61994C0068

1 Is it possible to apply the provisions of Council Regulation (EEC) No 4064/89 of 21 December 1989 on the control of concentrations between undertakings (1) (hereinafter `the regulation') to oligopolies? And what, in the light of the provisions of the regulation, are the limits to which the Commission is subject in appraising the lawfulness of concentrations?

Cases C-68/94 and C-30/95 provide the Court with an opportunity to answer those questions, among others, which are undoubtedly relevant for the purposes of applying the regulation. Those actions (one brought by the French Government and the other by the Société Commerciale des Potasses et de l'Azote (hereinafter `SCPA') and the Entreprise Minière et Chimique (hereinafter `EMC')) raise the issue, although from partially different perspectives, of whether Commission Decision 94/449/EC of 14 December 1993 (2) (hereinafter `the decision') is lawful.

2 In that decision, the Commission authorised a joint venture between Kali und Salz AG (hereinafter `K+S') and the Treuhandanstalt (hereinafter `Treuhand') aimed at combining the potash and rock-salt activities of K+S and Mitteldeutsche Kali AG (hereinafter `MdK'). The declaration of compatibility was, however, subject to compliance by those undertakings with certain conditions which consisted essentially in discontinuing their commercial links with SCPA and EMC.

The action brought by SCPA and EMC (Case C-30/95) specifically seeks the annulment of that part of the decision in which it imposes those conditions; the action brought by the French Government (Case C-68/94), on the other hand, also deals with other aspects and therefore has wider implications. Nevertheless, since the objectives of the two actions are only partially different and a number of the pleas and arguments put forward are largely identical, I consider it possible and indeed preferable, subject to the necessary distinctions, to deal with them together.

3 However, before recalling the terms in which the decision is couched and thus examining a dispute which, because of the issues of principle which it raises, may represent an important stage in competition case-law, I consider it appropriate to recall the background to the regulation, its essential features and the procedural framework within which control of concentrations is achieved. It will thus be easier to focus on certain general aspects which will then be referred to when the various issues of admissibility and substance raised by the actions in question are examined.

The regulation: background and content

4 Unlike the ECSC Treaty (Article 66), the EC Treaty does not contain any specific provisions on the subject of concentrations between undertakings. That gap is the result of a specific political determination to encourage the strengthening, even through concentrations, of Community undertakings. The Commission itself, in a memorandum of 1 December 1965, stated that it was, in principle, in favour of concentrations since it considered them expedient for strengthening European industry which was still too weak to compete on the international markets. In the same memorandum, it also made clear its intention to apply Article 86 rather than Article 85 to concentrations. That approach was endorsed by the Court which, in its judgment in the `Continental Can' case, recognised the existence of an abuse within the meaning of Article 86 when an undertaking, already in a dominant position, reinforced that position, for example through a merger, `in such a way that the degree of dominance reached substantially fetters competition'. (3)

However, the application of Article 86 could not represent an entirely satisfactory solution, if only because that provision allows action to be taken only retrospectively, and therefore only in relation to pre-existing dominant positions, whereas a system for controlling concentrations should, above all, prevent any anticompetitive effects. The Commission, therefore, not long after the judgment in the `Continental Can' case, submitted to the Council a first proposal for a regulation to that effect. Moreover, its subsequent practice indicated a willingness to reconsider the application of Article 85 to concentrations, using the mechanism of `prior appraisal'. (4) That trend seemed to be confirmed by the judgment in the `Philip Morris' case, (5) in which the Court accepted that the acquisition by an undertaking of a minority holding in the capital of a competitor undertaking might, in certain circumstances, give rise to a restriction incompatible with Article 85.

5 That judgment fuelled the debate on the subject and gave renewed impetus to the Commission which submitted a further proposal in 1988. It was on the basis of that proposal that the Council adopted the regulation, the final wording of which does not conceal the at times profound disagreements which marked its drafting stage. (6)

Even with the adoption of the regulation, the favourable view of forms of merger which were designed to enable European industry to compete effectively on the international market was not abandoned. Indeed, the preamble to the regulation stresses that concentrations, which are bound to increase as a result of the dismantling of internal frontiers in order to achieve the single market, must be welcomed as being in line with the requirements of dynamic competition and capable of increasing the competitiveness of European industry, improving the conditions of growth and raising the standard of living in the Community. However, the favouring of mergers and industrial reorganisation must not result in lasting damage to competition, so that Community law must be able to act against `concentrations which may significantly impede effective competition in the common market or in a substantial part of it' (fifth recital).

6 The regulation is based primarily on Article 87 since it contains detailed rules for the application of Articles 85 and 86. However, since, as also stated in the sixth recital, `Articles 85 and 86 are not ... sufficient to control all operations which may prove to be incompatible with the system of undistorted competition envisaged in the Treaty', it also has Article 235 as a legal basis.

Concentrations are defined in Article 3(1) by recourse to a very broad concept which, in addition to mergers, also covers situations in which the parties concerned may retain their legal, and even economic, autonomy. (7) The primary characteristic of a concentration, apart from cases of merger in the technical sense, is the acquisition of control, defined by Article 3(3) as the `possibility of exercising decisive influence on an undertaking' by `ownership or the right to use all or part of the assets of an undertaking' or by `rights or contracts which confer decisive influence on the composition, voting or decisions of the organs of an undertaking'. Operations falling within the scope of Article 3 are covered by the regulation only if they have a `Community dimension'. (8) Other cases of concentration fall within the jurisdiction of the national authorities.

7 Article 2(1) requires the Commission to take into account, with a view to establishing whether the concentration in question is compatible with the common market, `the need to maintain and develop effective competition within the common market in view of, among other things, the structure of all the markets concerned and the actual or potential competition from undertakings located either within or outwith the Community'; and a series of factors used to ascertain the actual market power of the undertakings involved. (9) The list contained in the provision leaves a wide discretion to the Commission, partly because it does not place the various factors to be taken into consideration in any order of priority. It is also significant that the regulation raises no presumptions of unlawfulness connected with the market shares held by the undertakings and no mathematical criteria that can be used for the appraisal, in contrast to other antitrust legislation. (10)

Article 2(2) and (3) provide that the appraisal of compatibility carried out on the basis of the aforementioned criteria is to depend, in the final analysis, on whether the concentration in question creates or strengthens a dominant position `as a result of which effective competition would be significantly impeded in the common market or in a substantial part of it'. (11)

8 The control procedure is divided into two stages. During the first stage, which begins with notification (12) and lasts one month, the Commission examines whether the concentration raises serious doubts as to its compatibility with the common market. (13) If that is not the case, or if the concentration does not fall within the scope of the regulation, the procedure ends at the first stage with the Commission's decision (Article 6(1)(a) and (b) of the regulation). In the absence of a decision, the concentration is deemed compatible with the common market (Article 10(6)).

Where, on the other hand, the concentration raises serious doubts and the Commission therefore decides to initiate formal proceedings (Article 6(1)(c) of the regulation), the decision must be taken within four months, otherwise the concentration is deemed compatible. The (final) decision may declare the concentration compatible with the common market, subject, if necessary, to compliance with certain conditions and obligations (Article 8(2)); or declare it incompatible. In the latter case, where the concentration has already been implemented, the Commission may require any action that may be appropriate in order to restore conditions of effective competition (Article 8(4)).

The procedure and the decision

9 On 14 July 1993, in accordance with Article 4(1) of the regulation, the Commission received notification of a proposed concentration between the German company K+S, a subsidiary of the chemicals group BASF, and MdK, a company in which all the potash and rock-salt extraction and processing activities of the former German Democratic Republic (GDR) are combined, whose sole shareholder is the Treuhand, an institution incorporated under public law whose task is to restructure the former GDR's State-owned enterprises.

It is appropriate at this point to make clear that, at the time of notifying the proposed concentration, K+S was by far the largest European producer of potash, with an output of more than 2 000 000 tonnes per year, followed by MdK with 1 430 000 tonnes per year. In total, therefore, German potash production in the year immediately preceding the concentration (1992) reached a level of 3 500 000 tonnes. In third place, with an output of around 1 000 000 tonnes per year, was Mines de Potasse d'Alsace (hereinafter `MPA'), owned by the French public corporation EMC which, moreover, holds 100% of the capital of SCPA, the company which markets the products in question. They were followed, with outputs below one million tonnes per year, by three Spanish undertakings, recently grouped together in a State-owned corporation known as INI, which market their own products through the same company, Coposa; and, finally, by the British company Cleveland Potash Limited (hereinafter `CPL').

10 Under the concentration plan, all the potash and rock-salt extraction and processing activities of K+S and MdK were to be transferred to Mitteldeutsche Kali AG GmbH (hereinafter `MdK GmbH'), a joint venture between K+S and the Treuhandt. K+S would have contributed to the joint venture all its own activities in the sector concerned, while the Treuhand would have contributed liquid capital of over DM 1 milliard. K+S was to hold 51% and the Treuhand 49% of the capital and voting rights.

After examining the proposal, the Commission decided, on 16 August 1993, in pursuance of Article 6(1)(c) of the regulation, to initiate formal proceedings since the concentration raised serious doubts as to its compatibility with the common market. The Commission therefore adopted, on 14 December 1993, the decision authorising the concentration, (14) making it subject, however, in pursuance of Article 8(2) of the regulation, to compliance with certain commitments.

11 The relevant product market, as identified in the decision, mainly concerns potash products for agricultural use. (15) Those products are mineral fertilisers which are either used on their own or applied to the soil together with other nutrients, particularly nitrogen and phosphate, in so-called compound fertilisers. In both uses, potash cannot be replaced by other nutrients. In the Commission's opinion, whatever their content (differing quantities of potash), form (standard or granulated) or intended use (agriculture or industry for the production of compound fertilisers), those products are, objectively, largely interchangeable as far as the customer is concerned, so that they can be considered together. (16)

The Commission also identified a second relevant market, distinct from that for potash products, which includes all magnesium products and, moreover, found that in that sector only K+S and MdK are really active on the Community market as a whole. (17) However, while reaching the conclusion that, following the merger, those companies would enjoy a monopoly in the Community as far as magnesium sulphate and kieserite were concerned, and that the merged company being formed would therefore have a dominant position on that special market, with a share of 92%, (18) the Commission did not draw any consequences from it.

12 With regard to potash products for agricultural use, on the other hand, on which the decision is focused almost entirely, the Commission drew a distinction between two relevant geographical markets: Germany and the rest of the Community. (19)

With regard to the German market, the Commission found that the concentration in question would have strengthened the dominant position of K+S, leading to a de facto monopoly. (20) Noting the insolvency of MdK which would be forced out of the market if it were not taken over by another undertaking soon, with the consequence that its share would in any case accrue to K+S, the Commission concluded that a dominant position would be created even in the event of the proposed concentration being prohibited. (21)

In those circumstances, the Commission found itself able to accept the parties' arguments based on application of the failing company theory under which the competent authorities cannot prohibit a concentration if one of the undertakings concerned is on the verge of bankruptcy. In the light of the situation which would arise in eastern Germany as a consequence of the closure of MdK, that conclusion must, according to the Commission, be regarded as consistent with the requirement of strengthening the Community's economic and social cohesion referred to in recital 13 in the preamble to the regulation. (22)

13 So far as the market consisting of all the Member States apart from Germany is concerned, the Commission pointed out that, as a result of the proposed concentration, two undertakings, K+S/MdK and SCPA, would be in a dominant position. (23)

The Commission's appraisal is based, on the one hand, on the consideration that supply outside the two main producers is fragmented and comes from operators who are not in a position to have any impact on the market shares held by the duopoly; (24) and, on the other, on the assumption that there will be no effective competition between K+S and SCPA because of the characteristics of the relevant market, the record of the past behaviour of the two companies and their close commercial links. (25) Those links consist essentially of: (a) cooperation in the export cartel Kali-Export GmbH, a company constituted under Austrian law and based in Vienna, which coordinates its members' sales of potash products in countries outside the Community (the members of that cartel, each with a 25% interest, are: K+S, MdK, SCPA and Coposa); (b) control of a joint venture in Canada, Potascan, of which K+S and EMC/SCPA each have 50% of the shares, and (c) long-established supply links by virtue of which SCPA distributes K+S products in France.

14 In the light of the above, the Commission took the view that the concentration would lead to the creation of a collective dominant position on the Community market apart from Germany, which could significantly affect competition. In order to avoid the negative consequences stemming from that conclusion, the parties agreed to enter into certain commitments which are set out in paragraph 63 of the decision and reproduced verbatim below:

`- Kali-Export GmbH, Vienna

K+S and the joint venture will withdraw without delay from Kali-Export GmbH in Vienna [...].

In the same way K+S and the joint venture will terminate the existing agency contract with Kali-Export GmbH in accordance with the termination arrangements provided for therein. After that date, the joint venture will enter into competition with Kali-Export GmbH via its own distribution organisation.

- Distribution

K+S and the joint venture will establish in the Community their own distribution organisation - where not already in existence - and will distribute their products through this distribution network in accordance with normal commercial practice. A distribution organisation will be established in France for potash products, including potash specialities. This will cover the whole of the French market and its nature and size will be commensurate with the importance of the French market. Its establishment will conform to the principle of economic efficiency.

The current cooperation with SCPA as distribution partner in the French market will be terminated. It will be possible, on the one hand, for SCPA to fulfil contracts already agreed with its own customers and, on the other hand, for the joint venture to build up its own distribution organisation. The sale to SCPA on normal market conditions is allowed.'

15 Noting the Commission's concerns about the negative effects of the concentration on the conditions of competition within the Community, K+S also undertook, as stated in paragraph 65 of the decision, `to adapt until 30 June 1994 the structure of Potacan in such a way as to enable each partner to market the potash obtained from Potacan independently of each other on the Community market'.

In that connection, however, it is made clear in paragraph 67 that `the Commission has decided not to impose a formal obligation based on the commitment regarding Potacan. It has taken note of this commitment and it will proceed on the assumption that K+S will use its best efforts to reach, in agreement with EMC/SCPA, an arrangement which will meet the abovementioned requirements'. Taking account of the fact that EMC/SCPA is not a party to the merger proceedings and that, moreover, K+S and EMC have notified the Potacan joint venture to the Commission under the provisions of Council Regulation No 17/62, the Commission adds that, in the event that `K+S is not able to reach an agreement with EMC, despite K+S's best efforts, an appropriate solution of the competition problems arising from the current form of the Potacan joint venture is to be found in the proceedings under Regulation No 17/62'.

16 Taking the view that the commitments in question would result in the severing of the existing links between K+S and SCPA/EMC, the Commission therefore adopted a decision declaring the concentration compatible, which provides in Article 1 that: `Subject to full compliance with the conditions and obligations contained in the parties' commitments vis-à-vis the Commission, set out in paragraph 63 of this decision, the proposed concentration between Kali und Salz AG, Mitteldeutsche Kali AG and Treuhandanstalt is declared compatible with the common market'.

The two actions: terms of the problem and level of submissions

17 As stated earlier, the decision has been challenged both by the French Government (Case C-68/94 (26)) and by SCPA and EMC (Case C-30/95 (27)). The latter parties seek the partial annulment of Article 1 in so far as it makes the declaration of the concentration's compatibility subject to the conditions and obligations set out in paragraph 63. Those companies also seek the annulment of that part of the decision in which it accepts the commitment (referred to in paragraph 65) entered into by K+S to modify the structure of Potacan by 30 June 1994.

Those companies have put forward six pleas in support of their action: by the first two, they claim that the conditions referred to in paragraph 63 have no connection with the maintenance of effective competition on the Community market; by the third, they complain that the Commission accepted the commitment concerning Potacan in breach of their rights; by the fourth and fifth, they claim that the Commission committed errors of fact and law, both in defining a geographical market comprising the Member States apart from Germany and in its assessment that an alleged collective dominant position existed on that market; finally, they allege that the provisions of the regulation do not apply to a collective dominant position.

18 It is important here to recall that, when the case was still pending before the Court of First Instance, (28) SCPA and EMC lodged an application for suspension of the operation of the contested decision in that it required K+S to sell its shares and/or to withdraw from Kali-Export and to terminate the contractual distribution relations binding it to SCPA, as well as an application for an order suspending the proceedings initiated by the Commission under Regulation No 17 in relation to the Potacan case.

After first making a provisional order, (29) the President of the Court of First Instance, by an order of 15 June, (30) upheld the companies' application in part, suspending the operation of Article 1 of the decision, in so far as it requires K+S and the joint venture to withdraw from Kali-Export.

19 Unlike SCPA and EMC, the French Government seeks the annulment of the whole of the decision (Case C-68/94). It asserts that the Commission failed to respect the principle of close and constant liaison with the authorities of the Member States, as laid down by the regulation; that it made errors in its appraisal of the effects of the concentration on the German market, in particular as regards the application of the failing company theory to this case, and that it also made an incorrect appraisal of the concentration on the Community market apart from Germany.

By this latter plea, which is in turn subdivided into several limbs, the French Government contests: (a) the definition of the relevant geographical market comprising the Member States apart from Germany; (b) the applicability of the merger regulation to collective dominant positions; (c) the misapplication in this case of the concept of collective dominant position; (d) the imposition of conditions and obligations on persons who are not parties to the concentration.

20 Clearly, with regard to the alleged errors in the appraisal of the concentration on the Community market apart from Germany, the arguments of SCPA and EMC are almost completely coterminous with those put forward by the French Government. As I have already stated, that makes it possible to consider the substance of the two actions together, drawing attention, where necessary, to the different arguments submitted.

Before examining the pleas put forward by the parties in detail, however, I consider it appropriate to make a few brief observations of a general nature with the intention of bringing to light, beyond the considerable interests at stake, the implications of these proceedings in principle.

21 I shall begin by noting that, in principle, given the nature of competition, the rules thereon leave a large degree of autonomy to the administrative authority, which is responsible for assessing the specific aspects of any particular case in order to reach the most appropriate decisions. Similarly, it is common ground that the task of the Community judicature in this field requires some latitude in assessing the economic value of the rules. (31)

Those remarks are all the more valid with regard to concentrations, the control of which, being necessarily preventive in character, requires an inherently discretionary appraisal on the part of the authority whose task it is to interpret and apply the regulation. Indeed, in all antitrust systems, even those endowed with forms of control based on structural data (market share held by the undertakings concerned and presumptions raised by it), there is a tendency to formulate judgments based also on qualitative factors and which take account of the specific features of each individual case. (32) However, the application of `qualitative' criteria must not become an unpredictable practice which would undeniably have the direct consequence of considerably reducing the scope for a review of legality. (33) Such review, moreover, appears all the more necessary if it is borne in mind that the twofold nature of the Commission's functions - a power of inquiry and investigation, coupled with a decision-making power - should impose on it an even stricter obligation to respect the right to a fair hearing. (34)

22 I make those observations precisely because the specific issues arising in these proceedings raise for the Court's consideration a general question concerning the limits of the Commission's discretion (35) and the consequential powers of the judicial authorities to act.

The Court should therefore assess extremely carefully what might be described as the `systematic' consequences which its decision may have, first and foremost as regards legal certainty, equal treatment and guarantees for the rights of third parties, but then also as regards its implications for future practice in the control of concentrations. I consider it appropriate to point out that judicial review is essential when laying down rules on concentrations, even in systems where, unlike ours, there is already a predetermined assessment criterion to guide the administrative authorities in their task, rendering their practice clear and predictable. (36)

23 In my Opinion I shall examine the objections of inadmissibility raised by the Commission with regard to the actions brought by the French companies (Case C-30/95) and relating to: (A) the possibility of partially annulling the decision, as sought by those companies; (B) their right to bring proceedings; (C) the possibility of contesting that part of the decision containing the commitment entered into by K+S to adapt the structure of Potacan (paragraph 65); and finally (D) the applicants' legal interest in bringing proceedings.

