles from Rotterdam has mushroomed..." and that "[t]he location of Rotterdam in relation to the waterway system is perfect". (109) It is no coincidence that Maersk and P&O Nedlloyd have both chosen Rotterdam to be one of their European hub ports and that both parties have invested heavily in intermodal services on that basis. (110)

112. The parties argue111 that the relative strength of a port operator in respect of hinterland traffic has no impact on the carrier's choice of transhipment port. This argument appears to be inconsistent with the parties' own position that hinterland and transhipment traffic are inseparably linked and that the choice of whether or not to call at a particular port is dictated primarily by hinterland considerations. As vessel sizes increase, the hinterland aspect will become even more important. (112) Obviously, therefore, an operator enjoying a competitive advantage with regard to hinterland traffic (good road and rail connections, good geographical location) will also enjoy a competitive advantage with regard to transhipment traffic. This is a conclusion shared by independent observers (113) and confirmed by port statistics, which show that transhipment traffic is strongly concentrated in just five main Northern European deep-sea ports - which also happen to have a high share of Northern European hinterland traffic (see paragraph 53, including footnote).

Conclusion

113. It may be concluded from the above analysis of future developments and trends that the concentration will lead to the creation of a dominant position of the parties on the relevant market even in the light of future developments considered up to and including 2005.

3. Countervailing buying power

114. The parties argue that the shipping lines possess significant countervailing buying power as regards the Northern European terminal operators (see section V.1 of the Statement of Defence).

115. The parties contend that recent shipping line consolidations and alliance structures have resulted in a high level of concentration on the demand side. This is, in the parties' view, exemplified by the Northern Europe - Far East trade, where five carrier groupings account for 82 % of the total capacity (see paragraph 148 of the Statement of Defence). The concentration amongst Hutchison's customers in Felixstowe and ECT's customers is even higher than the market average with four carrier groupings accounting for [60-70%]* of ECT's total throughput and four carrier groupings accounting for almost [70-80%]* of the throughput in Felixstowe. Because of the customers. greater share of fleet and trade capacity and their ability to switch large volumes to alternative ports, the terminal operators are faced with a considerable degree of buying and negotiating power. The parties further argue that in the light of the significant economic consequences of loss of an important customer, even the threat of shifting part of the volumes serves as a sufficient constraint on the pricing behaviour of the port operators.

116. As regards the level of concentration on the market, the parties rely on previous decisions where the Commission found that a specific degree of concentration on the demand side indicated buying power. In particular, the parties quote from a section of the Decision in BP Amoco/Castrol (114) (see paragraph 144 of the Statement of Defence), where the Commission, when assessing the creation of a possible oligopoly, found that the shipping lines had a strong buying power as regards the suppliers of marine lubricants.

117. In this respect, the market structure, as well as the relevant product and the purchase conditions assessed in the decisions cited by the parties, in particular in the BP Amoco/Castrol case, bear little resemblance to those at hand in the instant case. In BP Amoco/Castrol, five large suppliers of marine lubricants were pitted against each other in a bidding process and faced significant customer switches on a market where capacity was not constrained and where the physical presence of the supplier was not essential. This is fundamentally different from the instant situation: the stevedoring service is not an isolated product that can be purchased on a geographically unlimited market, where a switch from one supplier to another has few additional consequences for the purchaser's operations. On the contrary switches to other ports are dependent on several factors and require substantial changes to the services offered by the shipping lines. (115) The shipping lines' position as regards port operators is therefore completely different from their position as regards suppliers of products necessary for the operation of the vessels, such as lubricants.

118. In any event, it cannot be accepted that the demand side is highly concentrated. (116) While members of shipping alliances may in many instances negotiate collectively with a particular terminal operator, this is not always the case (117) - whether to do so is ultimately a matter of choice for each individual member of the alliance. It should also be noted that alliance members compete with each other both on price and on service and their interests do not therefore always coincide. This applies in all areas, including the choice of terminal operator. (118)

119. The concentration on the customer side, such as it may be, must in any event be compared to the concentration existing on the supply side. This is particularly relevant concerning the Northern Europe-Far East trade upon which the parties' estimate of concentration is based. As seen above (see paragraph 104 et seq.), there is currently a limited number of terminal operators able to accommodate the largest vessels being used on this trade. The concentration on the supply side is likely to increase further in response to the market trends identified above.

120. The shipping lines' ability to switch throughput (completely or partially) to alternative ports is dependent on several factors, such as available capacity, established loops and services and appropriate hinterland, feeder and relay infrastructure.

121. At present few of the main hub ports have sufficient capacity to accommodate more than limited new volumes. This may change in the future, but it is symptomatic of the current situation that the parties themselves have stated in submissions to the Commission that the capacity which will be freed by the move of P&O Nedlloyd/Grand Alliance traffic to the new Euromax terminal will only be sufficient to grow traffic in line with market growth.

