Opinion of the Advocate-General
1. In these appeal proceedings, the British Aggregates Association (‘BAA’) is requesting that the Court of Justice set aside the judgment of the Court of First Instance of 13 September 2006 (‘the judgment under appeal’) (2) dismissing the action brought by BAA against the Commission’s decision of 24 April 2002 not to raise objections to the tax on aggregates introduced by the United Kingdom (3) (‘the contested decision’).
2. That judgment, in so far as it rules on the admissibility of the action at first instance, is also the subject of a cross-appeal by the Commission.
I – The facts which gave rise to the appeal
3. The facts which gave rise to the present proceedings are described as follows in the judgment under appeal.
4. Aggregates are granular materials used in construction which are chemically inert. They may be used as they are, for example as construction fill, or may be mixed with binders such as cement or bitumen.
5. The aggregates levy (‘the AGL’) was introduced in the United Kingdom by certain provisions of the Finance Act 2001 (‘the Act’) which entered into force on 1 April 2002.
6. A number of amendments to the tax arrangements were then introduced by the Finance Act 2002. One of these was the introduction of an exemption for spoils resulting from the extraction of certain materials, including slate, shale, ball clay and china clay. In addition, provision was made for the phased introduction of the AGL in Northern Ireland.
7. The AGL is charged at the rate of GBP 1.60 per tonne where the taxable materials are commercially exploited within the United Kingdom. Aggregates which are exported or removed from the United Kingdom before being processed are exempt.
8. The Act, as amended, provides that aggregate is not taxable in four cases: (i) if it is expressly exempted; (ii) if it has previously been used for construction purposes; (iii) if it has already been subject to an aggregates levy; or (iv) if, on the commencement date under the Act, it was not on its originating site.
9. BAA is an incorporated association, the members of which are small independent quarrying companies in the United Kingdom.
10. By letter of 24 September 2004, the Commission received from two undertakings unconnected with BAA a complaint concerning certain aspects of the AGL – in particular, the exclusion of certain materials from its scope, the exemption for exports and the derogations relating to Northern Ireland – which they considered to be incompatible with the Treaty provisions on State aid.
11. On 20 December 2001, the United Kingdom notified to the Commission a State aid scheme entitled ‘Phased introduction of the aggregates levy in Northern Ireland’.
12. By letter of 6 February 2002, the Commission invited the United Kingdom to submit its comments on the complaint and to provide further information relating to the AGL. The United Kingdom responded by letter of 19 February 2002.
13. On 11 February 2002, BAA brought an action for judicial review of the AGL before the High Court of Justice of England and Wales, Queen’s Bench Division.
14. By letter of 15 April 2002, BAA submitted a complaint to the Commission in which it argued that the exclusion of certain materials from the scope of the AGL and the exemption of exports constituted State aid, and that the derogations relating to Northern Ireland, which had been notified by the United Kingdom, were incompatible with the common market.
15. By judgment of 19 April 2002, the High Court of Justice dismissed BAA’s action.
16. On 24 April 2002, the Commission adopted the contested decision. It found that the AGL did not entail any elements of State aid within the meaning of Article 87(1) EC, and that the exemption for Northern Ireland was compatible with the common market.
17. BAA lodged an appeal against the judgment of the High Court of Justice before the Court of Appeal (England and Wales), which ordered the proceedings to be stayed pending a ruling by the Court of First Instance in the action which BAA had meanwhile brought against the contested decision.
II – Proceedings before the Court of First Instance and the judgment under appeal
18. By application lodged at the Registry of the Court of First Instance on 12 July 2002, BAA sought annulment of the contested decision except in so far as it concerned the exemption for Northern Ireland. The United Kingdom was granted leave to intervene in the proceedings in support of the forms of order sought by the Commission.
19. The action was based on four pleas in law. By its first plea, BAA claimed that the Commission had infringed Article 87(1) EC. By its second plea, it alleged failure to state adequate reasons for the contested decision. By its third and fourth pleas, respectively, BAA alleged that the Commission had infringed its duty to initiate the formal investigation procedure and had failed to carry out its obligations in relation to the preliminary examination stage of the procedure.
20. The Commission first of all challenged the admissibility of the action, on the ground that BAA did not have locus standi to bring proceedings under the fourth paragraph of Article 230 EC. It argued that BAA had not demonstrated that it was individually concerned by the contested decision, in that it had failed to prove that the competitive position of at least one of its members had been significantly affected by the AGL. On the merits, the Commission maintained that BAA’s pleas were unfounded.
21. By the judgment under appeal, the Court of First Instance dismissed the action in its entirety, after declaring it to be admissible, and ordered BAA to pay the costs.