I shall then examine the substance of both cases. I shall deal first with the two pleas put forward by the French Government alone, namely, failure to observe the principle of close and constant liaison with the authorities of the Member States, which is required of the Commission pursuant to Article 19 of the regulation (A), and the justification for the Commission's appraisal of the effects of the concentration on the German market (B). Next, I shall examine the pleas put forward in both actions with regard to the appraisal of the effects of the concentration on the Community market apart from Germany (C). In this context, I shall deal first with the aspects inherent in the definition of the relevant geographical market (1), then with the possibility of using the provisions of the regulation in order to take action in relation to collective dominant positions (2), and with their applicability to this case (3), also in the light of the criteria developed by the Commission in its previous decisions. Finally, I shall consider the question whether it is lawful to impose conditions on persons who are not parties to the concentration (D).

I - Admissibility (Case C-30/95)

24 It is necessary first to establish the basis of the objections of inadmissibility raised by the Commission with regard to the French companies' action.

A - Partial annulment

25 The first objection concerns the application for partial annulment of the decision. The Commission maintains that, by removing the prerequisites for declaring the concentration compatible with the common market, the annulment of one or all of the conditions would inevitably oblige it to revoke the whole of the contested decision.

The applicants, on the contrary, assert that the conditions at issue could be separated from the rest of the decision and that their annulment would have no effect other than to render the decision unconditional. Moreover, since Article 8(5) of the regulation, which authorises the Commission to revoke its decision in the event of failure to comply with a commitment entered into by the parties, (37) is not applicable in this case, the situation feared by the Commission could in no way arise.

26 I am of the opinion that that objection must be examined together with the substance of the action. Only in that way will it be possible to establish whether annulment of the conditions could indeed affect the rest of the decision, rendering its complete annulment necessary.

In theory, moreover, it would certainly be possible to `split' the decision, as the applicants argue, also in the light of Article 10(5) of the regulation, which expressly contemplates the possibility of partial annulment. To that end, it should be established whether the conditions to which Community case-law makes exercise of the power of partial annulment subject are met. As the Court stated in its judgment in the `Flat Glass' case, (38) it is necessary `to determine whether the scope of the operative part of the decision, read in the light of the grounds of the decision, can be limited ratione materiae, ratione personae or ratione temporis in such a way that its effects are restricted but its substance remains unaltered'. In the course of that examination, it should be ascertained whether, as the Court has stated, `a partial annulment is therefore possible and is justified by the fact that, taken as whole, the decision is favourable to the interests of the undertakings concerned'. (39)

27 As already noted, the imposition of conditions to which compatibility of the concentration is made subject depends on whether, in the Commission's view, the concentration would lead, on the Community market outside Germany, to the creation of a duopoly that could impede competition, which only compliance with such conditions would prevent. In order to annul the decision in part, therefore, it would be necessary to show that that view was unfounded, that is to say, since the concentration would not have the effect which the Commission supposed, that there was no need, in order to declare the concentration compatible, to impose such conditions.

That confirms that the objection can only be examined together with the substance. It should, however, be emphasised at this point that a possible partial annulment could, at most, involve the application of Article 10(5) of the regulation, (40) that is, reopening of the time-limits and a new examination by the Commission.

B - Right to bring proceedings

28 The Commission further maintains that the contested decision is not of direct and individual concern to the applicants. Similar arguments have been put forward by K+S and MdK, who consider themselves to be the sole addressees of the conditions imposed by the decision.

This objection seems to me to be manifestly unfounded. Since the French companies are contesting a decision which is not formally addressed to them, it should be ascertained whether the decision is of direct and individual concern to them, as required by Article 173 of the Treaty. I would point out that the contested decision is undoubtedly of direct concern to the applicants, since the harm allegedly suffered by the French companies is attributable to that decision.

29 With regard to the requirement that the decision must be of individual concern, it should be pointed out that, as the Court has consistently held, `persons other than those to whom a decision is addressed may claim to be individually concerned within the meaning of the second paragraph of Article 173 only if that decision affects them by reason of certain attributes which are peculiar to them or by reason of circumstances in which they are differentiated from all other persons and by virtue of these factors distinguishes them individually just as in the case of the person addressed'. (41) In this case, it is obvious that the applicants' circumstances distinguish them clearly from all other persons.

First, it is significant that the applicants were heard during the pre-litigation procedure. It is clear from the consistent case-law of both the Court of Justice and the Court of First Instance on the right of third parties to bring proceedings in the matter of competition, anti-dumping and aid that persons on whom a regulation or the Treaty itself confers the right to take part in administrative procedures are entitled to institute proceedings `in order to protect their legitimate interests'. (42)

30 Secondly, in assessing the competitive situation on one of the two markets identified as relevant and the effects of the concentration, the Commission mainly and specifically took into account the position of EMC/SCPA as part of a duopoly with K+S. (43)

31 Thirdly, all the conditions imposed on the undertakings involved in the concentration in order for it to be declared compatible affect directly and almost exclusively the interests of the applicants and could seriously jeopardise their position on the market. (44) It is obvious that, in making the adoption of a decision concerning one person conditional upon the modification or even discontinuance of contractual relations or a partnership with a third person, the activities of both persons and not only those of the former are significantly affected.

32 Here also, the Commission's arguments cannot prevail. SCPA cannot properly be denied the right to bring proceedings on the ground that the decision was also of concern to the other company in the Kali-Export export cartel, the Spanish company Coposa.

First, SCPA is also directly and specifically concerned by the other conditions laid down in the decision, which, however, do not affect anyone else; and that is sufficient to differentiate its position from that of any other person, by the same token identifying it as that of an addressee of the measure. Secondly, it would not be possible to prevent a person who was indisputably (and individually) affected by a decision from bringing proceedings in order to protect his interests, merely because another person was in a similar position.

33 To accept the argument criticised here would be to risk undermining the right to judicial protection. Moreover, that argument does not seem to me to be supported by a correct interpretation of the case-law on the fourth paragraph of Article 173 of the Treaty. There is in fact no basis, except that of an excessively restrictive reading, (45) for assuming that the condition of being `directly and individually concerned' must be peculiar to only one person and that it may not apply to several persons who can be precisely and specifically identified as being immediately concerned by the measure. Indeed, in order for the right to bring an action to arise, that condition, namely, that the position of the person concerned must be specifically and distinctively differentiated from that of any other person, must be met, and in this case that differentiation cannot seriously be called into question.

34 I note, moreover, that there is again a causal link in this case, inasmuch as the position of the applicants was specifically taken into account, to the extent of having had a decisive influence on the institution's decision and having constituted the fundamental reason for the contested part of the measure. (46) In other words, the facts of the case show that the position of the applicant companies is differentiated, and not by coincidence.

35 Finally, the Commission's reference to the judgment in the Zunis Holding case (47) does not appear to be relevant. According to the defendant institution, from the point of view of the right to bring proceedings, the mere fact of being a shareholder in a company involved in a concentration is not sufficient to differentiate a person's circumstances in the same way as those of the addressee of a decision are differentiated.

I note first that the principles set forth in the judgment in the Zunis Holding case were applicable to the applicant undertakings which were shareholders in the companies directly involved in the concentration, and therefore in a completely different position from that of EMC and SCPA in this case. Moreover, in the aforementioned judgment, the Court did not by any means state that ownership of a parcel of shares is insufficient to differentiate the position of a person within the meaning of Article 173, as the Commission maintains, but merely pointed out that, for that purpose, the possible effect of a measure on the relations existing between a company's shareholders is not sufficient, but that, on the contrary, there must be specific circumstances present enabling the position of one of them to be regarded as `directly and individually concerned'. And among those circumstances, even on the basis of a reading between the lines of the judgment cited above, it would not be unreasonable to include the specific size of the holding. (48)

36 Those observations are applicable both to SCPA and to EMC. Moreover, the latter is also identified by the Commission as the addressee of the condition relating to Kali-Export. Indeed, despite the fact that only SCPA has an interest in the company in question, the decision concerns EMC and SCPA jointly, and they are therefore treated, rightly or wrongly, as a single business entity. (49) Consequently, the Commission's attempt to separate the position of EMC from that of SCPA in order to debar the former from bringing proceedings appears arbitrary at the very least, even apart from the fact that EMC has complained that it suffered damage `independently' of SCPA, particularly as a consequence of the dissolution of the export cartel. That dissolution deprives it of its sales network for non-Community countries, thus preventing it from achieving `major' exports.

It is hardly necessary to add that all the foregoing considerations dealing with the objection of inadmissibility relating to the conditions under which a person is entitled to bring proceedings against a measure also serve to refute the arguments of the intervening companies, which are based primarily on the applicants' lack of entitlement to bring proceedings.

C - Possibility of contesting the paragraph relating to Potacan

37 In its final objection, the Commission takes the view that the part of the decision concerning the structure of Potacan is not the proper subject of an action for annulment since it is not capable of producing binding legal effects of such a kind as to affect the applicants' interests.

I would point out first of all that, under the Court's case-law, in order to establish whether proceedings may be instituted against a measure under Article 173 of the Treaty, it is necessary to take account of its substance and to determine whether it is intended to produce binding legal effects of such a kind as to affect the applicant's interests by clearly altering its legal position. (50)

38 However, it is certainly unusual that the Commission, having initiated a procedure specifically for the purpose of ascertaining the compatibility of Potacan with the common market on the basis of Article 85 of the Treaty, should already contemplate, or rather suggest, a solution in the present context, encouraging one of the undertakings involved in the concentration to `use its best efforts to reach, in agreement with EMC/SCPA, an arrangement which will meet the abovementioned requirements', namely, the requirements specified by the Commission itself with regard to the adaptation of Potacan's structure. That can only raise serious doubts about the correctness of such an approach which obviously leads to a confusion of procedures that, although different in terms of their subject-matter and the parties, ultimately influence each other.

That said, it must nevertheless be acknowledged that the part of the measure at issue here does not seem to meet the requirements laid down by Article 173 for a measure to form the subject of an action against it; more specifically, it is not of such a kind as to produce binding legal effects vis-à-vis the applicant companies.

39 In the part of the decision at issue here, the Commission has merely `taken note' of a commitment entered into by K+S, albeit one expressly suggested by the Commission, to resolve the Potacan question in agreement with EMC. Moreover, no obligation is placed on the latter to initiate negotiations, still less to agree to the proposals put forward by K+S. As noted above, the part of the decision in question could, at most, produce an effect of some kind, albeit indirect, on the proceedings initiated pursuant to Regulation No 17; (51) however, even if they actually arose, those consequences would not have any relevance for the purpose of assessing whether or not the measure which is the subject-matter of these proceedings may be the subject of an action.

Nor, it seems to me, is it relevant in this case to cite the solution adopted in the `Woodpulp' case, (52) in which the Court held that, for the purpose of assessing whether it could be the subject of an action, an undertaking given by the applicants must be treated in the same way as directions requiring an infringement to be brought to an end, as provided for by Article 3 of Regulation No 17. Even disregarding the differences between the two cases, I note that the assumption on which that solution was based, namely, that the applicants merely assented to a decision which the Commission could have adopted unilaterally, does not apply in this case.

40 Consequently, I suggest that the objection raised by the Commission and the intervening companies should be upheld and that, as regards the point which I have just examined, the action brought by the French companies should be declared inadmissible.

D - No legal interest in bringing proceedings

41 Finally, the intervening companies have objected that the French companies' action is also inadmissible on the ground that there is no legal interest in bringing proceedings since they have already complied fully with the commitments to which the conditions in the decision relate.

I note first of all that, from a procedural point of view, the objection is inadmissible since there is no reference to it in the arguments raised by the Commission. That objection has, in any event, been overcome by the principle laid down by the Court, under which, in accordance with Article 176, even if it proves impossible to comply with the obligation imposed by that provision, `the application for annulment still constitutes an interest as the basis for possible proceedings for damages'. (53)

II - Substance

42 With regard to the substance of the case, I shall first examine the pleas put forward only by the French Government, which therefore relate to Case C-68/94. More specifically, I shall begin by assessing the merits of the argument alleging breach of the obligation to act in liaison with the national authorities (A), then turn to the complaints concerning the appraisal of the concentration on the German market (B). Finally, I shall consider the aspects connected with the alleged errors in the appraisal of the concentration on the Community market apart from Germany (C), including the question of the lawfulness of imposing conditions and obligations which affect third parties not involved in the concentration (D), aspects raised in both actions and therefore to be dealt with together as regards the arguments common to both.

A - Failure to comply with the obligation to act in liaison with the national authorities (Case C-68/94)

43 By this plea, the French Government complains that the Commission failed, in several respects, to observe the principle, enshrined in Article 19 of the regulation, of close and constant liaison with the national authorities of the Member States. In particular, it asserts that the data relating to the breakdown of the sales of the various operators in each Member State and, in the case of SCPA, to the distinction between the sales of its own products and those of products of other undertakings, data which it regards as essential for the purpose of assessing in full knowledge of the facts both the objections raised by the Commission against the notifying parties and the draft decision, were transmitted to the Member States only at the meeting of the Advisory Committee on 3 December 1993. Indeed, the French Government's repeated requests to that effect received only a partial response, by telephone, on 5 November 1993. Finally, it asserts that the inaccuracy of the data relating to SCPA's sales in Belgium and Luxembourg, as transmitted by the Commission to the Advisory Committee, prevented the representatives of the Member States from properly appraising the concentration. (54)

The Commission points out, first of all, that the principle of close and constant liaison with the Member States, laid down in Article 19(1) and (2) of the regulation, does not constitute a general and absolute principle, but must be reconciled both with compliance with the particularly short time-limits prescribed for initiating proceedings (Article 10 of the regulation) and with respect for the confidentiality of business information, relied on in this case specifically in connection with the sales effected in individual countries. Second, it is for the Commission, in application of the regulation, to decide which documents are the most important for transmission to the authorities of the Member States. At all events, in this case, both the authorities of the Member States and the Advisory Committee were, according to the Commission, in possession of all the information necessary in order to make a balanced appraisal of the impact on competition of the proposed concentration. Finally, the incorrect figure to which the French Government has drawn attention and which is manifestly no more than a serious typing error could not in any event be regarded as capable of invalidating the Advisory Committee's analysis.

44 It is appropriate to point out in the first place that Article 19 of the regulation requires the Commission to transmit `to the competent authorities of the Member States copies of notifications within three working days and, as soon as possible, copies of the most important documents lodged with or issued by the Commission pursuant to this regulation' (Article 19(1)). In addition, it requires the Commission to carry out the procedures set out `in close and constant liaison with the competent authorities of the Member States, which may express their views upon those procedures' (Article 19(2)). That article also provides, as far as is relevant here, for representatives of the national authorities to participate in a special advisory committee set up to deliver an opinion on the basis of `a summary of the case, together with an indication of the most important documents and a preliminary draft of the decision' (Article 19(5)).

However, it is common ground in this case that the Commission transmitted to the parties concerned, in sufficient time, as required by Article 19(1), both the objections addressed to the notifying parties (14 October 1993) and the draft decision relating to the concentration (26 November 1993), which, among other things, contained the data relating to the shares held by the operators of the sector on the Community market; in addition, on 5 November 1993, certain data relating to the sales volumes of individual operators in each Member State were transmitted to the French Government at its request, albeit by telephone. It is also common ground that the data requested by the French Government were transmitted to all the Member States only during the Advisory Committee meeting on 3 December 1993 and that one of the details, that relating to SCPA's sales in Belgium and Luxembourg, was incorrect.

45 Accordingly, it must be established whether, as the applicant government maintains, the belated transmission of the data relating to the volume of potash sales, combined with the inaccuracy of one of the details, constitutes a breach of the Commission's obligation to act in liaison with the national authorities.

Let me begin by pointing out that the belated transmission of those data cannot be justified in this case, either in the interests of confidentiality said to have been relied on by the undertakings involved in the concentration, since in that case those data should not have been transmitted, even belatedly, (55) or by a general reference to the discretion which the Commission enjoys in deciding which are the most important documents within the meaning of Article 19(1) of the regulation. It is true that transmission within the time-limit is not an end in itself, but is necessary in order to ensure the proper conduct of the procedure. From this point of view, the important thing is to determine whether the data in question were essential for the purpose of appraising the anticompetitive effects of the concentration and whether their belated transmission was such as to distort the procedure.

46 I admit to harbouring some doubts as to the fact that the data relating to the volume and breakdown of sales of the various operators in the sector, while admittedly important in order to obtain a complete picture of the effects of the concentration, are to be included among the most important documents within the meaning of Article 19(1) of the regulation. I am of the opinion that there is some basis for arguing, as the Commission does, that the data relating to the market shares held by each operator, which were contained in the draft decision, were sufficient in order to appraise the proposed concentration in full knowledge of the facts.

In any event, in view of the particularly short time-limit within which the Commission's decision must be adopted, as well as of the fact that those data were transmitted to the French Government by telephone as early as 5 November, I am of the opinion that the Commission cannot be accused of having failed, with regard to the French Government, to comply with its obligations under Article 19(1) and (2) of the regulation. Furthermore, the belated transmission of the information in question cannot in itself be considered such as to cast doubt on the validity of the decision since it is, on the contrary, necessary to that end for the alleged irregularity to involve an infringement of essential procedural requirements. I would point out in this connection that, as the Court has consistently held, `for ... an infringement of the right to be heard to result in an annulment it must ... be established that, had it not been for that irregularity, the outcome of the procedure might have been different'. (56) However, in this case, the applicant government has not by any means shown that the Commission's decision would have been different had it not been for the alleged breach of the obligation to act in constant liaison with the national authorities; it merely states that such a possibility could not be ruled out.

47 There remains to consider the fact, as the French Government also points out, that the Advisory Committee only received all the relevant data, including the incorrect figure, during the meeting on 5 December, that is, on the day when it was required to deliver its opinion. However, Article 19(5) of the regulation requires the draft decision and all the most important documents to be sent to the Advisory Committee at the time when it is convened, that is, at least 14 days before the meeting at which the opinion is delivered. (57)

The complaint in question, which, as the Commission has pointed out, really concerns the procedures for convening and consulting the Advisory Committee, does not seem capable of having substantial effects on the contested decision. First, the fact that the Advisory Committee, which was composed of experts from the sector, did not raise any objection in this regard should not be underestimated. Second, the applicant government has by no means shown that transmission of the data in question within the time-limit and absence of error in respect of a particular detail would have led the Advisory Committee to adopt a different opinion, still less that the Commission's final decision would have been different. (58) On the contrary, the arguments put forward by the Commission in this regard, and which have remained undisputed, would seem to rule out that possibility.

B - Effects of the concentration on the German market (Case C-68/94)

48 The French Government puts forward two pleas relating to the appraisal of the concentration on the German market: on the one hand, it accuses the Commission of, in this case, misapplying the so-called failing company theory, according to which the acquisition of an undertaking which is bankrupt (or verging on bankruptcy) must, subject to certain conditions, be authorised since it would not give rise to competition problems; on the other hand, it complains that the Commission did not make authorisation of the concentration subject to any conditions relating to the market in question, even though this led to the creation of an outright monopoly (98%) for the joint venture on that market.