122. The investigation has furthermore revealed that the port handling charges are only one, and not the most important, factor when the shipping lines choose which port to call at. That choice is instead dictated primarily by hinterland connections, availability of feeder and relay services, direct coastal access and draught of access channels and berths as well as the timetable and schedule of the service in question. Service levels are also important. Once a shipping line has started to operate a successful service calling at a particular port, they are therefore faced with a strong incentive to continue to call at the port in question. (119) Depending on the extent of operational restructuring required in the specific instance for a port switch, there will often be a large margin of manœuvre for the terminal operators to raise the port handling charges, without this increase being sufficient to justify a change of port for the shipping lines. A switch of port would therefore only be commercially defendable if the changes did not cause significant interruption or inconvenience to the shipping line's operation.

123. The parties have argued that there is no real impediment to switching ports, stating that the costs involved are not substantial and that the switch of port does not result in any operational delays. The Commission has however received information from the shipping lines stating that switches require significant reworking of loops and that changes in terminal cut-offs would cause difficulties for their customers. The loops are furthermore often agreed upon amongst the consortia members. In order not to diminish the efficiencies gained by the consortia arrangement, shipping lines would therefore be required to take account of the interests of fellow consortia members when considering whether to switch to another port.

124. In this context, the parties argue that switches in transhipment traffic can easily be made without upsetting the liner shipping operation. Transhipment, unlike hinterland traffic, is footloose in the sense that it can be loaded or unloaded at any port at which the shipping line calls for its hinterland traffic (see paragraphs 116 and 117 of the Statement of Defence).

125. Apart from the possible need to co-ordinate transhipment moves between consortium members, the choice of transhipment port is likely to be dictated primarily by service levels, available relay and feeder connections, and geographical location. Price is not a paramount consideration (see paragraph 122). Furthermore the risk of losing some transhipment volumes is unlikely to be sufficient to constrain the behaviour of the port operators. The parties themselves have stated that transhipment traffic is secondary to hinterland traffic in importance for terminal operators, in view of the lower margins and volumes associated with transhipment.

126. The parties have provided empirical evidence of port switches (partial switches and switches of entire services) between the years 1996 and 2001 in support of their view that shipping lines can and do exercise their market power by switching demand (see paragraph 154 of the Statement of Defence). With the exception of the switch by Maersk Sealand of some transhipment traffic from Rotterdam to Bremerhaven, the parties have not provided details of these moves. In particular, the evidence submitted does not show that the switch of port has been motivated by price considerations. On the contrary, available data suggests that the switches have been motivated primarily by operational considerations (for example, MSC's partial switch of larger vessels from Antwerp to Le Havre) or restructuring following new consortium arrangements. In the case of the switch of Maersk Sealand transhipment volumes (feeder traffic to Norway and the Baltic) from Rotterdam to Bremerhaven in 2000, the prime drivers were the improvement of the feeder services and the loss of capacity at the old Delta Sea-Land terminal. The relative dearth of switches and the fairly stable market shares of the leading ports between 1997 and 1999 would tend to demonstrate that switches are not undertaken lightly by shipping lines. (120)

127. The parties argue that various developments on the market, such as the alleged decline in the level of port handling charges and the establishment of shipping line owned facilities indicate that the shipping lines have already successfully exercised countervailing buying power as regards the port operators (see section V.1.5 of the Statement of Defence). The parties contend that increases in shipping lines' freight rates over the recent years have not been matched by increases in port handling charges and that the latter have even declined in real terms. According to the parties, the shipping lines are, furthermore, effectively hindering the terminal operators from invoicing the shippers directly for container handling services. The parties argue that this would enable the terminal operators to attract new traffic by offering attractive rates. Instead, the shipping lines have introduced Terminal Handling Charges ("THC"), whereby they charge the shippers for the stevedoring services carried out by the terminal operators. According to the parties, the THC is set at a level in excess of that of the actual port handling charge, implying that the shipping lines are able to achieve mark-ups on port charges.

128. As regards the port handling charges, it is not clear from the parties' arguments that any decrease in the average port handling charge per move has been caused by shipping lines' ability to constrain or dictate the pricing on the market. It is equally plausible that this is a result of increased volumes and volume discounts and rebates. The parties have submitted a chart (121) showing the development of revenue per selected customer in the Port of Felixstowe over the period 1992 to 2000. The parties interpret this table as providing support for their thesis that handling charges have declined over that same period as a result of the increased buying power of the shipping lines. The Commission does not share that view. The table shows inter alia, that [A]* paid 21% less per container in the year 2000 than it did in 1992. However over that same period [A.s]* throughput in Felixstowe increased by some 800%. Against that background, the decrease in price is far less than that which might have been expected as a result of volume rebates and lower unit costs. (122) Another line, [B]*, increased its throughput at Felixstowe by 344% over the period in question, without any reduction in the price per container. It is true that lines such as [C]* and [D]* were paying a lower price per container in 2000 than they did in 1992, without any commensurate increase in throughput. It should however be noted that both of these lines have been experiencing severe financial difficulties - it is therefore far more likely that the decline in the rates charged to these lines is a reflection of their inability to pay rather than evidence of buying power. The Commission notes in this context that the Hutchison Whampoa ports business showed an operating profit margin of 39,7% and a profit attributable to shareholders amounting to 24,6% of turnover in 1999. In the same year, Hutchison's United Kingdom port operations showed an 11% increase in earnings before interest expense and taxation (EBIT) over 1998 figures (source: Hutchison Whampoa Annual Report 1999, pages 4 and 18). In the following year (2000), the ports division reported an increase of 11% in EBIT over 1999 figures. The Annual Report 2000 also notes that ECT has contributed positive earnings to the Group since the latter's acquisition of ECT in November 1999 (page 20 of the Annual Report). The Commission finds these statements and figures difficult to reconcile with the parties' portrayal of terminal operators as being under intense price pressure from customers.