III – Proceedings before the Court of Justice and forms of order sought
22. By application lodged at the Registry of the Court of Justice on 27 November 2006, BAA lodged an appeal against the judgment of the Court of First Instance.
23. BAA claims that the Court should set aside the judgment under appeal. It also claims that the Court should annul the contested decision, save as regards the exemption for Northern Ireland. Lastly, BAA claims that the Commission should be ordered to pay the costs of both the appeal proceedings and the proceedings before the Court of First Instance.
24. The Commission contends that the Court should set aside the judgment under appeal and declare the action before the Court of First Instance to be inadmissible. In the alternative, it contends that the Court should declare the appeal to be inadmissible and/or dismiss it as unfounded. Lastly, it contends that BAA should be ordered to pay the costs.
25. The United Kingdom, which intervened in the proceedings before the Court of First Instance, contends that the appeal should be dismissed.
IV – Legal analysis
26. By its cross-appeal, the Commission challenges the judgment of the Court of First Instance in so far as it ruled on the admissibility of the action before it. The Commission maintains that that action was inadmissible and that, in consequence, the principal appeal is inadmissible also. It is therefore necessary to consider the cross-appeal first.
A – The cross-appeal
27. The Commission contends, first, that in declaring BAA’s action to be admissible, the Court of First Instance erred in law. According to the Commission, the AGL is a measure of general application which affects a potentially unlimited number of operators in the United Kingdom and cannot, therefore, be treated as being of individual concern to BAA’s members. It follows that the Commission’s decision approving the AGL is also of general application.
28. Secondly, the Commission contends that the Court of First Instance failed properly to examine the condition that the competitive position of BAA’s members must be significantly affected by the AGL.
29. I shall first summarise in brief the case-law concerning the conditions for the admissibility of actions against decisions adopted in the context of the procedure for monitoring State aid, and then move on to consider the two main arguments on which the Commission bases its plea that the action before the Court of First Instance was inadmissible.
1. The case-law concerning the conditions for the admissibility of actions against decisions adopted in the context of the procedure for monitoring State aid
30. In the context of the procedure for monitoring State aid, the decisions which the Commission adopts are addressed solely to the Member States concerned.
31. It has been settled case-law since Plaumann v Commission (4) that persons, other than those to whom a decision is addressed, may claim to be individually concerned for the purposes of the fourth paragraph of Article 230 EC only if that decision affects them by reason of certain attributes peculiar to them or by reason of circumstances which differentiate them from all other persons, and thus distinguishes them individually, as in the case of the persons addressed.
32. In the State aid sector, that rule was first clarified in Cofaz . (5) Drawing on the case-law concerning actions against decisions adopted by the Commission on the basis of Regulation No 17, (6) the Court recognised in Cofaz that where competitors of State aid beneficiaries have taken an active part in the formal investigation procedure, they have locus standi to challenge a decision adopted on completion of the formal investigation procedure , but only if their position on the market is significantly affected by the aid. (7) In subsequent judgments, the fact of having taken part in the procedure gradually ceased to be regarded as significant, and the Community judicature largely focused on evaluating the economic impact of the measure on the market, on the view that the effect on the applicant’s competitive position was the primary criterion for the purposes of assessing whether the action was admissible.
33. In the early 1990s, in the well-known Matra (8) and Cook (9) judgments, the Court – after pointing out the diversity of the objectives which, in the context of the procedure for monitoring State aid, characterise the preliminary examination under Article 88(3) EC and the formal investigation procedure under Article 88(2) EC – laid the foundations for an approach whereby the conditions for admissibility vary according to the procedural stage on conclusion of which the decision under challenge has been adopted. Without referring to Cofaz , but following essentially the same line of reasoning, the Court pointed out in Matra and Cook that, in the case of a decision by the Commission not to initiate the procedure under Article 88(2) EC, the only way that the persons intended to benefit from the procedural guarantees laid down in that provision can ensure that those guarantees are complied with is if they are recognised as having the right to challenge the Commission’s decision before the Community judicature. In both those judgments, the Court therefore concluded that the actions were admissible because they had been brought by persons who fell to be recognised as ‘parties concerned’ for the purposes of Article 88(2) EC.