49 With regard to the first complaint, I would first point out that American antitrust legislation allows the possibility, although within strict limits in order to prevent abuses, of authorising the acquisition of an undertaking which is insolvent, provided that the undertakings which are parties to the concentration can show, in the same way as in a test of proportionality, that: (a) the undertaking acquired will fail to meet its financial obligations in the near future; (b) it is unable to reorganise successfully under Chapter 11 of the Bankruptcy Act; (c) there are no alternative solutions to concentration which are less anticompetitive; and (d) the `failing' undertaking would withdraw from the market if the concentration were not implemented. (59)

As the Commission expressly points out in its decision, it did not apply the theory in question unreservedly, but made its own appraisal. In particular, while basing itself on that theory, it relied on the following three criteria: (1) MdK would have withdrawn from the market had it not been taken over by another company; (2) K+S would in any case have absorbed MdK's share after the latter's withdrawal from the market; (3) no alternatives were available that were less anticompetitive.

50 In substance, the French Government disputes the Commission's decision to apply a theory developed in and for another system, modifying some of its criteria. In particular, it asserts that the Commission took no account whatsoever of criteria (a) and (b) of the American theory; that it arbitrarily introduced the criterion of the absorption of market shares; and finally, that it has not sufficiently shown that MdK would in any case have been forced to withdraw from the market or that no other alternatives were available.

While acknowledging, moreover, that it is appropriate, in the appraisal of a concentration, to take into consideration the objective of economic and social cohesion referred to in Articles J.2 and J.3 of the Treaty, and also recalled in recital 13 in the preamble to the regulation, (60) the French Government maintains that that should not in any event result in the negation of the primary purpose of the control of concentrations, which is to protect competition.

51 However, the criteria, as laid down by the regulation and in the light of which the compatibility of concentrations with the common market must be appraised, are undoubtedly characterised by a rigorously economic approach which is therefore based mainly on considerations relating to the protection of competition. (61) Nevertheless, as the French Government has acknowledged, recital 13 also undoubtedly reveals the significant attention paid to the requirements of economic and social cohesion, so justifying the account taken of objectives regarded as equally fundamental by the Treaty, provided of course that the protection of competition is not jeopardised.

It is clear, moreover, that the fact the Commission was guided by a theory developed in another system, although partly modifying the criteria for its application, cannot be regarded as a ground for annulment of the decision. What is important here is that the theory in question should not be contrary to the regulation, that the analysis relating to it in the decision should not be flawed and that, in short, the criteria adopted should have been properly applied.

52 As is clear from the decision, the Commission took the view that a line of argument based on the failing company theory can in fact be taken into account `pursuant to Article 2(2) of the merger regulation, as far as the causality of a concentration for the creation or strengthening of a dominant position is concerned'. (62)

On that view, the contested introduction of the criterion by which it must be ascertained whether the acquiring undertaking would in any case obtain the market share of the undertaking acquired is specifically intended to show that the strengthening of the dominant position would be inevitable in that it would take place even if the concentration were prohibited.

53 I note first of all that the impermeability of the German market, which has been established, and the fact, also pointed out in the decision, that K+S was in a position to increase its production without any increase in costs and to become the sole supplier on the German market, lead inevitably to the conclusion that MdK's market share would, in the event of its withdrawal from the market, have accrued to K+S; that, moreover, is not seriously contested even by the French Government.

What the French Government contests is, rather, the application of that criterion, on the ground that an undertaking which was already in a dominant position would thus be enabled to strengthen its position, solely by reason of the probability that it would in any case, in the near future, ultimately acquire the market share of the other undertaking operating on the market, which would be taking things too far and would be contrary to the objective of protecting competition. That assertion, while irrefutable in itself, nevertheless overlooks the fact that the criterion in question assumes the fulfilment of two other criteria: that the undertaking in question is in any event bound to withdraw from the market and that there are no alternative, less anticompetitive methods of acquisition.

54 With regard to the latter two criteria, however, the French Government maintains, as I have stated, that they were not correctly applied in this case. In particular, as far as the probable withdrawal of MdK from the market is concerned, it asserts that the Commission has by no means shown that it would be impossible to keep MdK alive as a public undertaking by, for example, granting aid compatible with Articles 92 and 93 of the Treaty.

That argument cannot be accepted. It is true that the Treuhand, the public institution whose task is to privatise the former GDR's enterprises and the owner of MdK, holds 49% of the share capital in the undertaking created as a result of the concentration and that, as a consequence, the Treuhand (and therefore the State) has given and will be giving that undertaking substantial aid. It is also true, however, that the Commission's analysis is based on the following circumstances: (a) MdK's considerable losses cannot be permanently subsidised by the Treuhand from public funds, in so far as such a solution would in any event be incompatible with the provisions of the Treaty relating to State aid; (b) the costs of any restructuring would in any case be greater than the aid provided for by the concentration; (c) the statement by the Treuhand, in the context of privatisations already embarked upon, that it will close down all MdK's operations if it cannot find a private undertaking prepared to take them over. It was therefore in the light of the particular circumstances of this case (privatisation of a State-owned enterprise which cannot be regarded as viable according to the normal criteria for valuing companies) that the Commission considered that it had sufficient evidence to show that MdK was bound to withdraw from the market in the medium term.

55 Finally, I am of the opinion that the Commission's analysis must also be endorsed from the point of view of the absence of other, less anticompetitive solutions. It is common ground, as the privatisation procedure documents forwarded by the Treuhand to the Commission clearly show, that all the undertakings which had expressed an interest in acquiring MdK had subsequently abandoned their interest. Moreover, the Commission conducted a further investigation and, in particular, informed the undertakings concerned of the exact amount of aid from which the concentration would have benefited and which would therefore also have been granted in the event of purchase by a private undertaking. Despite that, no undertaking declared itself willing to purchase MdK.

Under those circumstances, it seems to me that the conclusion which the Commission reached can be endorsed. However, the fact that the Commission did not consider as a possible alternative the offer submitted by the Peine group, after the conclusion of the agreement on the merger contract, to purchase the Bischofferode mine, that is to say, one of MdK's four operational mines, remains to be examined. This was because Bischofferode apparently does not constitute a significant part of MdK, and also for reasons connected with the employment factor.

56 The latter aspect, which, moreover, eventually assumed considerable importance in the application of the `failing company defence' theory in the United States as well, (63) is in my view decisive, particularly in the light of the requirements of economic and social cohesion to which I have referred on a number of occasions.

The specific nature of the framework within which concentrations must be viewed, which inevitably includes issues of industrial policy, as well as the objectives of strengthening economic and social cohesion referred to in Articles 2 and 130b of the Treaty, may very well justify appropriate account being taken of employment and, in general, social aspects in the appraisal of concentrations. That is true at least in cases such as this, where there are considerable differences, specifically in terms of potential employment, between the various solutions. (64)

57 I would point out, moreover, that, in two recent judgments, even though their context and purpose were different in that they were more specifically concerned with the right to bring proceedings granted to the trade union representatives of the employees of the undertakings involved in the concentration, the Court of First Instance itself has held that, in applying the regulation, the attention given to the protection of competition may certainly be reconciled, in the context of the assessment of a concentration's compatibility, with appropriate consideration of its social consequences from the point of view of Article 2 of the Treaty. To that end, the Commission may, according to the Court, ascertain whether the concentration is liable to affect the position of the employees of the undertakings in question, `such as to affect the level or conditions of employment in the Community or a substantial part of it'. (65) The Court did not fail to refer, in those judgments, to recital 13 in the preamble to the regulation, cited above.

Finally, it is significant that in this case the protection of employment is set in a wider context characterised by the need to overcome the structural deficiencies of eastern Germany, where MdK is established. Even from that point of view, therefore, the decision is certainly consistent with the objective of economic and social cohesion referred to in recital 13.

58 The French Government's second complaint, which is closely connected with that just considered, relates, as I have stated, to the fact that the Commission did not attach any conditions and/or obligations to authorisation of the concentration as far as the German market is concerned, despite the fact that it gives rise to a monopoly on that market (98% of the market). In particular, the Commission should, in the French Government's opinion, have made the compatibility assessment conditional upon a specific commitment by the undertakings to leave the German market open to competition, on the model of the Nestlé/Perrier decision. (66)

In reply, the Commission states that the Nestlé/Perrier decision is not at all relevant, since in this case the opening up of the market to competition does not depend on the structure of that market, as it did in that case, but on buyer behaviour, and therefore on demand and not on supply. It notes, moreover, that the French Government has not specified what commitments might be imposed on K+S and MdK in order to `open up the market to competition'.

59 However, it is undoubtedly true that, as the French Government claims, it is not its task to propose solutions in this connection, since that is a matter for the Commission as the institution responsible for applying the antitrust rules. Nevertheless, the absence of suggestions on its part merely indicates the difficulty of finding conditions to attach to the concentration on the German market, which could open it up to competition.

It is sufficient to observe in this connection that, even assuming that, in order to open up the German market to competition, despite the special features noted by the Commission, it was still necessary to influence its structure, the fact remains that a solution such as that adopted in the Nestlé/Perrier case was not practicable. Indeed, as I have stated, no alternative methods of acquisition were available in this case. Accordingly, I do not consider that the Commission's appraisals can be refuted by the applicant government's objections.

60 In conclusion, even taking account of the foregoing considerations concerning the particular socio-economic features of this case, I suggest that the Court should dismiss the plea in question and that the Commission's appraisal of the effects of the concentration on the German market should be approved as a whole.

C - Effects of the concentration on the Community market apart from Germany (Cases C-68/94 and C-30/95)

61 Both the French Government (Case C-68/94) and SCPA and EMC (Case C-30/95) challenge the decision's appraisal of the concentration in relation to the second relevant geographical market, namely that comprising all the Member States apart from Germany. Indeed, the applicants' complaints concern the very definition of that market, the possibility of applying the regulation to collective dominant positions, the misuse, in this case, of the concept of collective dominant position and, finally, the imposition of conditions and obligations on persons who are not parties to the concentration.

It is appropriate to make clear that, since the arguments put forward by SCPA and EMC are for the most part included among those relied on by the French Government (which has, moreover, presented them more fully and in greater detail), I shall refer below, in particular, to the latter's arguments. Of course, I shall not omit to draw attention to the specific features of each of the actions in question.

1. Definition of the relevant geographical market

62 The applicant government and undertakings argue that insufficient reasons are given for identifying the Community (apart from Germany) as the second geographical market relevant for the purpose of appraising the concentration and that such identification is based on an incorrect and, in any case, partial analysis of the factors to be taken into consideration. In particular, the French Government observes that the criteria used by the Commission for defining the geographical market at issue here should, if they were consistent with those usually employed when applying the regulation, have led it to deny the existence of a single market consisting of the other Member States of the Community. That market, it maintains, is a very heterogeneous one, for several reasons.

In reply, the Commission asserts that it legitimately and correctly defined the Community as a whole, apart from Germany, as a distinct geographical market, a definition which it adopted by reference to this specific case, and in particular to the nature and characteristics of the product in question. That is why, according to the Commission, it is hardly relevant to refer to the criteria used in other cases.

63 I shall begin by noting that the definition of the relevant geographical market, a necessary and prior condition for any analysis of allegedly anticompetitive conduct, assumes even greater importance where, as in this case, it is sought to demonstrate the existence of a dominant position. (67)

Although the Commission undeniably enjoys a wide discretion on this question, it is clear that it is in any event for the Community judicature to verify the accuracy and validity of the analysis made in the decision and on which the finding of a collective dominant position on the Community market apart from Germany is based. (68)

64 It should be pointed out that the regulation does not give any clear guidance on the definition of the relevant geographical market, and only Article 9(7) deals with the criteria necessary for assessing whether there is a distinct market within a Member State and therefore whether the conditions are fulfilled for the Commission to refer the examination of the notified concentration to the competent authorities of that State. Those criteria are repeated in Section 5 of form CO: (69) `The relevant geographic market comprises the area in which the undertakings concerned are involved in the supply of products or services, in which the conditions of competition are sufficiently homogeneous and which can be distinguished from neighbouring areas because, in particular, conditions of competition are appreciably different in those areas. Factors relevant to the assessment of the relevant geographic market include the nature and characteristics of the products or services concerned, the existence of entry barriers or consumer preferences, appreciable differences of the undertakings' market shares between neighbouring areas or substantial price differences.'

Among the factors to be taken into consideration, according to the requirements of form CO, there is therefore no order of importance which would allow any one of them to be regarded as predominant or at any rate more significant than the others. Consequently, while the Commission's claim that the criteria applied in earlier decisions are to be regarded as irrelevant in this case cannot be accepted, they must be regarded as constituting only one frame of reference, since the validity and consistency of the Commission's decisions must be assessed by reference to the particular context within which each specific case is set.

65 That being so, I now turn to the various arguments put forward. First, in the applicants' opinion, the Commission regarded as homogeneous a market comprising countries with no domestic production (Belgium, the Netherlands, Denmark and Greece), producer countries in which production is structurally higher than consumption (Spain and the United Kingdom) and others in which consumption is structurally higher than production (France). Furthermore, the purchasing strategies of the importing Member States are substantially different, so that almost all intra-Community trade consists of flows in one direction only and not of reciprocal movements which normally characterise a genuinely homogeneous market.

According to the French Government, the Commission therefore lumped together totally different situations, whereas it could have established some degree of homogeneity only with regard to the non-producer countries. In its view, further proof of the lack of homogeneity of the market in question derives from the examination of the market shares held by the suppliers, which vary considerably from one country to another, as well as from the concentration of supply. Even the analysis of the characteristics of demand confirms, according to the applicant government, the non-existence in this case of a geographical market comprising all the Member States apart from Germany. In its view, the Commission's assessment concerning the interchangeability of the products is incorrect since the choice of potash products depends on the geological nature of the soil, the size of the agricultural area, consumer habits and agricultural policies, as well as on the presence of potash industries within the country. (70)

66 The Commission counters this by stating that the inclusion of producer and non-producer countries in the same market in no way implies an incorrect definition of the relevant geographical market, just as the presence of only one producer in a Member State is by no means decisive. Moreover, the existence of trade flows in only one direction does not preclude homogeneity of the market concerned, with which, according to economic theory, it is perfectly compatible. According to the Commission, it is not possible to come to a different conclusion on the basis of the different purchasing strategies employed by the importing Member States, still less on the basis of the different market shares held in the individual countries by the main operators, which is not a decisive factor in the definition of the relevant market since it may be due, for example, to pre-existing trade barriers or firmly established consumer habits.

With regard to the characteristics of demand for the products in question, the Commission considers that it took them sufficiently into account and found, on the one hand, a high degree of interchangeability, since in no country except Germany do users have a clear preference for local products and, on the other, that all Community potash producers had the capacity to produce the different types of potash. Consequently, unlike other cases in which it had found consumer preferences for well-known brands (71) or for domestic products (72) or extremely weak demand in a newly developing market, (73) in this case there is no evidence that the structure of demand points to the existence of distinct national markets.

67 Second, according to the applicants, the analysis of transport costs and prices within the reference market, as carried out by the Commission, is not at all reliable. As far as transport costs are concerned, the Commission did not actually conduct an analysis but merely stated apodictically that transport costs are not an obstacle to trade flows.

The survey of potash prices within the Community, is equally inadequate: the Commission failed to undertake, as it should have done, a comparative analysis of the prices charged by the various operators in each Member State. In order to demonstrate the absence of substantial differences in price levels between the various countries concerned, it supplied statistics which were five years old and therefore unreliable.

68 The Commission rejects the criticisms levelled at its analysis, objecting that the absence of trade flows between producer countries and importing countries is not always and not necessarily caused by transport costs. On the other hand, the existence of trade between some non-adjacent countries (Germany and Portugal, Spain and Belgium, Germany and Ireland) and, by contrast, its absence between neighbouring States (the United Kingdom and the Netherlands) show, in its view, that transport costs are by no means prohibitive. The Commission draws attention to the considerations set out in the contested decision (paragraph 42) which, it maintains, show that the level of those costs is not capable of preventing trade flows within the geographical market in question.

With regard to the prices charged in the Member States apart from Germany, the Commission points out that they are not markedly different. In fact, the maximum price difference within the Community is 10%, whereas in Germany it is as high as 20%.

69 Finally, according to the applicant government and undertakings, the Commission should at least have isolated, in its appraisal of the proposed concentration, the French and Spanish markets in which, because of their particular characteristics, the conditions of competition are unlike those of the other Community partners. In particular, according to the French Government, the Spanish market displays characteristics similar to those of the German market, whereas French market conditions are very unusual, determined by a production level below that of consumption and by the presence of a single large operator.

The very fact of not separating the markets of the two States in question from that of the remainder of the Community eventually results, in the applicants' opinion, in a wholly distorted analysis of the conditions of competition before and after the concentration, by giving a false picture of the balance of power between K+S and SCPA.

70 The Commission points out that the presence of significant trade flows, albeit in only one direction, the absence of marked differences between prices and the interchangeability of the various potash products could only lead to the inclusion of Spain and France in the Community market apart from Germany, all the more so as the distribution methods used in those two countries are identical to those employed in the rest of the Community (apart from Germany).

As far as France is concerned, the Commission also disputes that the conditions of competition on that country's market are different from the rest of the Community and points out that no importance can be attached, for this purpose, to the fact that K+S and SCPA/EMC have artificially kept the French market separate from the Community market. That does not in any way alter the fundamental homogeneity of the two markets, which stems from objective factors.

71 That being so, I shall begin by noting that the unilateral nature of the trade flows, which is also acknowledged by the Commission, is unquestionably an important factor in defining the relevant geographical market since it could be inferred from it that there are barriers to the entry of individual countries, which could prevent the marketing of products and thereby demonstrate the existence of national markets. (74) This question should therefore have been examined in more detail by the Commission, which instead merely noted the presence of significant trade flows between the Member States in question.

Likewise, the Commission's disregard of the fact that the largest importers in the sector had and have substantial market shares in their own countries and significantly different market shares in the other countries seems less than wholly convincing, as does the minor importance it attaches to the degree of concentration of supply.

72 The deficiencies to which I have just drawn attention are not, however, such as to invalidate the Commission's appraisal, which seems to be founded on objective factors that are essentially undisputed.

First, the conditions do not seem to have been met in this case for considering that the structure of demand indicates the existence of distinct national markets, particularly since there is, throughout the Community apart from Germany, a high degree of interchangeability between potash products and there are no particular consumer preferences for speciality products which are only available from local producers.

73 With regard to the analysis of transport costs, a factor commonly considered to be among the most significant in defining the concept of relevant market, as the Court stated in the `sugar' case, (75) it is certainly true that the Commission has often confined itself to general assertions. (76) However, it does not seem to me that, in its objections, the applicant government has succeeded in showing that the differences in costs are such as to constitute `barriers to entry': it is well known, as the Court has held, that different transport costs are not relevant for this purpose if they are `not discriminatory'. (77)

A similar observation may apply to the analysis of prices of the products in question, which are another important factor to be taken into account, according to the opinion expressed by the Commission in previous cases. (78) While it is true that the only information supplied by the institution as evidence of the absence of significant differences in price levels between States consists of a few statistical data provided by the FAO for the period from 1987 to 1989 and of the reference to the 1992 prices charged by K+S in Belgium, the Netherlands and Germany, it is also true that neither the French Government nor the applicant undertakings have countered the Commission's analysis with any evidence to the contrary. Indeed, they have actually affirmed that there is no substantial difference between prices, while attributing this to the transparency of the market and not to its homogeneity.

74 If one adds to the foregoing considerations the fact that the applicants have not disputed the substantial uniformity of the distribution methods adopted in the different States, which is such as to preclude the existence of barriers to entry similar to those existing in Germany, the Commission's analysis can only be regarded as essentially correct and sufficiently reasoned.

Nor would it be possible to come to a different conclusion on the basis of the arguments put forward by the French Government in order to show that France and Spain are independent markets which cannot be lumped together with the other Community countries, although the Commission's position is not entirely free from contradictions.