129. As to THCs, it is not clear from the available evidence that the terminal operators have previously been able to charge each individual shipper directly or that such a system would be realistic. However, if shipping lines are in fact passing on port handling charges to their customers as an element in the THC (with a markup), this would suggest that the shipping lines would have little real incentive to switch throughput to another port solely in order to reduce port handling charges. In the Statement of Defence, paragraphs 167-168, the parties claim simultaneously and self-contradictorily (a) that THCs do not vary by port (123) (and hence that shipping lines do not or cannot pass on to their customers the benefit of any difference in handling fees as between two ports or terminals) - and (b) that they do. (124)

130. The parties also argue that the emergence of carrier solely- or jointly-owned terminals even in ports which previously refused to establish such facilities, reflects the countervailing buying power of the shipping lines. However, firstly, only a handful of such terminals (125) are currently operational in Northern Europe and all have strong participation by incumbent terminal operators. Secondly, the setting up of a jointly owned facility developed and structured in such a way as to optimise the shipping line's services also severely reduces the incentive for the shipping line to switch to another port. The shipping line is therefore to a considerable extent locked in to using the jointly-owned terminal, which would tend significantly to negate any relative bargaining power which the shipping line might otherwise have been able to exert by threatening to switch its throughput to other terminals. This is particularly true for the jointly-owned facilities in which ECT has or will have a participation: Maersk Delta and Euromax. These terminals are intended to serve as major Northern European hubs for Maersk Sealand and P&O Nedlloyd respectively and represent a major investment and commitment by these lines. (126)

VII FINAL CONCLUSION

131. For the reasons set out above it may be concluded that the concentration would lead to the creation of a dominant position on the market for the provision in Northern Europe of stevedoring services for transhipment traffic carried by deep-sea container vessels, through which effective competition in the common market would be significantly impeded.

VIII UNDERTAKINGS SUBMITTED BY THE PARTIES

132. In order to resolve the competitive concerns identified by the Commission, the parties submitted undertakings on 8 June 2001. Those undertakings are set out in Annex II. They may be summarised as follows: ECT will divest its minority shareholding in the Maersk Delta BV joint venture ("MDBV"), over which ECT has joint control, to an independent buyer approved by the Commission.

The Parties will take measures to enable MDBV to handle third party business freely in competition with ECT from the date of the Commission's decision in this case: MDBV will be released from all contractual restrictions that currently prevent MDBV from handling third party business.

The parties will cause ECT to renounce the exercise of its veto rights in MDBV. RMPM will make available to MDBV capacity to allow MDBV to handle third party business equivalent to [a certain proportion]* of MDBV's estimated total capacity for each of the years 2001 to 2005.

RMPM confirms that land and necessary facilities at the planned Maasvlakte II area will be allocated according to an open, transparent and non-discriminatory procedure.

IX ASSESSMENT OF THE PROPOSED COMMITMENTS

133. The Commission has examined the proposed commitments and considers that they ensure the possibility of significant competition in one of the main transhipment hubs in Northern Europe, without giving rise to new competition concerns. The divestiture and interim measures will allow MDBV to handle third-party business independently in the Port of Rotterdam. The Port of Rotterdam is one of a very limited number of transhipment hubs in Northern Europe. There are only five effective transhipment hubs in Northern Europe at present; it is probable that the number of hubs will be reduced to only three in the future. Rotterdam is, by virtue of its natural advantages, likely to be among the survivors. Against that background, the introduction of effective competition in Rotterdam itself can be expected to have a considerable constraining effect on any market power that the concentration might otherwise have been able to exercise. This is all the more so because the cost for a shipping line of switching from ECT to Maersk Delta127 - switching not only within the same port, but within the same area of the port - is insignificant when compared to the cost of switching traffic from Rotterdam to another port. Further, the Maersk Delta terminal is already operational and has the backing of a strong corporate group (A.P. Møller). Once the undertakings have been implemented, it will therefore be in a position to offer immediate and effective competition to ECT.

134. The obligation to make available capacity is essential to ensure that MDBV has the means at hand to become an active competitive force. Without that specific undertaking, capacity constraints might effectively curtail competition from MDBV.

135. The above aspects of the commitments (a), (b) and (c) constitute conditions, as only by fulfilling these undertakings can the desired structural change on the relevant market be achieved.

136. The remaining aspects of the undertakings with the exception of (d) constitute obligations, as they concern the implementing steps necessary to achieve the structural change that is sought. In particular this relates to:

(a) the provisions relating to MDBV.s freedom to determine tariff, tariff structure and pricing and freedom to compete with other stevedoring companies as described in paragraphs 4 (b) and (c) of Annex II;

(b) the provisions relating to the trustee in paragraphs 7 to 12 of Annex II.

137. In this context, the Commission takes note of RMPM.s assurance that land at the Maasvlakte II development (on which construction may start in 2007) will be allocated in a fair, transparent and non-discriminatory manner (commitment (d) in paragraph 132). MDBV will thus have the same opportunity as other Rotterdam terminal operators to expand its capacity in the period after 2005.128

138. Taken as a whole, the undertakings submitted by the parties will prevent the emergence of a dominant position on the market for the provision in Northern Europe of stevedoring services for transhipment traffic carried by deep-sea container vessels.

139. In conclusion, the Commission considers that the undertakings submitted by the parties are sufficient to address the competition concerns raised by this concentration,

Has adopted this decision:

Article 1

Subject to full compliance with the conditions and obligations, as set out in Annex II, the notified operation whereby Hutchison Ports Netherlands BV and Rotterdam Container Participatie Maatschappij BV acquire joint control of Europe Combined Terminals BV is hereby declared compatible with the common market and the functioning of the EEA Agreement.

Article 2

This decision is addressed to:

Rotterdam Container Participatie Maatschappij BV Galvanistraat 15 PO Box 6622 3002 AP Rotterdam

Hutchison Ports Netherlands BV Nolst Trenité Plaza Weena 666 PO Box 190 Rotterdam 3000 AD

Notes:

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

2 OJ L 180, 9.7.1997, p. 1.

3 OJ

4 OJ 13, 21.2.1962, p.204/62.

5 OJ L 148, 16.6.1999, p. 5. This text is made available for information purposes only and does not constitute an

6 In the following "RMPM" and "RCPM" are used interchangeably, as RCPM is controlled by RMPM and has no business activity of its own.

7 OJ C 66, 2.3.1998, p.5.

8 For example, to secure the completion of the Delta 2000-8 development, something in which both Hutchison and RMPM had a clear interest.

9 Provided in the context of Cases COMP/37.688/D2 and COMP/JV.52, which information the parties have agreed may also be used for the purposes of the present case.

10 Save those affecting basic shareholder rights, where a 75% majority is required.

11 ABN is entitled to two seats, but has filled only one.

12 ABN increased its stake from 14% to 28% (a level of shareholding that would allow it, theoretically, to form a voting majority with either of the two strategic investors) at the very last minute, following the abrupt withdrawal of the financial institution ING.

13 Available capacity is becoming saturated.

14 An opinion shared by other knowledgeable observers: Baron Delwaide, president of the Antwerp Port Authority, is thus quoted as saying .There will be some shipowners which will choose Rotterdam because of Hutchison. (Lloyd's List of 18.5.2001, annex 8 to the parties. Statement of Defence of 30.5.2001).

15 In view of these lines' commitment to the European Rail Shuttle, the need to co-ordinate service patterns and schedules with alliance partners, and shipper preferences, it may be doubted whether it would be commercially viable for Maersk and P&O to transfer the bulk of their operations away from Rotterdam.

16 [Quote from minutes of a meeting of ECT's Supervisory Board]*

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

17 [Quotes from submissions to and correspondance with the Commission]*

18 Figures for the financial year 1999. Turnover calculated in accordance with Article 5(1) of the Merger Regulation and the Commission Notice on the calculation of turnover. To the extent that figures include turnover for the period before 1 January 1999, they are calculated on the basis of average ECU exchange rates and translated into EUR on a one-for-one basis.

19 Form CO, page 25, .Relevant market..

20 223 pages, published by OSC in the United Kingdom in 2000.

21 See, inter alia, pages 12-13 and Section 6 of North European Containerisation. The report identifies four regional markets: North Continent West (Rouen to Amsterdam), North Continent East (Hamburg/Bremerhaven), United Kingdom/Ireland and Scandinavia/Baltic.

22 For the reasons given at paragraph 25 above the Commission distinguishes between services provided to deep-sea container vessels and those provided to short-sea vessels and non-unitised (containercarrying) vessels. The activity (on which an overlap occurs) exercised by the parties to the concentration consists in the provision of stevedoring services to deep-sea container vessels.

23 North European Containerisation, page 13.

24 The parties state that the charge per transhipment move is on average [considerably]* lower than that for a hinterland move.

25 Form CO, page 53.

26 See, inter alia, the parties' response to the Commission's questionnaire of 20 March 2001.

27 North European Containerisation, pages 176-177.

28 Despite the Commission's efforts to provide unambiguous definitions of these traffic flows in its questionnaires to the notifying parties and third parties.

29 Statement of Defence of 30 May 2001 ("Statement of Defence"), at paragraphs 32, 33, 34 and 37.

30 North European Containerisation, pages 12-13. Drewry Shipping Consultants Ltd (DSC), in its report European Container Port Market Data (Dec. 2000), uses a slightly different breakdown: United Kingdom, Benelux, Germany/Scandinavia, France.

31 Felixstowe, Southampton and Thamesport are relevant for this case.

32 The other main ports present on this market are: Wallhamn, Oslo, Bergen, Helsinki, Kotka, Hamina, Rauma, St Petersburg, Tallinn, Riga, Ventspils, Klaipeda, Gydnia, Szczecin and Gdansk (see North European Containerisation, pages 146-149).