75 It must be acknowledged in this connection that, in the analysis of so-called `barriers to entry', (79) the existence of SCPA's statutory monopoly (even though limited to the transit of imports from countries outside the Community), which is regarded by the Commission as a major barrier to entry to French territory, (80) is not taken into consideration in any way thereafter in order to establish whether or not France may constitute a distinct market. Nor can such inconsistency be explained, as the Commission claims, by the distinction between barriers which prevent intra-Community trade and those which hinder trade with non-Community countries. Indeed it seems doubtful, to say the least, whether SCPA's distribution monopoly has any effect on the entry of products from non-Community countries into the market of the other Member States, since certain data supplied by the French Government, and not disputed by the Commission, show that the contrary is the case. (81)

The contradictions to which I have just drawn attention do not, however, appear sufficient to justify acceptance of the French Government's argument; according to the Commission's assessments, which have not been disputed, both France and Spain display, as regards competition, similar characteristics to the rest of the Community market from the point of view of product interchangeability, opening up of markets (confirmed by the large volumes of imports), price levels and distribution systems. Even in respect of those two countries, therefore, there is sufficient evidence for regarding the Commission's decision to include them in the market comprising all the Member States apart from Germany as justified.

76 In the light of the foregoing considerations, I am of the opinion that the definition of the geographical market given in the decision must essentially be endorsed, above all in its conclusions.

The Commission could, and perhaps should, have justified its assessments more fully; on closer examination they prove not to be without a few approximations. However, it does not seem to me that some degree of homogeneity in the conditions of competition on the market comprising all the Member States apart from Germany can plausibly be denied or, more particularly, that the French Government's objections can lend weight to the opposite conclusion. It should be borne in mind, moreover, that, as the Court has held, `homogeneity' of conditions of competition for the purposes of this case does not mean perfect equality of conditions in the different parts of the territory regarded as the relevant market. Rather, what is required is that the conditions of competition should be sufficiently homogeneous to appear non-discriminatory, (82) that is, such as to enable traders to engage in competition on an equal footing.

77 In this case, it does not seem to me that the applicant government's arguments make it possible to regard as not homogeneous, in the sense used here, the conditions of competition within the market defined as relevant by the Commission. Accordingly, I propose the Court should reject the plea put forward by the French Government.

(2) Possibility of using the regulation as the basis for action in relation to collective dominant positions

78 It will be recalled that the conditions to which authorisation of the concentration in question was made subject were specifically intended, on the basis of the analysis made by the Commission, to remove any basis for creating a situation of oligopolistic dominance by K+S and SCPA on the relevant market.

Both the French Government and SCPA and EMC, which in any event dispute the existence of a collective dominant position in the present case, assert that the regulation does not even apply to circumstances of this type.

79 The applicants maintain that it is not possible, even from a theoretical and general perspective, to rely on the concept of collective dominant position in the context of applying the rules on the control of concentrations. There are several reasons for this:

- first, it is precluded by the wording: Article 2 of the regulation, unlike Article 86 of the Treaty (`... one or more undertakings ...'), does not expressly provide for the case of a collective dominant position;

- second, neither the legal bases of the regulation (Articles 235 and 87 of the Treaty) nor the legislature's intention allow application of the provisions on concentrations to situations of oligopolistic dominance;

- finally, further confirmation is to be found in the procedural guarantees accorded to third parties by the regulation; in particular, the undertakings forming part of a collective dominant position, but not of the concentration, would have insufficient means of protection.

80 The Commission counters those arguments with the following:

- The wording of the regulation, and in particular the fact that it does not provide expressly for the case of a dominant position held by several undertakings, in no way precludes its application to cases involving the formation or strengthening of collective dominant positions. It is significant that Article 2(3) links the dominant position to the concentration, and not to the undertakings concerned, by taking into consideration its implications for effective competition.

- The legal bases of the regulation show that its objective is to fill a gap left in the rules on competition by Articles 85 and 86 with regard to the control of oligopolistic dominant positions. Nothing in the preparatory work or in the declarations cited by the French Government supports the conclusion that the Council intended to exclude application of the regulation for the purpose of declaring unlawful or eliminating collective dominant positions.

- A decision authorising a concentration is binding only and exclusively on the notifying undertakings. Since the act closing proceedings initiated pursuant to the regulation cannot cause any harm to third parties, even where they form part of the collective dominant position in question, it is inappropriate to speak of a lack or inadequacy of guarantees for third parties. In any case, the detailed procedural rules for implementing the regulation afford ample protection for the rights and interests of third parties.

- The possibility of prohibiting a concentration which would strengthen the oligopolistic framework of the market arises from the economic theory which holds that competition is at all events affected, provided that certain conditions are fulfilled, within an oligopolistic market. The use of the regulation as a basis for action in relation to a collective dominant position is therefore vital in order to ensure effective competition.

- The wording

81 I shall begin by observing that the applicants' view is difficult to dispute: the regulation, and in particular Article 2 thereof, contains no reference, even indirectly, to the concept of oligopolistic dominant position. That is all the more significant if it is borne in mind, as the applicants point out, that Article 86 of the Treaty provides, on the contrary, and expressly, that any abuse of a dominant position by `one or more undertakings' is incompatible with the common market.

Clearly, Article 2(3), which, as the Commission maintains, does indeed link the dominant position to the concentration and not to the undertakings concerned, refers to the dominant position of the concentration and not to the dominant position involving undertakings which are not parties to the concentration. In those circumstances, it would be too easy to evoke the well-known adage ubi lex voluit, dixit, ubi noluit tacuit.

82 However, while it is true that the wording can only lead to the conclusion that the regulation is aimed solely at discouraging concentrations which create or strengthen a dominant position held by a single undertaking, and does not cover those which result in dominance by several undertakings, (83) it is also true that the wording of Article 2 of the regulation cannot be considered decisive, since that provision must be read in the light of the regulation in its entirety and of its underlying objective.

This brings me to the other arguments relied on by the applicants and relating to the legal bases, to the legislature's intention (as it appears from the preparatory work and is reflected in the wording of the regulation) and to the procedural guarantees, as well as to the economic considerations put forward by the Commission, which are closely linked to the objectives of the rules on competition.

- Legal bases and intention of the legislature

83 Let it be said immediately that I do not share the applicants' view that the choice of legal bases is in itself a ground for concluding that the regulation does not apply to collective dominant positions. In principle, Articles 87 and 235 can indeed be used for that purpose. On the other hand, it seems to me that neither the fact that the regulation is based mainly on Article 235, rather than on Article 87, nor the declaration by the Council and the Commission on Article 22 of the regulation, in so far as it precludes the applicability of Articles 85 and 86 to concentrations, (84) can be relied on.

It is hardly necessary to add that it is difficult to see any other possible basis for preventive action against collective dominant positions created or strengthened as a result of concentrations, unless the view were taken, albeit unreasonably, that the Treaty itself precludes such a possibility.

84 Unlike the Commission, however, I am unable to infer from such a conclusion that the regulation in its present version actually authorises action against collective dominant positions. I shall merely observe that the fact that the legal basis chosen is undoubtedly capable of conferring specific powers does not in itself mean that those powers have in fact been conferred. The fact remains that, being founded on that legal basis, certain powers have been provided for and not others.

In the end, the point is whether the wording of the regulation confirms, either expressly or even implicitly, that the power to take action against collective dominant positions has been conferred on the Commission. However, it seems to me that the text as a whole tends to support a restrictive interpretation of Article 2.

85 That conclusion is supported, for example, as the applicant undertakings have asserted, by the wording of recital 15 in the preamble to the regulation, on the basis of which `concentrations which, by reason of the limited market share of the undertakings concerned, are not liable to impede effective competition may be presumed to be compatible with the common market' and `an indication to this effect exists, in particular, where the market share of the undertakings concerned does not exceed 25%'. In their view, the choice of that percentage would be inappropriate if the regulation also applied to collective dominant positions. (85)

However, it is quite true that the presumption of incompatibility of a dominant position on the basis of a total market share in excess of 25%, as contained in the recital, is not repeated in the operative part of the regulation. (86) It is also true that, in the decisions adopted pursuant to the regulation, market shares are only the starting point for appraising the effects of the concentration since many other factors must be taken into consideration. (87) However, if any significance is to be attached to recital 15, it would not be unreasonable to regard it as evidence of the legislature's intention not to include cases of collective dominance within the scope of the regulation, since the 25% share is certainly rather low for an oligopoly. (88)

86 Furthermore, the preparatory work would seem rather to confirm the legislature's intention not to apply the regulation to collective dominant positions. Although it is well known that the Commission's intention, as indicated during the process of drafting the act and expressed on several occasions, was to apply the rules in question to oligopolies, (89) it is also true, as the Commission has pointed out, that the present wording of Article 2(3) is the result of a difficult compromise reached specifically on the question of collective dominant positions. (90)

It is worth noting at this juncture that a comparison between the final amended proposal submitted by the Commission (91) and the definitive text shows that the reference to obstacles to the maintenance and development of effective competition, as set out in the former, was subsequently replaced by the concept of `dominant position', which undeniably refers to the wording of Article 86. Accordingly, it must be accepted that the final text is the result of a conscious choice determined, as the Commission has acknowledged, by the difficulty of reaching agreement on this point.

87 However, regrettable though it may be that the Council took its decision in accordance with the wishes of some of its members and against those of others, no-one except the legislature itself can do anything about it. I am not swayed by the Commission's argument that, as a result, in the absence of a provision expressly precluding the applicability of the regulation to collective dominant positions, that question has been left open. In any event, I do not believe that the Commission can take the view, as it claims, that it has the power to resolve issues deliberately left unresolved by the Member States.

To put it more clearly, to believe otherwise would amount to supporting a sui generis criterion of subsidiarity, according to which, whenever it deemed it appropriate or necessary, the Commission could deputise for other institutions and, in particular, for the legislature. It would certainly be more in keeping with the rules, from this point of view, if the Commission were to take the initiative in proposing an amendment to the regulation along the aforementioned lines. Failing that, I would be very hesitant to state that the Commission and the Court may, when applying and/or interpreting rules, fill any lacunae deliberately left by the legislature. (92)

88 In conclusion, even the preparatory work would seem to confirm that the legislature's intention, as reflected in the wording of the regulation, is that the latter should not apply to collective dominant positions. That finding is not, however, decisive for the purposes of this case. The fact remains that the preparatory work can, at most, provide guidance but is not conclusive since, in order to establish the correct interpretation of a provision, it is necessary to refer to its wording and purpose.

- Procedural guarantees

89 I now turn to an aspect which is unquestionably of fundamental importance in this context. Both the French Government and the applicant undertakings maintain, as I have already stated, that the fact that the regulation must not and cannot be applied to collective dominant positions is decisively confirmed by the absence from its provisions of procedural guarantees capable of protecting the rights of undertakings which, even though they are not involved in the concentration, are part of a collective dominant position.

I have already drawn attention, in examining the admissibility of the applicant undertakings' action, to the lack of justification for the Commission's view that the act closing proceedings initiated pursuant to the regulation is binding only on the parties to the concentration, with the result that it could under no circumstances cause damage to third parties. Without citing other examples, it is precisely this case which, in an extremely significant way, indicates the contrary. (93)

90 That being so, it is necessary at this juncture to establish whether the procedural guarantees provided for by the regulation in order to protect third parties can be considered appropriate even where the `third parties' are part of the oligopoly to which the Commission's decision relates.

Let me say at once that the participation of third parties in the proceedings, as provided for by the regulation, does not seem capable, in this case, of ensuring appropriate protection for the right to a fair hearing. First, the undertakings which are presumed to be members of the oligopoly but are not parties to the concentration are not given any specific information. Secondly, under Article 18(4), third parties are heard only in so far as the Commission or the competent authorities of the Member States deem it necessary also to hear other natural or legal persons (who must obviously be taken to include the undertakings not parties to the concentration) or where, on their initiative, the latter show a sufficient interest. Consequently, as Regulation No 2367/90 (94) (applicable ratione temporis to this case) confirms, those proceedings are non-mandatory and informal, given that they are conducted without the participation of the representatives of the Member States' authorities and without the formal requirements and guarantees laid down for the hearing of the parties. In short, they are proceedings which are capable of protecting the rights of those who really are `third persons' (in every sense of the term), but not the rights of those who are `parties' to the dominant position.

91 The analysis carried out by the French Government, which compared the procedural arrangements laid down by Regulation No 2367/90 pursuant to Article 18(4) of the regulation with the mechanism envisaged for proceedings initiated under Articles 85 and 86, merely confirms the inadequacy of the guarantees provided for the right to a fair hearing of undertakings which are members of an oligopoly but are not involved in the concentration. (95)

That analysis clearly shows that, unlike undertakings holding a collective dominant position as referred to in Article 86, undertakings participating in an oligopoly allegedly created or strengthened by a concentration in which, however, they are not involved, are treated as `third persons', with the negative consequences which inevitably flow from that as regards the right to a fair hearing.

92 However, it is hardly necessary to point out that respect for the right to a fair hearing in administrative proceedings is an essential principle of the Community legal order. As the Court has held on numerous occasions, `observance of the right to be heard is, in all proceedings initiated against a person which are liable to culminate in a measure adversely affecting that person, a fundamental principle of Community law which must be guaranteed even in the absence of any rules governing the procedure in question'. (96)

That lends some credence to the Commission's view that, since respect for the right to a fair hearing constitutes a fundamental right, any absence of specific procedural guarantees from the regulation is of no importance for the purposes of its applicability to collective dominant positions. In any event, third persons, as well as parties, remain fully entitled to receive the necessary protection.

93 That is certainly an attractive argument, but one with which I am nevertheless unable to agree. While it is true that the right to a fair hearing must in any event be observed, quite apart from individual express provisions to that effect in relation to any circumstances likely to arise, it is nevertheless true that it would be paradoxical to allow the Commission to take refuge behind such a screen when it has made no provision, despite the adoption of specific legislation on the subject, of which Regulation No 2367/90 is clearly an example, for the necessary guarantees in favour of those who may be injured by the application of the rules in question.

That said, I recognise that the Court could overcome the problem by stating that it would suffice to extend the guarantees provided for the parties to the concentration so as to cover the parties to the presumed oligopoly. The fact remains, however, that the absence of provisions to that effect in the regulation can only confirm, as the applicants have maintained, that the regulation is not, or at least was not intended by the legislature, to be applied to collective dominant positions. Finally, it is scarcely necessary to add that, in this case, observance of the right to a fair hearing, to which the applicant undertakings were (according to the Commission) in any event entitled, was certainly not ensured. (97)

- Economic considerations and objectives of the regulation

94 I now turn to the economic arguments put forward by the Commission, which are closely linked to the objective of ensuring effective competition.

Essentially, the Commission's view is that it is vital to bring oligopolistic situations under control (98) since they could result in damage to competition at least as severe as that caused by an individual dominant position. The applicability of the regulation to collective dominant positions is based, in the final analysis, on the need to control oligopolies in order to safeguard competition. (99)

95 I shall begin by noting that the Commission's concern is justified, especially if it is examined in the light of the economic and legal implications of oligopolistic situations. It is hardly necessary to point out that an oligopoly, that is to say, a market supplied by only a handful of undertakings, (100) can produce harmful effects very similar to those generally attributed to a monopoly. (101) Added to that is the undeniable fact that, on the one hand, there is a certain correlation between concentration and market power, understood specifically as the power to raise prices by reducing production and increasing profit margins, and, on the other, that concentrations can foster collusion between competitors and the adoption of parallel, even if not concerted, conduct.

The Commission has also pointed out that the problem could not be resolved by relying on Articles 85 and 86 of the Treaty either, since those provisions do not always or in all circumstances allow effective action to be taken against oligopolies. Indeed, at times an oligopoly cannot be contained by recourse to the prohibition of agreements, decisions and concerted practices since, as the Court made clear in its judgment in the `Woodpulp' case, parallel conduct, which sums up the very concept of an oligopoly, `cannot be regarded as furnishing proof of concertation unless concertation constitutes the only plausible explanation for such conduct. It is necessary to bear in mind that, although Article 85 prohibits any form of collusion which distorts competition, it does not deprive economic operators of the right to adapt themselves intelligently to the existing and anticipated conduct of their competitors'. (102) Moreover, reliance on Article 86 is limited by the nature of the provision itself which can affect only pre-existing situations of dominance.

96 However, the lacunae to which I have just drawn attention do not seem to me sufficient to justify the application of the regulation to oligopolistic dominance; nor does the `Woodpulp' judgment appear to me to contain any indications of the kind sought by the Commission: the fact that parallel conduct exists which cannot be punished on the basis of Article 85 (103) does not in itself mean that such conduct must in any event be punished; nor, above all, can it justify a broad interpretation of the regulation, as proposed by the Commission.

Moreover, the preceding observation highlights very significantly the inconsistency of the position adopted by the Commission in this case: as we have seen, it relies on the inappropriate nature of the legislative instruments furnished by Articles 85 and 86 for taking action against oligopolies, a fact which, in its view, justifies a broad interpretation of Article 2(3) of the regulation. However, on closer examination, the action taken by the Commission against the alleged K+S/SCPA/EMC duopoly consisted of an `order' to the first-named undertaking to sever its contractual links with the other two. In those circumstances, that is, faced with obvious signs of `concertation', why not use Article 85? (104)

97 I note, finally, that not even the practice followed in applying the regulation, referred to on a number of occasions in the course of the proceedings, offers any arguments in support of a different solution. In fact, in the Nestlé/Perrier decision, in which the Commission did not conceal its intention to develop an independent concept of collective dominance, the applicability of the regulation to oligopolies was based mainly on Article 3(g) of the Treaty. (105)

It is quite true that that provision is referred to in general terms in recital 1 in the preamble to the regulation and that, having the objective of protecting effective competition, it can only apply to any kind of anticompetitive situation, whether attributable to one or to more than one undertaking. It is equally true, however, and should be borne in mind, that that provision also states that it is to be implemented only `as provided for in this Treaty and in accordance with the timetable set out therein'.

98 All the factors which I have so far examined raise numerous doubts as to the possibility of relying on the concept of collective dominant position in the context of the provisions of the regulation on concentrations. In the final analysis, I consider that the regulation is not applicable to oligopolistic dominant positions and that this interpretation of the relevant provisions is the most consistent, not only with their wording, but also with the rationale on which they are based and, more generally, with the overall scheme of the system. While finding some justification in economic terms for the opposite solution, it would have a number of negative consequences; and above all, in view of the absence of similar reference parameters in the form of legislation and/or regulations to those existing in other antitrust systems, it would ultimately foster uncertainty. That would, moreover, be contrary to the requirement, emphasised on several occasions, that Community legislation must be unequivocal and its application must be predictable for those who are subject to it. (106)

Of course, as I noted earlier, the Commission's concerns relating to the negative consequences which the non-application of the regulation to oligopolies could have as regards the penalisation of anticompetitive conduct cannot and must not be overlooked. Moreover, it would be more appropriate, in my opinion, to translate such concerns into action in their natural context and to make them the subject of a proposal for an amendment aimed at completing the relevant provisions.

(3) Mistaken reliance in this case on the concept of collective dominant position

99 By this plea, both the French Government and the applicant companies maintain that a number of errors of law and of fact were committed in the appraisal of the concentration on the Community market apart from Germany.

In general, the French Government complains that the Commission failed, in assessing whether a collective dominant position existed, to follow either the criteria developed in the case-law on Article 86, or those established in its previous decisions, which, moreover, are more flexible than those suggested by the Community judicature. In support of its view, the French Government also makes a series of specific criticisms concerning all the factors taken into account in the decision, at the same time emphasising the importance of the criteria which the Commission omitted to take into consideration. Those arguments, it maintains, expose the errors by which the decision is vitiated as regards the existence of a collective dominant position and, in any event, the inadequate reasons on which it is based.