33 A view shared by OSC - see North European Containerisation, page 23.

34 Page 1867 of the file.

35 Page 2884 of the file.

36 Page 3741 of the file.

37 Pages 2803 and 2804 of the file.

38 Page 2315 of the file.

39 Page 2316 of the file.

40 See, inter alia, OSC Briefing Paper, page 9 (.Review of market share.).

41 A view confirmed by market studies (for example, Briefing Paper and North European Containerisation), by the notifying parties and by third parties (customers and competitors).

42 Felixstowe, Thamesport, Southampton, Le Havre, Zeebrugge, Antwerp, Rotterdam, Bremerhaven, Hamburg, Gothenburg.

43 The Maersk Sealand terminal had an estimated capacity in 1999 of 2 100 000 TEU (twenty-foot equivalent units - the most common container size). The other Algeciras terminal, Isla Verde, had an estimated capacity of 200 000 TEU (Source: Drewry, European Container Port Market Data, December 2000, page 24).

44 Source: A.P. Møller (reply to Commission's request for information).

45 The Algeciras facility is one of the main European hubs of Maersk Sealand and forms an essential link in that line's shipping operations. There is no evidence that Maersk Sealand would contemplate interrupting and inconveniencing its own shipping operations in order to create additional capacity to accommodate third party volumes.

46 See, inter alia, North European Containerisation, page 23.

47 North European Containerisation, page 176.

48 OSC Briefing paper, page 13.

49 [Throughput estimates by OSC]*

50 [Calculation of the increase]*

51 A post-merger HHI above 1800 is regarded as evidence of a highly concentrated market (see US Department of Justice Merger Guidelines - 1992 version).

52 [Calculation of the HHI]*

53 Figures for 1999 adjusted for information provided by terminal operator / port authority.

54 Briefing paper, page 11.

55 In this context it should be noted that the same five ports have a share of "only" 75% of North European hinterland traffic. The level of concentration with regard to transhipment is thus much higher than the level of concentration with regard to hinterland. It should further be noted that four ports (Rotterdam, Hamburg, Felixstowe and Antwerp) accounted for approximately 81.5% of transhipment traffic, but "only" 67% of hinterland traffic [Source: OSC Briefing Paper, table 10].

56 Source: Maersk Delta business plan.

57 Overall growth or contraction of demand, new entrants (and exits) on the upstream (terminal services) and downstream (liner shipping) markets, restructuring of shipping alliances and services, switches between terminals, increase or decrease of productivity at individual terminals, etc.

58 North European Containerisation, page 185.

59 The source quoted is .Ocean Shipping Consultants January 2001. That report (referred to in this document as the "Briefing Paper") does not contain the figures in question. In the table, ECT is shown as having handled [less than 4 000 000]* TEU in 2000 - this figure differs from that provided by Hutchison Whampoa in its Annual Report 2000 (page 20), where ECT is said to have handled [4 200 000-4 500 000]* TEU in 2000, i.e. an increase of some [250 000-350 000]* TEU over 1999 throughput.

60 It should first be noted that the table in paragraph 102 of the Statement of Defence is mathematically flawed (different market totals in columns 1 and 2). The actual transfer of [less than 75 000]* TEU is already taken into account in the year 2000 actual throughput figure for ECT - there is thus no need for any further correction to that figure.

61 The parties' "figures" have been obtained by increasing the 1999 actual throughput of the respective ports by 6 % (estimated annual increase of demand in Northern Europe). There is no independent confirmation (for example, port statistics, market studies, information from port or terminal operators) of these "figures".

62 That is to say, an increase in transhipment throughput of [around 25 000]* TEU over 1999 figures, despite the switch of [250 000-350 000]* TEU to Bremerhaven in 2000.

63 Source: A.P. Møller (reply to Commission's request for information).

64 North European Containerisation, page 207.

65 European Container Port Market Data, page 6.

66 For competitors' shares see Table 3 below.

67 It is precluded from doing so by agreement between the joint venture partners.

68 Although the parties have argued that ECT does not have joint control, the joint venture agreement unequivocally requires unanimous approval of any changes to the jointly agreed pricing policy set out in the agreement. This pricing policy applies to the provision of services to the Maersk Sealand shipping line, i.e. the only services that will be provided by Maersk Delta BV until 2007. ECT's control through this veto right is reinforced by a non-compete clause which explicitly prohibits Maersk Delta from competing with ECT for third-party business. Joint control is also assured by veto rights over major investments and acquisitions - these veto rights go beyond that which is required to safeguard basic shareholder rights (which are in any event protected by a specific provision in the joint venture agreement).

69 It is common practice within corporate groups to invoice internal services at cost price with no, or only an insignificant, mark-up.

70 The Commission has taken the year 2005 as the cut-off date for taking account of developments in the market. Developments beyond that date are generally too uncertain to be relied on in the present assessment.

71 Figure drawn from the OSC/Drewry reconciliation submitted by the parties. This figure does not include any capacity relating to the Maersk Delta terminal so no further adjstment to exclude Maersk Delta capacity is necessary.

72 [Details regarding the North Sea Terminal]*

73 Carel van den Driest (ECT CEO), quoted in P&O Nedlloyd press release 14.11.2000: "In addition Euromax will ensure room for growth for current and new customers at the [ECT] Delta Terminal for many years to come".

74 The Maersk Delta terminal will require a five-year phasing-in period - it is reasonable to assume a similar phasing-in period for Euromax.