The applicant companies, which also dispute some of the factors taken into consideration in the decision, relying on arguments similar to those of the French Government, maintain in particular that the Commission has failed to prove the existence of a causal link between the concentration and the alleged creation of a collective dominant position.

100 The Commission counters that by saying that the criteria which it used in the contested decision in no way conflict with those employed in earlier decisions. In order to establish the existence of a collective dominant position in this case, it contends that it acted essentially on the basis of three criteria: the degree of market concentration which would result from the concentration, the structural factors relating to the nature of the market and the characteristics of the product, and the structural links between the undertakings concerned.

In addition, it asserts that the present case is distinguished by certain specific features from other cases in which it has taken action. First, prior to the concentration there was already no effective competition between the largest two suppliers of potash products because of the absence of K+S from the French market and because of the links existing between it and SCPA. Second, as a result of the enhancement of their competitive potential following from the merger with MdK, K+S and SCPA would find themselves in a dominant position in relation to other potash suppliers.

101 More generally, the Commission maintains, in replying to the French Government's objections, that the criteria developed in connection with Article 86, while not absent from the decision, are not decisive in regard to the control of concentrations. In this context, unlike that of Article 86 where it is defined primarily on the basis of market shares, the existence of a collective dominant position can only be the result of an analysis based on the future. The market shares held are therefore only a starting point; for the rest, it is essential to ascertain the scope for present and future competitors to act as a counterbalance to any dominant position held by the new undertaking. It is precisely following such an analysis that many concentrations, although having significant market shares and almost certainly such as to lead to a declaration of incompatibility with Article 86, have been approved under the regulation.

With regard to the criteria established in its decision-making practice, the Commission points out, on the one hand, that the decisions on which the French Government bases its arguments are all, with the exception of the Nestlé/Perrier decision, (107) later than the contested decision and, on the other, that it is the characteristics of each specific case which determine whether greater weight or even decisive significance should be attached to certain criteria rather than to others. It was therefore by reference to the particular features of this case that it considered the abovementioned criteria to be of paramount importance.

102 Let me begin by pointing out that the concept of collective dominant position, as developed in connection with Article 86 of the Treaty, which is to the effect that the hegemony from which abuse may result may be attributable either to one undertaking or to more than one, (108) has been clarified over the years. In particular, in its judgment in the `Flat Glass' case, the Court of First Instance made it clear that several undertakings hold a collective dominant position when, while independent, they are united on the market by particularly strong economic links, which could, for example, be the case where they jointly have a technological lead affording them the power `to behave to an appreciable extent independently of their competitors, their customers and ultimately of their consumers'. (109) The Court of Justice essentially confirmed that definition in its judgment in the Almelo case, stating that, in order for a collective dominant position to exist, the undertakings concerned `must be linked in such a way that they adopt the same conduct on the market'. (110)

With regard to the criteria developed in the decision-making practice on the subject, it is quite true, as the applicants maintain, that, in appraising concentrations, the Commission has developed a concept of collective dominance which is broader than, and in any case partly different from, that defined in the case-law relating to Article 86. It must, however, be acknowledged that, although it refers to the same factors as those used by Court in order to identify collective dominance, the regulation avoids any definition of the term, thereby not excluding a predominance of `structural' appraisals and hence the possibility of developing a concept somewhat independent of that developed in relation to Article 86. (111)

103 Clearly, a new concept of collective dominance, different from that developed in the case-law relating to Article 86, while possibly justified by the specific nature of the control of concentrations, is liable to render application of the concept in question to the system of rules governing concentrations even more problematical and unpredictable. Moreover, although Article 2(1) of the regulation, always assuming that the criteria listed therein for appraising concentrations in respect of their compatibility with the common market are also applicable to this case, does not specify any order in which the relevant factors are to be taken into account, the fact remains that the Commission cannot disregard the criteria listed and that the discretion which the regulation allows it must not be taken so far as to become arbitrary. In short, the Commission is clearly precluded from inventing whatever criteria it considers to be most appropriate in relation to each particular case.

In those circumstances, moreover, the Commission's practice, while unquestionably complex and not free from contradictions, may nevertheless prove to be a useful instrument for identifying possible inconsistencies and contradictions in the Commission's decision and in its defence and thus, by the same token, show conclusively whether or not the criteria used in this case were correct. If the issue is viewed in those terms, it seems to me that the Commission is not justified in claiming that there is no need to take account of the criteria and appraisals contained in decisions subsequent to the one at issue in this case. The Commission's practice, whether in earlier decisions - such as in the Nestlé/Perrier case often cited by the applicants (112) - or in later ones, is certainly not a parameter for determining the lawfulness of the contested decision, but is at least a guide to the method hitherto used by the Commission when applying the regulation to collective dominant positions.

104 That said, I now turn to the criticisms levelled by the French Government, and by the applicant undertakings, at the criteria applied in the decision in order to establish the existence of a collective dominant position liable, as forecast, to result in a significant distortion of competition.

In my view, it is appropriate, immediately after considering the preliminary aspect - the degree of market concentration - to deal with the issue of the structural links which, as the Commission itself has admitted, were the main factor which it relied upon in order to find the existence of a joint dominant position held by K+S/MdK and SCPA. I shall then examine the other criteria used in the decision, as well as the relevance of criteria employed in other decisions but not used at all in this case.

- Degree of market concentration

105 The Commission's appraisal is based, first and foremost, on the increase in the degree of market concentration (already high), which is considered such as to lead, in the future, to the reinforcement of a collective dominant position liable significantly to impede competition.

The French Government and the applicant undertakings dispute that appraisal, asserting that the increase is not substantial since the market shares held by K+S and SCPA rose, as a result of the concentration, from 54% to 61%. The French Government also contends that the Commission's analysis is arbitrary inasmuch as, on the one hand, it fails to take account of the fact that, following the concentration, the number of potash suppliers fell from ten to nine and, on the other, it did not take properly into consideration the role of two important suppliers, CPL and Coposa. Finally, the Commission's analysis is not reliable, in its view, because it calculates the market shares in two different ways, depending on the result to be achieved: on the basis of the first method, it takes into consideration only the sales volumes of the individual operators in the Community; on the basis of the second method, the sales made on the French market are instead included in SPCA's sales volume and accordingly deducted from K+S's sales volume.

106 It must be said, first of all, that the mere quantitative increase in the market shares held by the undertakings concerned, upon which the Commission laid stress, even excessively, in the decision, is not in itself sufficient to constitute the creation of a collective dominant position, as, moreover, the Court of First Instance emphasised in its judgment in the `Flat Glass' case, a point on which it was very clear. That is all the more so in this case, since the concentration led to an increase of only 7%.

It must therefore be acknowledged that the increase in the degree of market concentration resulting from the acquisition is not particularly important, being limited in terms of both the number of competitors operating on the market (which fell from ten to nine) and the increase in the total market share held by K+S and SCPA on the same market. In those circumstances, the importance attached in the decision to the criterion in question appears excessive. (113) In any event, the analysis made in this respect could, at most, show that the degree of market concentration is not such as to preclude automatically the existence of a collective dominant position.

- Structural links between K+S and SCPA

107 The parties agree that the central aspect of the contested decision is the finding of `the existence of exceptionally close links between the two companies [K+S and SCPA] extending over a long period of time', (114) links so close in fact as to suggest that there is no effective competition between the two undertakings. It was those very links which were the decisive factor taken into account by the Commission in its own appraisal; indeed, it considered that the severance of those links, through the conditions imposed, could bring to an end the alleged collective dominant position.

108 According to the French Government, the Commission made a series of errors of assessment in its examination of the structural links between K+S and SCPA, links which are not in fact capable of establishing a causal connection between the concentration and the creation of a K+S/SCPA duopoly since they are not sufficient to prove that the concentration in question created the conditions for collusive conduct by the two undertakings. In particular, (1) although the Commission identified the Community apart from Germany as constituting the relevant market, the links in question affect only France; (2) the decision should have included the Spanish undertaking Coposa in the oligopoly; (115) (3) the weak presence of K+S on the French market is not sufficient to show the existence of dominance by that undertaking together with SCPA; and (4) it has not been proven that the acquisition of MdK by K+S could lead to the creation of a duopolistic dominant position held by K+S and SCPA.

The applicant undertakings focus their attention principally on the latter aspect, pointing out that there are certain substantive contradictions in the Commission's position, in particular the fact that the concentration, according to the Commission itself, would have created, and not merely strengthened, a collective dominant position on the relevant market.

109 With regard to the first of those arguments, I would observe that the links between the companies concern, directly or indirectly, only the French market; it is therefore only in that market that they could be regarded as likely to have a bearing on the competitive relationship between K+S and SCPA. This is true, in the first place, as regards the severance of the distribution links between K+S and SCPA, which operate exclusively in the context of the French market. But it is also true, in the second place, as regards the interest in the Kali-Export cartel. According to the Commission, that interest, while not concerning sales on the Community market but only exports to non-Community countries, is such as to interfere with sales in France. That assertion cannot be verified, for lack of evidence; (116) however, if it were proven, it would, at the very most, demonstrate the absence of effective competition between K+S and SCPA on the French market, that is to say, on a market which, in the opinion of the Commission itself, does not constitute the relevant market. It follows that, if the Commission's argument were well founded, even the interest in the export cartel would, in the Community context, affect only the French market which, I repeat, is not a relevant market for the purposes of the decision.

In this regard, the conclusion reached by the French Government, that the Commission has not shown how the acquisition of MdK by K+S would lead to the creation of a duopoly capable of impeding competition on the Community market apart from Germany, seems to be well founded. It is not sufficient, for that purpose, to take into account the links existing between K+S and SCPA in France, which is not an independent market.

110 With regard, on the other hand, to the Spanish company Coposa, the French Government asserts that the Commission wrongly excluded it from the alleged oligopoly consisting of K+S and SCPA. In any case, it maintains, the Commission cannot argue, in order to demonstrate that the identification of the Community apart from Germany as the geographical market was correct, that Coposa's exports to France are significant and then state, in order to demonstrate the existence of a collective dominant position, that Coposa is not present to any significant extent in France and that in any event its sales are channelled.

Let me say at once that I am unconvinced by the argument put forward by the Commission in its defence that the links between K+S and SCPA are such as to characterise their relations towards all the other operators, and that the possible inclusion of Coposa in the oligopoly is not, in any case, a determining factor for the purposes of the decision since it would, by definition, only result in a strengthening of the collective dominant position. It is only too obvious that the exact identification of the members of an oligopoly is a matter of priority in terms of its appraisal, as is ascertaining the conditions of competition outside that oligopoly. (117) In this case, therefore, it seems difficult, on the basis of the available data and applying the criteria used for the inclusion of K+S and SCPA in the collective dominant position, to exclude Coposa from that oligopoly. It is sufficient to point out in this connection that Coposa is also a partner in the Kali-Export cartel and that it sells in France mainly through SCPA.

111 Nor does it appear to me to be decisive, for the purpose of establishing the absence of competition between K+S and SCPA, and therefore the existence of a collective dominant position held by them, that, as the French Government maintains in its third argument, K+S is not present to any significant extent in France and that it uses SCPA's distribution organisation for a proportion of its sales in that country. The first fact could have several explanations, such as, for example, the structure of the French market, consumer preferences or the German company's commercial strategy. Moreover, if more importance were to be attached to this aspect, it would be necessary to ask why similar situations, that is, absence from important national markets, arise for all the major operators in the sector; and, in particular, why, in the contested decision, with regard to the German market, the Commission attributed SCPA's absence not to the links with K+S, but to structural factors specific to that market.

With regard to K+S's product sales through SCPA, it should be emphasised that, although a large proportion of French imports of K+S products passes through SCPA, that represents such a small volume (1.4% of the relevant market) that it cannot constitute sufficient evidence to substantiate the Commission's argument.

112 That brings me to the French Government's final argument relating to the implications of the takeover of MdK by K+S: in particular, it maintains that the Commission has by no means proved that that acquisition gave rise to a duopolistic dominant position detrimental to competition. The applicant undertakings have also raised a similar argument, particularly emphasising the contradiction inherent in the position adopted by the Commission which, on the one hand, applied the failing company theory on the basis that MdK would have withdrawn from the market in any case, since its plight was now irreversible, and, on the other, took the view that the takeover of MdK by K+S, and the resulting concentration, were likely to create a duopolistic dominant position which might impede competition.

On this point, the Commission's rejoinder is that the creation of a joint dominant position by K+S/MdK and SCPA is in fact a consequence of the concentration, since the latter altered considerably the structural conditions of the market. In particular, the Commission draws attention to the following factors: (a) MdK accounts for 25% of the Community's total potash production; (b) at the time of the decision, the rate of utilisation of MdK's production capacity was down to 50%; and finally (c) the 7% market share is a crucial factor in view of the fact that supply outside the group comprising K+S/MdK and SCPA is fragmented and that the group's market share, according to the Commission's analysis, is bound to increase.

113 I observe, first of all, that the Commission's aforementioned statements do not answer the applicants' objections. Both the French Government and the undertakings maintain that there is an absence of any causal link between the creation of oligopolistic dominance and the concentration, an absence which stems from the fact that, according to the analysis made by the Commission in this regard, MdK was in any case bound to withdraw from the market. In short, how could the takeover of MdK result in the creation of a collective dominant position when - with or without a concentration - MdK's market share would in any case have been taken over, following its withdrawal from the market, by the other operators in the sector?

Even assuming that the Commission's analysis, according to which it is unreasonable to take the view that MdK's share of the Community market apart from Germany would almost certainly have accrued to K+S, (118) is correct, and that one of the criteria for the application of the failing company theory is therefore not met, the fact remains that the entire argument built up by the Commission appears, from the absolutely vital point of view of a causal link between the concentration and the creation of a joint dominant position by K+S and SCPA, to be rather weak. It follows that the significance attached in this context to MdK's 7% share of the relevant geographical market is by no means decisive.

- Characteristics of the eliminated competitor

114 According to the French Government, the Commission failed to analyse, as it had in the Nestlé/Perrier decision, the characteristics of the competitor eliminated from the market as a result of the concentration, namely, in this case, MdK. In the French Government's opinion, examination of those characteristics would necessarily have led to the conclusion that MdK's elimination from the market could not have any influence on the concentration since, moreover, according to the Commission's analysis, it was an undertaking which was in any case bound, albeit in the medium term, to withdraw from the market.

The French Government's argument is not without substance. In the decision given in the Nestlé/Perrier case, in which the eliminated competitor (Perrier) accounted for a share fluctuating between 35.9% and 44.2% of the market, the Commission took the view that `the reduction from three to two suppliers (duopoly) is not a mere cosmetic change in the market structure. The concentration would lead to the elimination of a major operator who has the biggest capacity reserves and sales volumes in the market. Perrier sources and brands would be divided between the two remaining suppliers. In addition, the reduction from three to only two national suppliers would make anticompetitive parallel behaviour leading to collective abuses much easier'. (119) The Commission thus drew attention to the implications and effects of the takeover of Perrier which, by virtue of its size and market share, was an essential factor in ensuring effective competition.

115 In the present case, on the other hand, the analysis in the decision seems, to say the least, to contain some gaps. While not seeking to attach excessive importance to the share of the Community market held by MdK (7%), (120) it is impossible not to have some doubts as to the actual effect of MdK's withdrawal on competition on the market. Moreover, the decision makes no reference to the `impressive production capacities' mentioned by the Commission in its defence (121) and which, moreover, appear difficult to reconcile with the characterisation of the undertaking as an economic entity that cannot be expected to survive.(122) Finally, there is nothing in the decision to suggest that MdK plays an essential role, even beyond its actual weight, in maintaining competition.

In conclusion, I take the view that the Commission has not shown that MdK's withdrawal from the Community market apart from Germany is likely to result in, or at least contribute to, the creation of a duopoly consisting of K+S/MdK and SCPA and capable of significantly impeding competition.

- Position of the competitors

116 The French Government claims that the decision contains a contradictory and inadequate analysis of the position of the competitors operating on the relevant market. In particular, it draws attention to the inconsistency of the Commission's attitude in minimising the market position of CPL and Coposa while specifically laying emphasis, for the purpose of identifying the relevant geographical market, on exports from Spain and the United Kingdom to the other Member States. The French Government also disputes the Commission's view that the channelling through SCPA of imports from countries outside the Community shows that the competitive pressure exerted by those countries on the duopoly is almost non-existent.

From this point of view also, the French Government's argument seems to highlight a number of objective gaps in the decision.

117 In the first place, the method used by the Commission, which, in order to show the absence of competitive pressure on the alleged oligopoly, puts forward as evidence the channelling of imports from non-Community countries into France through SCPA, (123) cannot be accepted. Above all, France is not, for the purposes of this case, an independent market. Consequently, the analysis of the factors involved in the appraisal of the concentration cannot take account solely of the French market, but must rather, and in any event, take into consideration the territory defined by the decision as the relevant market, namely, the market comprising the Community as a whole apart from Germany.

The Commission's observations on the significance of imports from the CIS, which, according to the figures supplied by the French Government (124) and not disputed by the Commission, have progressively increased, reaching 11% of sales within the Community, a level which is by no means insignificant, seem to be just as inadequate and contradictory. Moreover, the nine operators remaining on the market following the concentration hold shares which are such as to give rise to a situation in which the influence of the competitors of the duopoly allegedly consisting of K+S/MdK and SCPA would undoubtedly appear to be significant. (125)

- Comparison between the position of K+S/MdK and that of SCPA

118 In its criticism of the criteria applied in this case in order to establish the existence of oligopolistic dominance, the French Government also disputes both the excessive importance attached by the Commission to the total market share held by K+S and SCPA and the fact that it did not take account of the imbalance between the two undertakings, an imbalance which, in previous decisions, had instead constituted a significant reason for denying the existence of an oligopolistic dominant position.

The Commission points out that a market share of 60% clearly indicates the existence of a collective dominant position, in particular when, as in this case, there is an appreciable difference from the market shares held by the competitors. While acknowledging that there are considerable differences between the undertakings in question, it disputes the theory that a duopoly is conceivable only when the undertakings involved have similar characteristics.

119 With regard to the first aspect mentioned above, I note to begin with that the Commission has in the past ruled out the possibility of establishing thresholds for presuming the existence of collective dominance; and that it has in any case, in recent decisions, (126) authorised concentrations with combined shares equivalent to those calculated in this case. Consequently, while it is impossible, in theory, to deny that a 60% share is of some importance, such a share cannot, at least not by itself, be of decisive importance and constitute proof of the existence of an oligopolistic dominant position.

Furthermore, the judicial precedents cited by the Commission are not relevant, inasmuch as they refer to individual dominant positions, (127) which justifies the presumptive significance attached to a share of between 45% and 80% of the market. Moreover, in those judgments the market share is always analysed together with other factors such as the difference (of at least 50%) between the largest and second largest undertakings operating on the market.

120 Nor do I find myself able to share the Commission's view on the second aspect mentioned here, that is, the failure to take into consideration the absence of symmetry between the undertakings presumed to hold a collective dominant position.