75 JV between MSC and Le Havre on a 50/50 basis. [...]*

76 Referring to the parties. arguments in their Statement of Defence (paragraph 94 and the footnotes contained therein) the following should be noted: The Annex to the answer provided by the Port Authority of Antwerp (pp. 2260-2262 of the File) shows the expansion plans. As it is clear from the cover letter and the Annex, only the first two phases are under construction and can reasonably be expected to be completed by 2005. For the 3rd phase of Deurganckdock east and west, construction has not yet been ordered (page 2262). It is also clear from the answer provided by the Antwerp Port Authority that the original timing is no longer realistic, as the Parliament has suspended the expansion plans (see answer 4 on page 2261). Taking the above into consideration, one can add the capacity additions resulting from the 1st and 2nd phase expansions at Deurganckdock east and west to the capacity of 2000. [...]*

77 [...]*

78 [...]*

79 [...]*

80 [...]*

81 [...]*

82 Information provided by Port Authority.

83 The main container terminal in Southhampton is jointly owned by P&O Ports. Southampton is also one of the main hubs of the P&O Nedlloyd shipping line. A significant proportion of the throughput of Southampton is therefore likely to constitute captive production (provision of services by P&O Ports to P&O Nedlloyd). The same applies for Aarhus, which is controlled by A.P. Møller (Maersk) and is one of the main European hubs of the Maersk Sealand shipping line.

84 See paragraph 103.

85 Briefing Paper, table 14: Hessenatie is forecast to have a [10-15%]* share of capacity in 2005 (NB: this figure includes captive capacity relating to the Hessenatie/MSC joint venture - the figure is thus overstated).

86 North European Containerisation, page 55: "... demand growth has been disappointing".

87 The business plan for the Maersk Delta joint venture envisages a five-year phasing in period.

88 MSC capacity excluded [...]*

89 Total JV Hessenatie/MSC capacity excluded [...]*

90 Maersk Delta and Euromax capacity excluded [...]*

91 Capacity at North Sea Terminal excluded (Maersk and Eurogate][...]*

92 With regard to developments at Vlissingen, the Commission notes the following information in the 1999 Annual Report of RMPM (page 13): "Plans are being made in Flushing (Vlissingen) for the construction of the Westerschelde Container Terminal. The intended operator is the Belgian company Hessenatie. If the plans go ahead, construction and operation of the infrastructure will probably be executed on the instruction of Exploitatiemaatschappij Schelde Maas (ESM). Zealand Seaports and the RMPM each have a 50 percent stake in ESM. (emphasis added).

93 See MSC reply, page 2326 of the File.

94 North European Containerisation, page 209.

95 It may be useful to contrast these more recent estimates with an earlier report (The Competitive Position of Felixstowe versus Rotterdam Terminals, September 1998) in which OSC predicted that a typical Far East vessel in 2010 would have a 6 500 TEU capacity, while the largest vessel would be of 8 000 TEU size.

96 DynaLiners 13/2001, 30 March, page 5.

97 Containerisation International, March 2001, page 27.

98 North European Containerisation, page 207.

99 North European Containerisation, page 210.

100 See, inter alia, Lloyd.s List of 3 April 2001, page 18.

101 For example North European Containerisation, page 213, Trinity III Extension Development, pages 21 et seq. (a report prepared by OSC for Hutchison).

102 Significant concentration has already taken place in respect of the Far East trades (on which the largest vessels are deployed): in 1995 Felixstowe, Rotterdam and Hamburg accounted for 34% of all calls in Northern Europe by Europe - Asia liner shipping services; by 1999 this proportion had increased to 61%.

103 According to press reports (Lloyd's List, 17 May 2001, page 5) Lloyd's Register and OSC both take the view that Ultra Large Container Ships (ULCS) of 12 500 TEU capacity may well be deployed by the middle of this decade. This deployment would be driven by the significant cost savings that the ULCS are expected to provide (up to 19% over an 8 000 TEU vessel). It should be noted that these cost savings are predicated on a reduction in the number of port calls (as compared, presumably, with the number of port calls currently made by the average 8 000 TEU vessel).

104 See, inter alia, North European Containerisation, pages 206-207.

105 North European Containerisation, page 222.

106 North European Containerisation, pages 206-208. See also page 183, where OSC comments, regarding vital requirements for transhipment hubs: .The entire economics of transhipment are based upon scale economies for deep-sea vessels. It is thus vital that these largest vessels can be berthed on any tide with minimum disruption. Problems in this regard have handicapped both Antwerp and Hamburg in recent years. These requirements can only increase as very large vessels further consolidate market share and the economics of Ultra Large Container Ships come to the fore..

107 Official rating; the parties contend that the actual capacity is considerably greater.

108 The other North European ports called at by AE1 are Felixstowe and Bremerhaven.

109 North European Containerisation, pages 59-60.

110 Joint investment in the European Rail Shuttle.

111 Statement of Defence, paragraph 36.

112 The economics of the largest vessels are predicated on spending as much time at sea as possible - it follows that only those ports that are able to generate sufficient hinterland volumes will be contenders for calls by these vessels and hence actual and potential competitors on the transhipment market.