I note first that the absence of symmetry between the undertakings concerned, which is undisputed in this case, (128) has been considered decisive by the Commission in other cases in order to exclude the existence of a duopoly, (129) whether from the point of view of the market shares taken as a whole and in relation to the markets of individual countries, or from that of `economic power', or, finally, from that of production capacity. It should be borne in mind in this context that in the Nestlé/Perrier decision the Commission made clear the implications of similarity in size and nature of the undertakings concerned and the reasons for taking it into account in its appraisal of the effects of the concentration, (130) pointing out in particular that similarity in size and nature of the undertakings is a clear deterrent to competition and makes it easier to engage in coordinated and collusive behaviour. Consequently, I consider that the Commission's defence, by merely treating the difference between K+S and SCPA as irrelevant for the purpose of establishing the existence of a duopoly, is inadequate. Nor do I consider that the Commission is justified in claiming that the practice which it has followed in applying the regulation is completely irrelevant in this case. Although practice is certainly not a mandatory reference parameter, but can be adapted according to the circumstances of each individual case, it certainly cannot be regarded in this context as wholly meaningless.

121 All the considerations expounded so far bring to light a series of gaps and contradictions in the appraisal and in the application of the criteria developed by the Commission, leading to the conclusion that the reasoning behind the decision is not consistent. The Commission has failed to prove to a sufficient standard, even on the basis of its own practice, the existence on the relevant market of a collective dominant position for K+S/MdK and SCPA which would be created as a result of the proposed concentration.

That leads me to deal only briefly with the other arguments put forward by the French Government with regard to the other factors taken into consideration in the decision. More specifically, I shall focus on the manner in which the aspects concerning the economic power of the customers, the barriers to entry and the market and product characteristics, as well as parallel conduct, were examined.

- Economic power of the customers

122 The French Government criticises the Commission for failing to take any account at all of the criterion relating to the customers' economic power, and in particular for failing to analyse the nature and influence of the producers and importers of potash, which could (and should) have led to the conclusion that the creation of an oligopolistic dominant position by K+S/MdK and SCPA was highly unlikely.

The Commission acknowledges the importance of the customers' economic power as a `counterbalance' to the duopoly and therefore capable of discouraging any parallel conduct; however, it adds that that is only one factor among others to be taken into consideration in order to reach a definitive appraisal of the effects of the concentration in question.

123 There is no doubt that in this case no account was taken of the influence and characteristics of the various types of customer for the products in question. That makes it more difficult to establish whether, as the French Government maintains, the large multinational producers of compound fertilisers in fact have a degree of purchasing power capable of preventing the members of the oligopoly from strengthening their position or whether, on the contrary, as the Commission asserts, it is the bulk blenders of fertilisers who are prominent among the customers. Nor is it specified whether those bulk blenders are actually small and medium-sized undertakings which are not in a position to exert adequate pressure on the producers, or whether, as the applicant government asserts, their activity is also carried out by the subsidiaries of the large multinationals. Clearly, an answer to this question would have been essential in order to determine whether or not the customers have effective bargaining power.

The Commission's position therefore seems weak, all the more so as the customers' negotiating power is undoubtedly an important factor in ascertaining the conditions of competition, since it may discourage the adoption of anticompetitive behaviour. Moreover, it is to that effect that the Commission's analysis, in both the Pilkington-Techint/SIV case and the Nestlé/Perrier case, (131) must be interpreted.

124 In view of the abovementioned uncertainties, other aspects raised by the French Government can be left aside, even though they relate to the customers' characteristics. I shall merely add, for the sake of clarity, that the statement by the Court of First Instance in the `Flat Glass' judgment that the Commission `is not required to discuss in its decisions each argument advanced by the undertakings concerned', (132) contrary to that institution's contention in the course of the proceedings, does not provide any support for the Commission's argument.

The customers' power is not an argument relied on by the applicants at the administrative stage and is not included in the decision, but is an important factor to be taken into account in determining whether or not an oligopolistic dominant position exists. As such, it must necessarily be examined, failing which the decision may contain a lacuna such as to affect its substance. Besides, precisely in the judgment which I have just cited, the Court found, by reference to Article 86, that the analysis of those aspects of the market which the applicants alleged had not been taken into consideration `was absolutely essential for an assessment of the question of dominance', (133) thus confirming the necessity for the Commission to base its findings on investigations which are as full and exhaustive as possible.

- Barriers to entry

125 The Commission is criticised for not making any analysis, in the part of the decision concerning the existence of a collective dominant position, on the presence of barriers to entry, a factor regarded as highly important in the Commission's practice. That is all the harder to explain since attention was drawn to the absence of such obstacles in the decision itself for the purposes of defining the relevant geographical market.

The Commission acknowledges the absence of barriers to entry within the Community. However, it adds that in this case there are two types of barriers to entry for undertakings from countries outside the Community, namely, anti-dumping measures for imports from the CIS, and SCPA's statutory monopoly whereby all imports into France from non-Community countries are channelled through that undertaking.

126 Let me say at once that I very much doubt whether the aforementioned factors are capable of constituting barriers to entry. Anti-dumping duties are certainly not barriers to entry since, as the Commission itself has admitted, they are intended to restore competition and not to obstruct trade. (134) In addition, their actual effect on the level of imports has by no means been established, and certainly not on the basis of objective data. If anything, the appraisals relating to the individual national markets, contained in the decision (paragraph 64), raise the opposite presumption.

With regard to SCPA's monopoly, moreover, the defective approach already noted on several occasions, that is, the Commission's tendency to make an analysis limited to France, as if the French market constituted the relevant geographical market for the purpose of the decision, features again here. In addition, even without considering in detail the extent of the effects of such a monopoly, the existence, as the Commission itself acknowledges, (135) in individual Community countries of flows of imports from non-Community countries on such a scale as to represent a predominant share of the market, leads rather to the conclusion that the presence of barriers to entry has not been established.

- Characteristics of the market and the product

127 Finally, with regard to the characteristics of the market and the product, usually regarded as significant factors to be taken into account in the analysis, I shall merely point out, with the French Government, that among the Commission's assessments, which are, moreover, characterised by a certain conciseness, there is no detailed and specific analysis either of the prices charged by K+S and SCPA, (136) which are an absolutely essential factor in establishing market transparency (which, as is well known, makes uniform conduct easier) and in assessing elasticity of demand, or of the cost structures of the two undertakings, which, if they are different, may constitute an incentive to competitive behaviour.

It is unnecessary to point out that such omissions, relating to decisive aspects, contribute towards both the weakness of the Commission's analysis and its results. I am therefore of the opinion that there is no need to examine in detail the other `structural' factors mentioned, such as those relating to the homogeneity of the product, on which, moreover, I do not share the French Government's position, as I have already pointed out in examining the definition of the relevant geographical market.

- Existence of parallel conduct

128 As the French Government has rightly asserted, the Commission's assessments in this regard are by no means convincing. Indeed, the circumstances from which the Commission seeks to derive evidence for the existence of joint dominance by K+S/MdK and SCPA are of no significance, especially if they are compared with those relied on in other decisions.

It is only too obvious that the agreement between K+S and SCPA, declared incompatible with Article 85 in 1973, (137) must be regarded as too weak a basis on which to infer, after an interval of more than 20 years, that there is no competition between K+S and SCPA.(138) Similarly, from the point of view under discussion here, the considerations relating to the minor presence of K+S on the French market are not decisive and may well have a different explanation, as I have pointed out in examining the structural links between K+S and SCPA.

129 Those observations are also supported by the fact that the factors taken into account for these purposes in previous decisions were of a completely different nature. In the Nestlé/Perrier decision, for example, after drawing attention to the much-reduced level of price competition, the Commission found that there was a high degree of parallel pricing by the three producers concerned by the concentration, which was combined with the development of instruments of transparency capable of facilitating coordination and mutual monitoring of pricing policies. (139) Moreover, in that same case, the Commission took into consideration, as evidence of the links between the undertakings concerned, their response to the takeover bid by an outside operator, a response regarded as `a joint entry deterrence action'. (140) In the Pilkington-Techint/SIV case, on the other hand, the Commission took the view that certain agreements between the undertakings concerned, which had been condemned in 1988, that is, only five years prior to the decision, and by the Bundeskartellamt only a few months prior to the decision, were such as to foster anticompetitive parallel conduct. (141)

The factors in question were, even prima facie, considerably more significant and conclusive as evidence than those taken into account in this case.

130 In the light of all the considerations set out above, I am of the opinion that the Commission has not consistently applied, in this case, the criteria which it has developed in order to ascertain the existence of a collective dominant position. That is not all; even on the assumption that, as already suggested, the concept of collective dominant position can be adapted to the specific features of the control of concentrations, as well as to the characteristics of this particular case, it is none the less necessary for the Commission to adduce, in the context of the control of concentrations, sufficient evidence to establish the existence of a causal link between the concentration and the collective dominant position. (142)

There is no such evidence in this case. Indeed, there is substantial evidence for rejecting the Commission's view. In conclusion, I am of the opinion that the Commission has by no means shown that the concentration examined creates a position of joint dominance by K+S/MdK and SCPA which could significantly impede competition.

D - Imposition of conditions and obligations affecting third parties not involved in the concentration (Cases C-68/94 and C-30/95)

131 Assessment of the plea alleging the unlawful imposition of obligations on persons who are not parties to the concentration clearly presupposes, on the one hand, that the collective dominant position is relevant for purposes of applying the regulation and, on the other, that the Commission has in fact shown that a collective dominant position exists in this case.

On the basis of the conclusions which I have reached so far, in particular with regard to the use in this case of the concept of collective dominant position, it would not seem necessary to examine the plea in question. However, for the sake of completeness, I consider it appropriate to examine the arguments put forward by the applicants in this context as well.

132 The French Government maintains that, even assuming it to be applicable to collective dominant positions, the regulation does not allow a declaration of compatibility to be made subject to conditions and obligations which, as in this case, directly affect the position of third parties who are not involved in the concentration. To that end, it asserts that Article 8(2) of the regulation authorises the Commission only to accept, from the notifying undertakings, commitments concerning the concentration in the strict sense and not commitments which substantially affect third parties as well.

The applicant undertakings, which agree with the French Government that it is impossible, on the basis of the regulation, to impose conditions and obligations affecting third parties, essentially claim that, even if the existence of a collective dominant position is considered proven in this case, the fact remains that the conditions laid down are neither suitable nor appropriate for the maintenance of effective competition on the relevant market.

(1) Possibility of applying Article 8(2) to undertakings which are not parties to the concentration

133 Let me say at the outset that the French Government's view must be endorsed. Article 8(2) of the regulation authorises the Commission to make the declaration of compatibility subject only to `conditions and obligations intended to ensure that the undertakings concerned comply with the commitments they have entered into vis-à-vis the Commission with a view to modifying the original concentration plan'. That wording indicates that it does not fall within the Commission's powers to take action with respect to operators who are not involved in the concentration, even if it considers that they form an oligopoly together with the undertakings which notified the plan.

In other words, Article 8(2) clearly and expressly limits the Commission's power to the right to obtain from the undertakings concerned commitments entered into with a view `to modifying the original concentration plan'; it therefore precludes the adoption of any structural changes in the competitive context in which the concentration is created. The power to modify the plan cannot therefore be transformed, in the absence of an express provision to this effect, into a more general power to restructure an entire market and/or industrial sector.

134 That solution does not seem to me to be capable of serious contradiction by the Commission's arguments that: (a) the decision imposed conditions and obligations only on the undertakings which were parties to the concentration, and those conditions and obligations were intended to ensure compliance with the commitments entered into by the parties which notified the plan; and (b) the prohibition of a concentration on the basis of the fact that the conditions in question might affect the interests of third parties is contrary to the objectives pursued by the regulation and is not compatible with the principle of proportionality.

With regard to the latter aspect, I refer to what I stated earlier when examining the applicability of the regulation to collective dominant positions, bearing in mind economic considerations and the objectives of that regulation. (143) As far as the former aspect is concerned, it is sufficient to observe that although, as the Commission has maintained, it is in the very nature of the control of concentrations that the commitments entered into by the notifying parties may, however indirectly, affect third parties, in this case the conditions at issue certainly do not merely affect third parties, but concern them objectively and subjectively, at least to the same extent as they concern the parties themselves. Nor does the practice relied on by the Commission allow the emphasis of the problem to be shifted: the decisions referred to do not impose any conditions directly concerning third parties and are therefore not comparable with the contested decision. (144) Moreover, and in any event, the fact remains that `a mere practice cannot override the provisions of the Treaty', (145) or, I would add, provisions contained in secondary legislation which clearly conflict with the Commission's practice. As I have already pointed out on another occasion, it would be paradoxical to treat a breach of the rules as legitimate merely because of its repetition!

(2) Whether the conditions imposed in this case for the purpose of maintaining effective competition are appropriate

135 This point has, in particular, been raised by the applicant undertakings which claim that the conditions in question are neither necessary nor appropriate for the purpose of maintaining effective competition on the relevant market; as such, those conditions are unsuitable for attaining the objective referred to in Article 8(2) of the regulation.

In substance, SCPA and EMC cite the arbitrary nature of the conditions laid down by the contested decision, which did not and do not in any way concern the concentration plan, but relate solely to contractual relations and partnerships subjectively and objectively unconnected with it, and, furthermore, are to be complied with in a geographical context different from the relevant market. (146)

136 I note as a preliminary point that the effectiveness from the point of view of competition of the conditions laid down appears doubtful, both on account of the geographical discrepancy, to which attention has already been drawn, between the subject-matter of the conditions and the relevant market, and on account of the lack of evidence showing that those conditions are capable of ensuring that the concentration complies with the provisions of the regulation. From that angle, it would seem possible to endorse the applicant undertakings's view, namely, that, in determining the conditions to be imposed pursuant to Article 8(2) of the regulation the Commission should act on the basis of serious and detailed appraisals and not of hypothetical considerations such as those relied on by it in its defence to justify the conditions imposed in this case.

Since, it will be recalled, I have concluded that the action brought by the applicant undertakings is inadmissible in so far as it seeks annulment of the conditions relating to the Potacan joint venture, the brief analysis which follows will focus on the conditions concerning the withdrawal from the Kali-Export partnership and the termination of the distribution contracts, merely referring, of course, to those points already dealt with exhaustively in the analysis of the structural links between K+S and SCPA. (147)

- Condition concerning the withdrawal of K+S from the Kali-Export cartel

137 The applicant undertakings argue first of all that the existence of Kali-Export and the shares held by K+S, MdK and SCPA in the capital of that undertaking are not likely to have the slightest effect on the conditions of competition within the Community. In any case, the Commission has by no means established the anticompetitive effect of K+S's interest in the Kali-Export cartel.

The Commission counters that the structure of the market in question was such that the possibility of cooperation between K+S and SCPA in the Kali-Export cartel having an indirect effect on the anticompetitive behaviour of the members of the cartel on the Community market could not be ruled out. In particular, it points out that: (a) K+S operates only on the margins of the French market, (b) CPL had access to the French market, increasing its sales there by 13% in six years, only after withdrawing from the Kali-Export cartel, and (c) Coposa does not operate to any significant extent on the French market where, moreover, a substantial proportion of the potash produced by it is sold by SCPA.

138 The arguments put forward by the Commission, as I have just outlined them, do not seem to me to be capable of answering the objections raised by the applicant undertakings. The fact remains that the Commission has failed to show that the weak presence of K+S on the French market is attributable to its membership of the Kali-Export export cartel, while that situation, as the applicant undertakings have pointed out, could well be explained by historical and geographical factors. Moreover, although, as the Commission has maintained, the principles relating to the application of Articles 85 and 86 cannot be applied to concentrations, the Commission can certainly not base its appraisal on mere presumptions and probabilities. It has a duty, in any case, to show that the conditions in question are necessary in order to ensure the maintenance of effective competition.

In this case the Commission has not only failed to explain convincingly how the absence of cooperation in a cartel which produces its effects on the market outside the Community is likely to ensure effective competition on the Community market apart from Germany, but it has also failed to prove that the weak presence of K+S or Coposa on the French market is attributable to membership of that cartel. With regard to the fact that CPL allegedly increased its exports to France only after its withdrawal from Kali-Export, it is self-evident that this could be attributable specifically to a decision by CPL to concentrate on the Community market or, as claimed by the applicants and not disputed by the Commission, to the operating difficulties experienced by CPL during the period when it was a member of Kali-Export.

- Condition relating to termination of the distribution contracts between K+S and SCPA

139 In the opinion of the applicant undertakings, the Commission has in no way shown how and why termination of the distribution contracts between K+S and SCPA would be likely to ensure the maintenance of competition on the Community market apart from Germany. In their view, those links actually produce effects only on the French market.

Moreover, the links in question consist of a contract of 28 June 1985 for the distribution of products containing, not potash, but kieserite. For the rest, SCPA confines itself to purchasing potash used by EMC or intended for sale outside the French market. Finally, SCPA purchases from K+S potash products containing magnesium and sold in France, but those products, as is clear from the decision, are not covered by the relevant product market.

140 In view of the situation which I have just described, and which, moreover, is not disputed by the Commission which merely relied on the importance of the establishment by K+S of its own distribution organisation on the French market, it is only too obvious that K+S and SCPA had no special relationship for the distribution of potash products. However, it is worth pointing out that the wording of the decision led K+S to terminate the only existing distribution link with SCPA: the contract relating to kieserite, that is to say, I repeat, to a product which in the decision was not considered to be covered by the relevant market.

That factor, more than any other assessment, helps to confirm the validity of the applicants' argument concerning the direct impact of the decision on their interests, and also reinforces doubts as to the expediency of the conditions imposed for the purpose of eliminating the duopolistic dominant position allegedly created by the concentration.

Conclusions

141 The foregoing considerations show, in my opinion, that in this case mistakes were made in the interpretation and application of the provisions of the regulation on concentrations.

It is necessary, at this stage, to establish whether those mistakes warrant annulment of the decision as a whole or only its partial annulment, that is, only of the part containing the conditions concerning the applicant companies. For reasons which I shall explain below, I am of the opinion that the latter solution is the most appropriate.

142 First of all I regard as unfounded the Commission's argument that partial annulment of the decision would alter the institutional balance since the Community judicature is not competent to amend the Commission's decision by substituting itself for the Commission in the appraisal of the concentration.

I shall merely observe in this connection that the right of the Court to annul only partially a decision given in subjecta materia derives from the regulation itself, as well as from the Court's settled case-law. (148) It would moreover be paradoxical to maintain that the exercise by the Court of a power expressly conferred on it undermines the rights and powers of the Commission. Allow me instead to reaffirm that, for the purposes of partial annulment, the conditions laid down on the matter in the case-law, which are, in substance, intended to ensure that partial annulment does not result in the contested decision being emptied of all substance, must be met. (149)

143 In view of the foregoing, I am of the opinion that it is possible to annul the decision in part and that this solution can be adopted in respect of any of the pleas put forward by the parties which I have considered well founded and which the Court sees fit to accept. Whether the Court inclines towards inapplicability, in principle, of the regulation to collective dominant positions, or whether it considers that a collective dominant position is non-existent or in any event insufficiently proven, the premiss on which the Commission based its imposition of the contested conditions on the applicant undertakings would be negated. Those conditions, let us remember, were laid down in order to prevent the concentration from creating a situation of oligopolistic dominance which could impede competition on the Community market.

Those observations would, in my opinion, also apply should the Court come to the conclusion, once the problems connected with the application of the provisions of the regulation to oligopolies have been dealt with and the existence of a collective dominant position has been established, that Article 8(2) of the regulation does not authorise the Commission to impose conditions which directly affect the interests of third parties not involved in the concentration. Similarly, if the conditions in question were considered invalid for the purpose of ensuring the maintenance of effective competition on the relevant market, the most appropriate solution would be, once again, to annul them and thus annul the decision in part.