113 See for exampe, North European Containerisation, at page 213.

114 Commission Decision in Case No. COMP/M.1891- BP Amoco/Castrol, OJ C301, 21.10.2000, p. 23

115 See paragraphs 120 and 123 to 126.

116 See to this effect Commission decision in Case M.1651 - Maersk/Sea-Land, OJ C 313, 30.10.1999, p. 6.

117 Thus for example, Eurogate: "With our shipping line customers we have individual contracts as well as contracts that have been commonly negotiated and signed individually by each shipping line". [Translation of German original]. And further (Gothenburg): "Negotiations are done individually with the shipping companies".

118 As an example, P&O Nedlloyd has chosen to establish a hub in Rotterdam; Hapag-Lloyd has instead invested in Hamburg. Both operators are members of the Grand Alliance. The parties themselves point out that there will be no obligation on Grand Alliance members calling at Rotterdam to use the joint ECT/P&O Nedlloyd Euromax facility in that port, even though P&O Nedlloyd may wish them to do so.

119 P&O Nedlloyd (reply to Commission's questionnaire): "... whilst, in theory, it is open to carriers to switch ports to obtain a wider choice of stevedoring service providers, in practice this is difficult. Apart from the issue of lack of port substitutability, the more calls a carrier makes at a given port the more arrangements it will have in place to deal with these calls. [...] Therefore, once a carrier has established ports at which it calls, it becomes economically more difficult for the carrier to switch ports for a significant portion of its cargo and for transhipment activities. Further, as vessel sizes increase, this will bring about a corresponding decline in the number of port calls a carrier will make and this will again serve to increase the importance of the ports of Rotterdam and Felixstowe". And further, CP Ships: "Flexibility is limited as service quality and cost is very much influenced by a lines infrastructure and it can be quite difficult to shift resources, or find new resources, to accommodate a change of ports".

120 In a Lloyd's List special report of 25 April 2001 ("World Top Container Ports", page 6), the managing director of Maersk Benelux (joint venture partner in Maersk Delta) is quoted as saying that the switch of Maersk transhipment volumes from Rotterdam to Bremerhaven .cost a lot of money. (emphasis added) and that there is therefore no guarantee that the volumes diverted to Bremerhaven will return to Rotterdam. This contradicts the notifying parties' assertion that switches of transhipment volumes can be made easily.

121 Electronic mail of 10 April 2001.

122 It should be recalled that the greater part of the costs of a terminal operator are fixed (infra- and superstructure costs).

123 "Any gain a shipping line achieves from reduced port charges as a result of a switch of port would directly benefit the shipping line concerned as the shipping line would not be able to vary the THC from the level established by the freight conference" (paragraph 168). This implies that the THC is the same in both ports (the port from which the switch is made and the port to which the switch is made).

124 "[The shippers] will choose port by reference to the THCs. (paragraph 167).

125 Maersk Delta (Rotterdam), North Sea Terminal (Bremerhaven). Possibly also Southampton (P&O Ports) and Aarhus (Maersk). The Altenwerder terminal in Hamburg has a 25% participation by the Hapag Lloyd shipping line. It is questionable whether Altenwerder should be included in the ranks of jointly (carrier) owned terminals.

126 It should also be noted that these two carriers are tied to Rotterdam through their investment in the European Rail Shuttle, which provides regular shuttle trains (container loaded) to several European destinations, such as Milan, Mainz and Mannheim, Prague and Bratislava.

127 For example, in reaction to an attempt by ECT to increase prices.

128 The Maasvlakte II (approximately 2500 ha) is the only significant capacity development (in Rotterdam) foreseen for the period after 2005.

Annex 1

Use Spring H eigh ts & Range

Assume w indow = th e full p e riod o f time w hen the req uired d ep th is ava ilable

<tables>

NOTES

UR = Unrestricted N/A = Insufficient Depth

Hamburg: Depth ex Chart latest small correction 0669/01. Cuxhaven primary port data used. UKC assumed.

Hamburg 2: Calculated for projected channel depth.

Bremerhaven: Depth ex chart latest small correction 0669/01. Wilhelmshaven primary port data used. UKC assumed.

Bremerhaven 2: Calculated for projected channel depth.

Rotterdam ECT (Delta): Assumed.

Rotterdam ECT (Home): Depth & UKC given by Maassluis 19/02/01.

Rotterdam (Other): Depth appear to remain as for ECT (Home) until Waalhaven. Berth depths vary considerably and have not been taken into account.

Antwerp 1: As previous information

Antwerp 2: Tide table for Antwerp (Prosperpolder) used. Spring tide curve used which allows shorter period than Neap curve.

Antwerp 3: Tide table for Antwerp (Prosperpolder) used. Spring tide curve used which allows shorter period than Neap curve. Calculated for projected channel depth.

Zeebrugge: As per previous information.

Zeebrugge 2: Calculated independently of previous information, using same criteria as for other ports.

Le Havre: UKC assumed.

Southampton: UKC assumed.

Thamesport

Thamesport 2: Calculated for projected channel depth.

Felixstowe

These times were calculated by Hutchison Marine Department, can vary by 2 - 3 percentage points depending on how you interpret the tidal graph.