144 In conclusion, it would be possible, in all the various solutions which I have just outlined, to annul the conditions without emptying the decision as a whole of all substance. (150) It is hardly necessary to point out in that respect that the appraisal of the effects of the concentration on the Community market, of which the disputed conditions form part, is completely and objectively independent of that concerning the German market, a fact which could form the basis of a limitation ratione materiae of the operative part of the decision in such a way as to satisfy the requirements laid down in the relevant case-law.

Nor, on the other hand, would partial annulment of the decision be likely to undermine its scope. Furthermore, in this case, taken as a whole, the contested decision is favourable to the undertakings concerned, a factor which, according to the Court's case-law, is relevant for the purpose of granting an action for partial annulment. (151)

145 With regard to costs, I propose, pursuant to Article 69(3) of the Rules of Procedure, that the parties should be ordered to bear their own as far as Case C-68/94 is concerned. As regards Case C-30/95, however, I am of the opinion, on the basis of the same provision, that the fact that the Commission has been essentially unsuccessful requires that it be ordered to pay the costs, including those incurred by the intervening undertakings. The governments which have intervened in the cases in question should, on the other hand, be ordered to bear their own costs.

146 In the light of the foregoing observations, I therefore conclude by proposing that the Court should:

- in Case C-68/94:

(a) annul Article 1 of Commission Decision 94/449/EC of 14 December 1993 relating to a proceeding pursuant to Council Regulation (EEC) No 4064/89 (Case No IV/M.308 - Kali + Salz/MdK/Treuhand), in so far as it makes the declaration of the concentration's compatibility subject to compliance with the conditions and obligations referred to in paragraph 63;

(b) dismiss the rest of the action;

(c) order the parties, and the German Government which has intervened, to bear their own costs;

- in Case C-30/95:

(a) declare the actions brought by SCPA and EMC admissible, save for the plea concerning the undertaking in relation to the Canadian company Potacan;

(b) annul Article 1 of Commission Decision 94/449/EC of 14 December 1993 relating to a proceeding pursuant to Council Regulation (EEC) No 4064/89 (Case No IV/M.308 - Kali + Salz/MdK/Treuhand), in so far as it makes the declaration of the concentration's compatibility subject to compliance with the conditions and obligations referred to in paragraph 63;

(c) order the Commission to pay the costs, including those of the intervening undertakings, and the French Government to bear its own costs.

(1) - OJ 1990 L 257, p. 14.

(2) - Commission Decision relating to a proceeding pursuant to Council Regulation (EEC) No 4064/89 (Case No IV/M.308 - Kali + Salz/MdK/Treuhand), published in OJ 1994 L 186, p. 38.

(3) - Judgment in Case 6/72 [1973] ECR 215, paragraph 26.

(4) - For that view, see, for example, Santamaria, Diritto commerciale comunitario, Milan, 1995, p. 92 et seq., which cites as the first example of that practice Commission Decision 85/78/EEC of 12 December 1984 (Mecaniver-PPG) (OJ 1985 L 35, p. 54).

(5) - Judgment in Joined Cases 142/84 and 156/84 BAT and Reynolds v Commission [1987] ECR 4487. However, for examples of opposition to applying the `Philip Morris' judgment beyond that particular case, see, inter alia, Bellamy, `Mergers outside the Scope of the New Merger Regulation. Implications of the Philip Morris Judgment', in Fordham Corp. Law Institute, 1988, p. 22.

(6) - See Frignani-Waelbroeck's reconstruction, Disciplina della concorrenza nella CE, Turin, 1996, pp. 685 to 686.

(7) - Literally: `where: (a) two or more previously independent undertakings merge, or (b) - one or more persons already controlling at least one undertaking, or - one or more undertakings acquire, whether by purchase of securities or assets, by contract or by any other means, direct or indirect control of the whole or parts of one or more other undertakings'.

(8) - This condition is fulfilled `where ... the combined aggregate worldwide turnover of all the undertakings concerned is more than ECU 5 000 million', or `... the aggregate Community-wide turnover of each of at least two of the undertakings concerned is more than ECU 250 million', unless `each of the undertakings concerned achieves more than two-thirds of its aggregate Community-wide turnover within one and the same Member State' (Article 1(2)).

(9) - They are: `the market position of the undertakings concerned and their economic and financial power, the alternatives available to suppliers and users, their access to supplies or markets, any legal or other barriers to entry, supply and demand trends for the relevant goods and services, the interests of the intermediate and ultimate consumers, and the development of technical and economic progress provided that it is to consumers' advantage and does not form an obstacle to competition'.

(10) - In the United States, the 1984 Merger Guidelines introduced the so-called Herfindhal-Hirschman Index (HHI) which totals the squares of the market shares of the individual undertakings involved in the concentration. In order to establish the actual increase in the degree of market concentration resulting from the operation, the product of the market shares of the undertakings involved is doubled. If the HHI Index is under 1 000 after the concentration, the operation will, as a rule, be authorised. If it is between 1 000 and 1 800, the concentration will be examined and may be prohibited where it involves an increase of more than 100. Finally, if the HHI is above 1 800, the concentration may be prohibited even if it entails an increase of less than 50.

(11) - The word `significantly' has given rise to lively debate in academic literature: among the most authoritative interpretations is that which makes it subject to the de minimis test, to the effect that only concentrations which have substantial effects on competition, reducing it significantly and permanently, can be declared incompatible. See to that effect, Langeheine, `Substantive Review under the EEC Merger Regulation', in Fordham Corporate Law Institute, 1990, p. 484 et seq. In that regard, see also Bellamy and Child, Common Market Law of Competition, 4th edition, London, 1993, p. 336 et seq. With regard to the meaning of `effective competition', the Court's case-law makes reference to the actual context in which competition would occur in the absence of the limitation in question: see, in that regard, judgment in Case 31/85 ETA/DK Investment [1985] ECR 3933, paragraph 11. See also Dechery, `Le règlement communautaire sur le contrôle des concentrations', in Revue trimestrielle de droit européen, 1990, p. 317, in particular p. 323.

(12) - The concentration must be notified not more than one week after the conclusion of the agreement, or the announcement of the public bid, or the acquisition of a controlling interest.

(13) - Appraisal of a concentration with a `Community' dimension clearly presupposes definition of the reference economic sector, whether in terms of the product or in terms of the geographical area to be taken into account. The relevant instructions for this purpose are contained in form CO set out in Annex I to Commission Regulation (EEC) No 2367/90 of 25 July 1990 on the notifications, time limits and hearings referred to in Commission Regulation (EEC) No 4064/89 (OJ 1990 L 219, p. 5), since repealed and replaced by Commission Regulation (EC) No 3384/94 of 21 December 1994 (OJ 1994 L 377, p. 1). That form lists all the information to be provided to the Commission at the time of notification.

(14) - The favourable opinion of the Advisory Committee on Concentrations, expressed by a majority at the meeting of 3 December 1993, is published in OJ 1994 C 199, p. 5.

(15) - The Commission concluded that, in the industrial potash and salt sectors, the concentration would not create or reinforce a dominant position as a result of which effective competition would be significantly impeded in the Community or in a substantial part thereof (paragraph 11 of the decision).

(16) - See paragraphs 13 to 29 of the decision.

(17) - See paragraphs 30 and 45 of the decision.

(18) - See paragraph 69 of the decision.

(19) - See paragraphs 31 to 44 of the decision.

(20) - Leaving aside the amounts earmarked to meet own requirements, the overwhelming majority of which consist of intra-group supplies made by K+S to BASF, K+S has a market share of approximately 79% and MdK one of approximately 19%. That gives MdK GmbH a total market share of 98% (paragraph 46 of the decision).

(21) - See paragraph 70 of the decision.

(22) - See paragraph 95 of the decision.

(23) - As is pointed out in paragraphs 51 and 52 of the decision, the merged company K+S/MdK and the French company SCPA would together account for 80% of potash production in the Community (K+S 35%, MdK 25% and SCPA 20%). Excluding internal consumption by the undertakings concerned, and therefore on the basis of sales alone, that gives a combined market share of approximately 50% (K+S 17%, MdK 8% and SCPA 25%). However, since SCPA also markets considerable quantities of potash from other producers, including from countries outside the Community, taking into account all sales controlled by K+S, MdK and SCPA in the Member States apart from Germany, that gives a combined market share of 60% (K+S 16%, MdK 7% and SCPA 37%).

(24) - See paragraphs 54 to 56 of the decision.

(25) - See paragraphs 57 to 62 of the decision.

(26) - In this case, the German Government has intervened in support of the Commission's conclusions.

(27) - In this case, the French Government has intervened in support of the applicants' conclusions, whereas K+S and MdK have been granted leave to intervene in support of the Commission's conclusions.

(28) - SCPA and EMC's action was brought before the Court of First Instance and referred by it to the Court of Justice because the French Government's action (Case C-68/94) relating to the same decision was already pending before the latter. Indeed, since the same applicant companies had requested that the two cases be dealt with together by the Court of Justice, the Court of First Instance, by order of 1 February 1995, declined jurisdiction in accordance with the third paragraph of Article 47 of the Statute and transmitted the documents to the Court of Justice.

(29) - Order in Case T-88/94 R Société Commerciale des Potasses et de l'Azote and Entreprise Minière et Chimique v Commission [1994] ECR II-263.

(30) - Case T-88/94 R Société Commerciale des Potasses et de l'Azote and Entreprise Minière et Chimique v Commission [1994] ECR II-401.

(31) - On the role of the Court of Justice in competition matters, Trabucchi's observations are still very helpful: `Sviluppa della giurisprudenza e suo significato nella disciplina comunitaria sulla concorrenza', in Rivista di diritto civile, 1973, I, p. 497 et seq.

(32) - See, with reference to the United States, Leddy, `The 1992 US Horizontal Merger Guidelines and Some Comparisons with EC Enforcement Policy', in ECLR, 1993, p. 15.

(33) - From the point of view of formal legality, such a review means checking the aspects connected with jurisdiction, grounds, procedural regularity and observance of the right to a fair hearing; from that of substantive legality, on the other hand, it takes the form of scrutiny of the accuracy of the economic and market analysis, of the anticompetitive effects and the correctness of the legal consequences (from the point of view of the characterisation of the facts, for example) drawn from that analysis, without, of course, encroaching on the scope of the discretion which the Commission enjoys in the application of the rules on competition. See, in this connection, the papers by Canivet and Biancarelli in Le contrôle juridictionnel en matière de droit de la concurrence et des concentrations (proceedings of the seminar organised by the Court of First Instance of the European Communities on 22 and 23 November 1993), Luxembourg, 1994, pp. 25 to 38 and 55 to 65.

(34) - Judgment in Joined Cases 100/80 to 103/80 Musique Diffusion Française v Commission [1983] ECR 1825.

(35) - In this connection, I note that the favourable view of concentrations, to which I have referred above, is confirmed in the pragmatic approach taken by the Commission in its application of the regulation, which is characterised by a preference for the adoption of `negotiated' solutions, which are in any case favourable to the plans in question. In the end, it has frequently resorted to the power to approve concentrations subject to compliance with certain conditions and obligations and, conversely, made very limited use of the power of prohibition. In that regard, see Van Bael-Bellis, Il diritto della concorrenza nella CE, 1995, pp. 555 to 556.

(36) - See, on the experience of the United States, the remarks by Hawk, `Public Private Enforcement of Merger Law in the United States', in Le contrôle juridictionnel en matière de droit de la concurrence et des concentrations, cited in footnote 33, p. 81 et seq., in particular pp. 83 and 84, who notes that `the legality of mergers is ultimately determined in the courts according to the judicial standards developed under Section 7 of the Clayton Act, although the Guidelines are increasingly consulted'.

(37) - Under that provision, `the Commission may revoke the decision it has taken pursuant to paragraph 2 where: (a) the declaration of compatibility is based on incorrect information for which one of the undertakings is responsible or where it has been obtained by deceit; or (b) the undertakings concerned commit a breach of an obligation attached to the decision'.

(38) - Judgment in Joined Cases T-68/89, T-77/89 and T-78/89 SIV and Others v Commission [1992] ECR II-1403, paragraph 320.

(39) - Judgment in Case 17/74 Transocean Marine Paint Association v Commission [1974] ECR 1063, paragraph 21.

(40) - Literally: `Where the Court of Justice gives a judgment which annuls the whole or part of a Commission decision taken under this regulation, the periods laid down in this regulation shall start again from the date of the judgment'.

(41) - Judgments in Case 25/62 Plaumann v Commission [1963] ECR 95, in particular p. 107; Case C-198/91 Cook [1993] ECR I-2487, paragraph 20; Case T-83/92 Zunis Holding v Commission [1993] ECR II-1169, paragraph 36.

(42) - Judgments in Case 26/76 Metro v Commission [1977] ECR 1875, paragraph 13; Case 191/82 Fediol v Commission [1983] ECR 2913, paragraph 28 et seq.; Case 169/84 Cofaz and Others v Commission [1986] ECR 391, paragraph 23. In more recent case-law, see the judgments in Case T-37/92 BEUC and NCC v Commission [1994] ECR II-285, paragraph 36, and Case T-114/92 BEMIM v Commission [1995] ECR II-147, paragraph 26.

(43) - Judgments in Case C-354/87 Weddel v Commission [1990] ECR I-3847, paragraphs 20 to 23; and Case T-2/93 Air France v Commission [1994] ECR II-323, paragraph 45, in which the Court held that France's action against a Commission decision concerning a concentration between two of its competitors was admissible by reason, inter alia, of the fact that `in assessing the competitive situation on the two markets identified as being the markets concerned after the concentration, the Commission mainly took into account the position of Air France'.

(44) - See, albeit with reference to a different case, the judgment in Cofaz v Commission, cited in footnote 42, in particular paragraphs 28 and 29.

(45) - As was made clear by the Court in Plaumann v Commission, cited in footnote 41, `The words and the natural meaning [of Article 173] justify the broadest interpretation. Moreover provisions of the Treaty regarding the right of interested parties to bring an action must not be interpreted restrictively' (p. 219).

(46) - See, in this respect, Barav and Vandersanden, Contentieux communautaire, Brussels, 1977, p. 172.

(47) - Judgment in Case T-83/92 [1993] ECR II-1169.

(48) - See the judgment in Zunis Holding, cited above, paragraphs 34 to 36; also Cook-Kerse, EC Merger Control, London, 1996, pp. 200 to 201.

(49) - See paragraph 60 of the decision.

(50) - With regard to competition, see the judgments in Case 53/85 AKZO Chemie v Commission [1986] ECR 1965 and Case 60/81 IBM v Commission [1981] ECR 2639; more generally, on the concept of a contestable act as provided for in Article 173, see, along the lines of that provision, the judgments in Case C-325/91 France v Commission [1993] ECR I-3283, paragraph 9, and Case C-476/93 P Nutral v Commission [1995] ECR I-4125, paragraph 28.

(51) - Regulation No 17 of the Council of 6 February 1962 (OJ, English Special Edition 1959-1962, p. 87).

(52) - Judgment in Joined Cases C-89/85, C-104/85, C-114/85, C-116/85, C-117/85 and C-125/85 to C-129/85 Ahlström Osakeytiö and Others v Commission [1993] ECR I-1307, paragraph 181.

(53) - Judgment in Case 76/79 Könecke [1980] ECR 665, paragraph 9.

(54) - SCPA's total sales in Belgium and Luxembourg were in fact 22 000 tonnes, whereas the figure given in the documents transmitted by the Commission was 221 000 tonnes.

(55) - The Commission itself has stated in its observations that these were mostly statistical data which were easily obtainable and that, in any case, Article 19 of the regulation guarantees Member States access to the file.

(56) - Judgment in Case C-142/87 Belgium v Commission [1990] ECR I-959, paragraph 48. To the same effect see, among others, the judgments in Case 30/78 Distillers Company v Commission [1980] ECR 2229, paragraph 26, and Case 234/84 Belgium v Commission[1986] ECR 2263, paragraph 30.

(57) - Article 19(5) provides, however, that the Commission may shorten that period `as appropriate in order to avoid serious harm to one or more of the undertakings concerned by a concentration'.

(58) - In this connection, it is worth recalling that `the Commission shall take the utmost account of the opinion delivered by the Committee. It shall inform the Committee of the manner in which its opinion has been taken into account' (Article 19(6)). The Commission may therefore disregard the Committee's opinion, or review its own position and adopt a final decision which is different from the draft decision submitted for the Committee's opinion. In this regard, the case which led to the adoption of Commission Decision 94/208/EC of 31 January 1994 declaring a concentration to be compatible with the common market (Case No IV/M.315 - Mannesmann/Vallourec/Ilva) (OJ 1994 L 102, p. 15) strikes me as exemplary. In that case, the draft decision on which the Advisory Committee delivered its opinion was to the effect that the concentration was incompatible with the common market (see the Opinion of the Advisory Committee in OJ 1994 C 111, p. 6), whereas the Commission's final decision was to the opposite effect, although it is not known whether this was by reference to the opinion of a minority of the Advisory Committee. On the point examined in the body of the text, see also, with regard to the Advisory Committee established on the basis of Regulation No 17, the judgment in Case T-69/89 Radio Telefis Eireann v Commission [1991] ECR II-485, paragraphs 21 to 27.

(59) - See, on the 1992 Guidelines, Leddy, cited above in footnote 32.

(60) - This recital provides that `the Commission must place its appraisal within the general framework of the achievement of the fundamental objectives referred to in Article 2 of the Treaty, including that of strengthening the Community's economic and social cohesion, referred to in Article 130a'.

(61) - See Article 2(3). The Commission's practice seems to have followed this line. Prior to the decision at issue in this case, no decisions can be found in which the considerations of a social nature referred to prevailed over the need to safeguard competition. I refer in this connection to the Aérospatiale-Alenia/de Havilland decision in which the Commission did not accept the parties' argument based on the likely elimination of de Havilland from the market in the event that the concentration was not approved, instead considering that there was no such likelihood [Commission Decision 91/619/EEC of 2 October 1991, Case No IV/M.053 - Aérospatiale-Alenia/de Havilland (OJ 1991 L 334, p. 42)].

(62) - See paragraph 71 of the decision; emphasis added.

(63) - The failing company theory arose from concerns relating to economic policy and corporate efficiency. Subsequently, however, in its application, political and social considerations emerged, connected, inter alia, with the need to safeguard employment. See, in this respect, the judgment in United States v General Dynamics Corp., 415, US 486, 507 (1974); and Areeda Turner, cited above, vol. IV, 1980, p. 941.

(64) - In this case, whereas the concentration involves activities representing 3 000 jobs in total, the Peine group's offer was limited to the purchase of an operation affecting only 536 jobs, that is, 18% of the jobs guaranteed by the concentration. Although primarily a matter for the legislature, the balance between the competition policy objectives and other aims of the Treaty is not outside the scope of the Court's case-law either: see, for example, the judgment in the Europemballage and Continental Can v Commission case (cited above in footnote 3), and also the judgment in Joined Cases 154/78, 205/78, 206/78, 226/78 to 228/78, 263/78 and 264/78, 31/79, 39/79, 83/79 and 85/79 Valsabbia and Others v Commission [1980] ECR 907.

(65) - Judgments in Case T-96/92 CCE de la Société Générale des Grandes Sources and Others v Commission [1995] ECR II-1213, paragraphs 28 and 29, and Case T-12/93 CCE de Vittel and Others v Commission [1995] ECR II-1247, paragraphs 38 and 39.

(66) - Commission Decision 92/553/EEC of 22 July 1992 relating to a proceeding under Council Regulation (EEC) No 4064/89 (Case No IV/M.190 - Nestlé/Perrier) (OJ 1992 L 356, p. 1).

(67) - With reference to the concept of relevant market tout court, the Court, in its judgment in Case T-29/92 SPO and Others v Commission [1995] ECR II-289, pointed out that this aspect is more important for the purpose of applying Article 86 than for the purpose of applying Article 85. The remarks made on that occasion seem to me to be equally relevant to a dominant position arising from a concentration.

(68) - Judgment in the `Flat Glass' case, cited above in footnote 38, paragraph 360.

(69) - Form CO, as provided for by Regulation No 2367/90, cited above in footnote 13.

(70) - The applicant undertakings, on the other hand, seem to agree with the Commission with regard to the interchangeability of products, but assert that this depends on how the relevant product market is defined and therefore does not affect the identification of the relevant geographical market.

(71) - Commission Decision 91/403/EEC of 29 May 1991 (Case No IV/M043 - Magneti Marelli/CEAc) (OJ 1991 L 222, p. 38).

(72) - Commission Decision of 6 November 1990 (Case No IV/M004 - Renault/Volvo) (OJ 1990 C 281, p. 2).

(73) - Nestlé/Perrier decision, cited above in footnote 66.

(74) - On the other hand, reasoning a contrario, trade patterns characterised by mutual interpenetration have always been regarded by the Commission as an important, although not decisive, indication of belonging to the same geographical market: see the Mannesmann/Vallourec/Ilva decision cited above in footnote 58, paragraphs 33 to 37. In this respect, and to the same effect, see Cook and Kerse, cited above in footnote 48, p. 145 et seq.

(75) - Judgment in Joined Cases 40/73 to 48/73, 50/73, 54/73 to 56/73, 111/73, 113/73 and 114/73 Suiker Unie v Commission [1975] ECR 1663, paragraph 372.

(76) - For example, in the decision it asserts that `transport costs ... tend to vary ... In any case, they do not appear to be of a level capable of preventing significant trade flows within the Community outside Germany' (paragraph 42 of the decision).

(77) - See the judgment in Case 27/76 United Brands v Commission [1978] ECR 207, paragraphs 52 and 53.

(78) - See, for example, the Magneti Marelli/CEAc decision, cited above in footnote 71.

(79) - Public monopolies are prominent among barriers to entry: see, for example, the Commission Decision of 29 April 1991 (Case No IV/M.063 - ELF/Ertoil) (OJ 1991 C 124, p. 13).

(80) - See section 53 of the defence.

(81) - Data cited in section 78 of the reply.

(82) - Judgment in the United Brands v Commission case, cited above in footnote 77, paragraphs 44, 52 and 53.

(83) - For this view, see, for example, Siragusa-Subiotto: `Il controllo delle concentrazioni a livello CEE: una prima analisi critica delle decisioni della Commissione', in Giurisprudenza Commerciale 1992, I, 233, in particular pp. 271 and 272. The position of some writers is more qualified, since they merely draw attention to the difficulty of accepting the opposite view: thus, among others, Cook and Kerse, cited above in footnote 48, p. 134 et seq.; Korah: EC Competition Law and Practice, London, 1994, p. 242; Bellamy and Child: Common Market Law of Competition, London, 1993, pp. 343 and 344.

(84) - Apart from the fact that, as the Court has consistently held, verbal declarations have limited legal significance (see, for example, the judgment in Case C-292/89 Antonissen [1991] ECR I-745, paragraph 18) and that the declaration in question cannot in any event be so construed as to restrict the scope of Articles 85 and 86, the fact remains that such a declaration certainly cannot be used in support of the conclusion that the legal basis constituted by Article 87 of the Treaty is in fact limited in scope, with the result that the concept of collective dominant position cannot be linked to Article 86.

(85) - For this view, see also Cook and Kerse, cited above in footnote 48, p. 134.

(86) - The Commission has drawn attention to this in, for example, the Decision of 18 December 1991 (Case No IV/M.165 - Alcatel/AEG Kabel) (OJ 1992 C 6, p. 23), paragraph 22.

(87) - See, for example, Commission Decision 94/359/EC of 21 December 1993 declaring a concentration compatible with the common market (Case No IV/M.358 - Pilkington-Techint/SIV) (OJ 1993 L 158, p. 24). I would point out, moreover, that, in the Court's case-law on Article 86, a market share, while constituting important evidence of the existence of a dominant position where equal to or greater than 80%, is not necessarily a decisive factor, but must be taken into consideration together with other factors: to this effect, see, among others, the judgment in Case 247/86 Alsatel v Novasam [1988] ECR 5987, paragraph 19.

(88) - In this connection, for example, the German Law against restrictions on competition, which is actually based on a structural approach linked to a system of presumptions by reference to the market shares held by undertakings, provides for a threshold of at least 50%. More specifically, Paragraph 23(2)(2) of that Law provides: `For the purposes of the control of concentrations, a group of undertakings shall also be regarded as being in a dominant position where:

1. it comprises three or fewer undertakings which hold the highest market share and account, in aggregate, for a 50% share, or

2. it comprises five or fewer undertakings which hold the highest market share and account, in aggregate, for a two-thirds share, unless they show that conditions of competition are such that significant competition can be expected to take place between them even after the concentration, or where all those undertakings together do not hold a dominant position in relation to the remaining competitors ...'.

(89) - It is sufficient to refer, for this purpose, to the Sixteenth Report on Competition Policy, 1986, in particular section 333.

(90) - See p. 36 of the rejoinder.

(91) - COM(88) 734/Final.

(92) - Trabucchi, cited above in footnote 31, p. 503, referring to the case-law on competition, has stated that `the frequent reference to economic data and the significance attached to them in reaching judicial decisions are by no means contrary to the essential function of the Court, which is to comply with and enforce the Treaty and the legislation adopted under it ... When the rule exists, it must be complied with fully ...'.

(93) - It is only too obvious that the damage suffered by the applicant companies as a result of the decision is the direct consequence of the behaviour adopted by the undertakings involved in the concentration in compliance with the Commission's requirements. This was perceived moreover by the President of the Court of First Instance hearing the application for interim measures. In the Order in Case T-88/94 R (cited in footnote 30), which suspended the operation of Article 1 of the contested decision in so far as it requires withdrawal from Kali-Export, the President notes that the possibility cannot be excluded that the condition in question `might injure the rights of third parties, in this case those of the two other partners in Kali-Export ... which were not parties to the proceedings before the Commission' (paragraph 28).

(94) - Cited above in footnote 13. In the same way as Regulation No 99/63/EEC of the Commission of 25 July 1963 on the hearings provided for in Article 19(1) and (2) of Council Regulation No 17 (OJ, English Special Edition 1963-64, p. 47), this regulation governs only the hearings of the parties to the concentration (Articles 13 and 14), whereas for third parties, such as the applicant undertakings in this case, according to the Commission, it provides for an arrangement for them to be heard upon application, without any procedural guarantees (Article 15). In this regard, the French Government has pointed out that such hearings are in practice merely contacts with Commission officials and can in no way satisfy the requirements of informing and protecting persons who are fully involved in the proceedings, such as the `parties' to the oligopoly which would be created following the concentration. It must be made clear that Regulation No 2367/90 has been replaced by Regulation No 3384/94 (cited in footnote 13), the provisions of which are, in certain respects, more protective. As far as is relevant here, however, the terms of the problem have remained essentially unchanged. While it is true that third parties who have submitted written observations may, pursuant to Article 16 of the new regulation, take part in an official hearing, it is also true that such a possibility is at the Commission's discretion.

(95) - As the French Government points out (pp. 20 to 23 of the application), proceedings under Article 86 are, in accordance with Regulations No 17 (cited in footnote 51) and No 99/63 (cited in the previous footnote), always inter partes with the undertakings deemed to be in a dominant position, which are, from the very outset, informed of the proceedings initiated against them and given the opportunity to defend themselves.

(96) - Judgment in the Belgium v Commission case, cited in footnote 56, paragraph 27. This means, as the Court has pointed out, that `the person against whom an administrative procedure has been initiated must have been afforded the opportunity, during that procedure, to make known his views on the truth and relevance of the facts and circumstances alleged and on the documents used by the Commission to support its claim that there has been an infringement of Community law'. See to the same effect the earlier judgment in Case 85/76 Hoffmann-La Roche v Commission [1979] ECR 461, paragraphs 9 and 11.

(97) - It is worth pointing out that, following repeated requests on their part, the representatives of the applicant companies were received by Commission officials only on 1 December, that is, two days before the meeting of the Advisory Committee and thus when the die was already cast. Even if it were true that, as the Commission asserts but the applicants deny, SCPA's representatives were received by the officials of the institution as early as 3 November 1993, the position would remain unchanged, given the essential `opacity' of the whole procedure for third parties, all the more so as the Commission states that it was only on 1 December that the undertakings given by K+S were discussed with the French company's representatives.

(98) - The need for this is confirmed by the fact that all the main antitrust systems apply the rules on the control of concentrations to oligopolistic dominance as well: for a comparative analysis, see Briones: `Oligopolistic dominance: is there a common approach in different jurisdictions? A review of the decisions adopted by the Commission under the Merger Regulation', in ECLR, 6, 1995, p. 334 et seq.

(99) - It is precisely from this perspective that certain writers have by no means ruled out the possibility of applying the regulation to cases of oligopolistic dominance where effective competition is significantly impeded by several undertakings which, together, are in a position to act independently of their competitors and customers. See, for example, Jones, Gonzales-Diaz: `The EEC Merger Regulation', London 1992, p. 168 et seq.; and Venit: `The Evaluation of Concentrations under Regulation 4064/89: The nature of the beast', in International Mergers and Joint Ventures, Annual Proceedings of the Fordham Corporate Law Institute, 1991, p. 519, in particular p. 540 et seq.

(100) - Based on the pioneering study on the subject by Cournot: Recherches sur les principes mathématiques de la théorie des richesses, Paris 1848, the economic theory of the oligopoly has given rise to a large number of remarkable contributions: for a summary see Pigassou: Les oligopoles et le droit, Paris, 1983, Ch. III, pp. 151 to 260.

(101) - Areeda and Turner: Antitrust Law, vol. III, Boston/Toronto, 1978, p. 359: `oligopoly with monopoly-like results'. This view finds support in economic theory which identifies, in similar situations, the same conditions of exercise of market power justifying the application of antitrust provisions to monopoly situations; and which even considers that the oligopoly may be a source of greater dangers for competition than individual dominant positions, since the latter tend to dwindle away in time under the effect of competition from smaller undertakings, whereas a strong oligopoly may, in certain circumstances, last for a long time. See, in that regard, Scherer and Ross: Industrial Market Structure and Economic Preference, Boston, 1990.

(102) - Judgment in the `Woodpulp' case, cited above in footnote 52, paragraph 71.

(103) - On this question, in all its implications, see the detailed Opinion of Advocate General Darmon in the `Woodpulp' case [1993] ECR I-1445. See also, for a recent exhaustive study of the case-law on the subject, Soames: `An Analysis of the Principles of Concerted Practice and Collective Dominance: A Distinction without a Difference', in ECLR 1996, p. 24.

(104) - The answer to this question may lie in the need for expediency. It would be confirmed, for example, by the fact that one of the undertakings originally provided for, namely that relating to Potacan, which was subsequently converted into an informal solution examined above (under Admissibility), was in fact the subject-matter of a proceeding initiated pursuant to Regulation No 17.

(105) - Decision 92/553, cited above in footnote 66. The decision goes on to state that, if the regulation were not applicable to oligopolistic situations, it would follow that the national authorities, which before its introduction would have been able to penalise such situations under their national provisions, would have abandoned such control altogether `without any substitute for it at Community level' (paragraph 115).

(106) - Judgment in Case 70/83 Kloppenburg [1984] ECR 1075, paragraph 11.

(107) - Decision 92/553, cited above in footnote 66.

(108) - Despite this, the Court initially seemed reluctant to recognise the concept of collective dominance, partly because of the difficulty of identifying, where not specified in the legislation, the constituent elements of a firm definition of that concept. Thus, on one occasion, it seemed to restrict the application of Article 86 to cases of individual dominant positions. In the judgment in Hoffmann-La Roche (cited in footnote 96), the Court stated that `a dominant position must also be distinguished from parallel courses of conduct which are not peculiar to oligopolies in that in an oligopoly the courses of conduct interact, while in the case of an undertaking occupying a dominant position the conduct of the undertaking which derives profits from that position is to a great extent determined unilaterally'. See in this respect Korah: `Concept of a dominant position within the meaning of Article 86', in Common Market Law Review, 1980, p. 395, in particular p. 398.

(109) - Judgment in the `Flat Glass' case, cited above in footnote 38, paragraph 358. To the same effect, see the judgment in the Hoffmann-La Roche case, cited above in footnote 96, paragraphs 38 and 48.

(110) - Judgment in Case C-393/92 [1994] ECR I-1477, paragraph 42. The principles set forth in that judgment were subsequently confirmed by the Court in Joined Cases C-140/94 to C-142/94 DIP and Others v Comune di Bassano del Grappa and Comune di Chioggia [1995] ECR I-3257, paragraphs 24 to 26. See, in this respect, the Opinion in that case of Advocate General Fennelly who pointed out, among other things, that the concept of collective dominance cannot be used as a substitute for the lack of any direct evidence of the existence, and still less of any abuse, of a dominant position (point 64).

(111) - For this view, see Venit, cited in footnote 99, p. 527 et seq., in particular pp. 530 to 531. This approach is also supported by the idea that the concept of `impeded competition', as contained in Article 2(3) of the regulation, would allow the Commission to develop a new concept of `dominance' based on a probability analysis with reference to the characteristics of the market. See Brittan: `The Law and Policy of Merger Control in the EEC', in European Law Review, 1990, p. 354.

(112) - In this connection, I note moreover that the wording and scheme of the Nestlé/Perrier decision do not seem exempt from criticism, mainly because they tend to develop a test for identifying collective dominant positions which is legally rather vague. See in this regard the remarks of Winckler and Hansen: Collective dominance under the EC Merger Control Regulation, in Common Market Law Review (1993) 30, p. 787 et seq., in particular p. 828, who take the view that `if the standard used in Nestlé/Perrier is, as suggested, one of "probability" or "facilitation", then the Commission has not explained in sufficient detail what types of economic conduct ("collective abuses") it wishes to avoid making "much easier"' (this statement refers to paragraph 120 of the decision).

(113) - See paragraphs 52 to 56.

(114) - Paragraph 57 of the decision.

(115) - This is the company which markets the potash produced by certain Spanish undertakings which were recently grouped together in a State-owned corporation known as INI.

(116) - The Commission claimed in its pleadings that it had sufficiently demonstrated, in paragraphs 57, 59, 61 and 67 of the decision, that such interference exists. A reading of the paragraphs in question, however, shows that the Commission merely drew attention in them to the structural links existing between K+S and SCPA.

(117) - In this connection, I would point out, for example, that in a checklist of factors relevant to the control of concentrations, adopted in 1990 by the Federal Cartel Office in Germany (published in English in Fordham Corporate Law Institute, 1991, p. 161 et seq.), the absence of competition between the members of an oligopoly and the `outside' competitors is treated as an essential factor for characterising the oligopoly as dominant.

(118) - See paragraph 79.

(119) - Decision 92/553, cited above in footnote 66, paragraph 120.

(120) - This share, moreover, is the same as that held by SIV, whose acquisition by Pilkington and Techint was authorised, without the imposition of any conditions, by Decision 94/359, cited above in footnote 87.

(121) - See p. 32 of the defence.

(122) - Paragraph 76 of the decision.

(123) - See pp. 33 and 34 of the defence and p. 54 of the rejoinder.

(124) - See p. 42 of the rejoinder.

(125) - See the figures in the table on p. 43 of the French Government's reply, which the Commission has not disputed.

(126) - See, for example, Decision 94/359, cited above in footnote 87.

(127) - They are the judgments in the Hoffmann-La Roche case, cited above in footnote 96; Case 322/81 Michelin v Commission [1983] ECR 3461; and Case C-62/86 AKZO v Commission [1991] ECR I-3359.

(128) - See the Commission's defence, pp. 37 to 58.

(129) - For example, in the Pilkington-Techint/SIV decision, cited above in footnote 87, paragraph 55; conversely, in the Mannesmann/Vallourec/Ilva decision (cited above in footnote 58), the similarity in size and nature of the undertakings involved in the concentration, which held shares of 36 and 33% respectively, was regarded by the Commission as an incentive to adopt parallel conduct.

(130) - Decision cited above in footnote 66, paragraph 123.

(131) - In that decision, cited in footnote 66, the power of the customers was considered, after a detailed analysis which is lacking in this case, to be insufficient to constrain significantly the power of the suppliers (see, in particular, paragraphs 77 to 89).

(132) - Judgment in the `Flat Glass' case, cited above in footnote 38, paragraph 363.

(133) - Ibid., paragraph 363.

(134) - See p. 61 of the rejoinder.

(135) - See p. 61 of the rejoinder.

(136) - See, in this connection, the point I have already made with regard to the identification of the relevant geographical market.

(137) - OJ 1973 L 217, p. 3. See paragraph 57 of the decision.

(138) - Even without taking account systematically of the principle laid down in the judgment in the `Flat Glass' case, in which it is made clear that `for the purposes of establishing an infringement of Article 86 of the Treaty, it is not sufficient ... to "recycle" the facts constituting an infringement of Article 85, deducing from them the finding that the parties to an agreement or to an unlawful practice jointly hold a substantial share of the market, that by virtue of that fact alone they hold a collective dominant position, and that their unlawful behaviour constitutes an abuse of that collective dominant position' (judgment cited above in footnote 38, paragraph 360).

(139) - See paragraphs 59, 121 and 122 of the decision.

(140) - See paragraph 127 of the decision.

(141) - Decision 94/359, cited above in footnote 87, paragraph 32. It is worth adding here that the factor in question was not considered such as to lead to the imposition of conditions, still less to the refusal of authorisation of the concentration.

(142) - If this were not so, the finding of the existence of a collective dominant position detrimental to competition would ultimately be made to depend on mere assumptions detached from any specific evidence and therefore wholly unreliable.

(143) - See points 94 to 98 above.

(144) - The Commission refers, in particular, to the Decisions of 4 May 1993 in the KNP/BT and VRG case (OJ 1993 L 217, p. 35); 19 December 1991 in the Courtaulds/SNIA case (OJ 1991 C 333, p. 16); and 31 July 1991 in the Varta/Bosch case (OJ 1991 L 320, p. 26).

(145) - Judgment in Case C-327/91 France v Commission [1994] ECR I-3641, paragraph 36.

(146) - The scope of those conditions shows, rather, that the Commission could, if it had seen fit, have acted against the alleged participants in the duopoly (SCPA/EMC and K+S) by relying on other instruments, and in particular on Article 85, as it did, moreover, in a parallel case to this one, concerning the joint Canadian subsidiary Potacan, which is the subject of an investigation under Regulation No 17 (cited in footnote 51).

(147) - See points 107 to 113 above.

(148) - Judgment in Case 2/56 Geitling v High Authority [1957] ECR 3.

(149) - See already, to that effect, the judgment in Case 37/71 Jamet v Commission [1972] ECR 483, paragraphs 11 and 12.

(150) - In those circumstances, moreover, that possibility would be available to the Commission, by virtue of the reopening of the time-limits provided for in Article 10(5) of the regulation, by laying down new and different conditions or else by showing more fully and coherently how the original conditions would take effect from the point of view of protecting competition. However, while such a course of action is quite permissible in theory, it does not seem likely to me, since it implies re-examining a concentration which has already produced its effects and which would therefore render an appraisal in the light of the previously existing state of affairs very difficult.

(151) - Judgment in Transocean Marine Paint Association, cited above in footnote 39, paragraph 21.