Method Used

<table>

Annex 2

Non Confidential Version

Re: undertakings in Case No COMP/JV.55 - Hutchison/RCPM/ECT

I Introduction

1. On 31 January, 2001, pursuant to Article 4 of Council Regulation (EEC) No 4064-89 (the .Merger Regulation.) Hutchison Ports Netherlands BV and Rotterdam Container Participatie Maatschappij BV have notified their acquisition of 35 percent each of the shares in ECT Beheer BV. The latter holds 100 percent of the share capital in Europe Combined Terminals BV ("ECT"). The notification has been submitted without prejudice to the notifying parties' view that the notified transaction does not constitute a concentration.

2. In connection herewith and with the same reservation expressed in their notification, the notifying parties and Rotterdam Municipal Port Management ("RMPM") (collectively referred to as "the Parties'") make the following commitments pursuant to Article 8(2) of the Merger Regulation. The commitments are subject to the Commissions finding that the notified transaction is compatible with the common market pursuant to Article 8(2) of the Merger Regulation (the "Decision").

II DIVESTITURE OF ECT.S SHAREHOLDING IN MDBV

3. The Parties undertake that ECT will divest its shareholding in Maersk Delta BV ("MDBV") to an independent buyer approved by the Commission within a maximum period of [...]* from the date of the Decision. This period may be extended by an additional period of maximum [...]* at the Parties. reasoned request, provided that the trustee appointed pursuant to section 3 below can satisfactory establish that ECT has, despite its best efforts, been unable to carry out such divestiture within the initial period. When considering such extension of the deadline for the divestiture of the shares, the Commission shall take into account the procedural process set out in Article 4 of the joint venture agreement between ECT and Maersk Benelux BV (the Joint Venture Agreement).

III RELATIONS WITH MDBV

Changes to contractual restrictions

4. From the date of the Decision the parties undertake that:

(a) In order to permit MDBV to handle third party business, the Parties will immediately and unconditionally release MDBV from any contractual restriction, which currently prevents MDBV from handling third party business and inform MDBV thereof.

(b) The Parties will cause ECT to renounce the exercise of ECT's veto rights in MDBV and confirm MDBV.s freedom to determine tariff, tariff structure and pricing.

(c) The Parties will cause ECT to confirm MDBV's freedom to compete with other providers of stevedoring services, including ECT.

Assurance of available capacity for MDBV

5. RMPM is committed to enable MDBV to develop such capacity as to allow it to handle third party business equivalent to [a certain proportion]* of MDBV.s total capacity for each of the years 2001 to 2005. RMPM undertakes to make available to MDBV at normal business conditions the land, facilities, barge loading areas and quay walls necessary to achieve this objective. For the purpose of calculating MDBV's need for capacity, [a certain document]* will be the relevant starting base. Also and in order to ensure that no undue advantage is taken from RMPM's commitment, calculations will disregard upward capacity not stemming from natural growth, such as acquisitions or excessive traffic flows originally handled in other ports and switched by [a certain shipping line]* to [a certain terminal]*.

6. RMPM confirms that land and necessary facilities at the planned Maasvlakte II area will be allocated according to an open, transparent and non-discriminatory procedure.

IV TRUSTEE1

Appointment of a trustee

7. No later than two weeks after the Commission has adopted the Decision, the Parties shall propose to the Commission a trustee suitable to monitor and implement the commitments referred to above. The trustee must be independent from the Parties. The appointment of the proposed trustee is subject to approval of the Commission.

8. Should the Commission reject the proposed trustee, the Parties will propose the name of a new trustee within 2 weeks after being informed of the rejection. If the new trustee is rejected by the Commission, the Commission shall nominate a suitable trustee whom the Parties will appoint or cause to be appointed.

Mandate of the trustee

9. Within 1 week of the date on which the Commission has approved a trustee, the Parties shall enter into a mandate agreement with the trustee. The mandate agreement shall confer on the trustee all the rights and powers necessary to exercise its tasks and describe the obligations and the remuneration. The terms of the mandate agreement will have to be approved by the Commission.

10. The trustee shall monitor and advise the Commission on the implementation of this commitment and ensure its compliance by the Parties. The trustee will provide, every 6 month, written reports to the Commission on the progress of the discharge of its duties of its mandate, identifying any respect in which it has been unable to discharge its mandate. The parties shall receive a non-confidential copy of such reports.

11. If the Parties have not entered into a binding contract with a purchaser approved by the Commission within the time period specified above (including the maximum additional time period), or have not completed the sale within a further period of 1 month, the trustee shall have the power to carry out the divestiture within a further time period of 2 months. The trustee shall carry out the divestiture on the best possible terms and conditions in accordance with good business practice. He is not bound by any minimum price.

12. The parties shall provide the trustee with all such assistance and information as the trustee may reasonably require.

13. The trustee is bound to respect the confidentiality of all information provided to him or otherwise obtained by him in the course of the exercise of his mandate. Notwithstanding the trustee's overall responsibility to carry out his task as an independent unrelated third party, the trustee shall have, to the extent possible given the nature of his task, due regard to the commercial interest of the Parties and undertake in the mandate to do so.

Note:

1 The following provisions are drawn up in accordance with the Commission Notice on remedies acceptable under Council regulation (EEC) 4064-89 and under Commission Regulation (EC) No 447-98, notably paragraphs 53-57 thereof.