34. As is generally known, in recent appeal proceedings in which the Commission sought a review of the principles established in Cook and Matra , together with the subsequent application of those principles by the Court of First Instance, the Court of Justice confirmed and refined that case-law. In paragraphs 34 to 37 of Commission v Aktionsgemeinschaft Recht und Eigentum ( ARE ), (10) the Court summarised the conditions for admissibility which currently govern the bringing of an action against a decision which has been adopted without initiating the procedure under Article 88(2) EC. Those conditions vary depending on the objective sought by the person bringing the action. If the applicant is seeking to protect his procedural rights under Article 88(2) EC and is essentially trying to make sure that the Commission opens the formal investigation procedure, the action is declared admissible, provided only that the applicant is able to demonstrate that he is a con cerned party for the purposes of that provision. If, on the other hand, the person bringing the action questions ‘the merits of the decision appraising the aid as such’, the mere fact that he may be regarded as a concerned party is not sufficient, and the applicant must prove that he meets the more stringent conditions for admissibility laid down in Plaumann : for example, by proving that his market position would suffer significant adverse effects if the measure provided for in the contested decision were applied.
35. To summarise, on the basis of the case-law referred to above, the conditions governing the admissibility of actions against Commission decisions on State aid vary according to the procedural stage at which they are adopted and the objective sought by the person bringing the action, and the margin of variation essentially reflects the difference in the degree of damage to his own interests which that person is required to prove.
36. It is against the background of that case-law that the pleas relied on by the Commission in support of its cross-appeal must be considered.
2. The alleged failure by the Court of First Instance to take into account the fact that the AGL is a measure of general application
37. The Commission contends that, when considering the admissibility of the action before it, the Court of First Instance correctly ruled out application of the Cook and Matra case-law, after pointing out that BAA was not seeking merely to ‘challenge the Commission’s refusal to initiate the formal investigation procedure’, but ‘also [to call] into question the merits of the contested decision’. (11) According to the Commission, the Court of First Instance erred, however, in finding that the decision concerned BAA individually. The Commission points out that the AGL is a legislative act of general application which introduces a fiscal charge, and that the conditions for levying that charge are established in objective and abstract terms. That being so, the AGL is liable to have a negative impact on a potentially unlimited number of undertakings and not just on the members of BAA. According to the Commission, the Court of First Instance failed to consider the nature of the measure at issue, and merely assessed the effects of that measure on three of BAA’s members, without the selection of those undertakings being justified by circumstances peculiar to them and capable of differentiating them from all other persons liable for the tax. The Commission maintains that, if the approach of the Court of First Instance were to be confirmed by the Court of Justice, the result would be that State aid decisions concerning measures of general application would be open to challenge by an indefinite number of persons and, in consequence, the condition relating to individual concern laid down in Article 230 EC would be rendered meaningless.
38. In so far as the plea under consideration is separate from the plea alleging that the Court of First Instance erred in its assessment of the AGL’s impact on the competitive position of BAA’s members, it seems that it ought, logically, to be construed as an assertion on the part of the Commission that, whenever the contested measure is of general application, the criterion established in Cofaz – that the aid measure must significantly affect the applicant’s market position – is not of itself sufficient for the purposes of assessing the admissibility of the action.
39. Let me say straightaway that, to my mind, there is nothing in the case-law of either the Court of Justice or the Court of First Instance to support that argument.
40. In the first place, the precedents referred to by the Commission do not seem to me to be of critical relevance. They relate, on the one hand, to actions brought by potential beneficiaries of measures declared to be incompatible with the common market and, on the other, to an action brought by an undertaking which was acting in its capacity as a competitor of potential beneficiaries of an authorised aid scheme. Those precedents therefore relate to situations which differed in law and in fact from the present case.
41. Nor does it seem to me that those precedents are open to the interpretation suggested by the Commission.
42. In Kwekerij van der Kooy and Others v Commission (12) and Arbeitsgemeinschaft Deutscher Luftfahrt-Unternehmen and Hapag-Lloyd v Commission , (13) the Court of Justice and the Court of First Instance, respectively, dismissed the actions brought before them upon finding that, in relation to the potential beneficiaries of the national provisions declared to be incompatible with the common market, the contested decisions constituted ‘measure[s] of general application covering situations which are determined objectively and entail legal effects for categories of persons envisaged in a general and abstract manner’, and, consequently, that those decisions affected the applicants solely in their objective capacity as potential beneficiaries of those provisions. (14) In Kahn Scheepvaart v Commission , (15) the Court of First Instance declared to be inadmissible an action brought by an undertaking which was a competitor of the potential beneficiaries of measures to be adopted in the context of a general aid scheme authorised by the Commission. In the grounds of that judgment, the Court of First Instance pointed out, inter alia, that the ‘adoption of the contested decision could have only a potential and indirect effect’ on the applicant’s competitive position, because, since the decision related to ‘the approval of a general aid scheme whose potential beneficiaries are defined only in a general and abstract manner, the existence of an actual beneficiary, and hence that of an active competitor of that beneficiary, presupposes the practical application of the aid scheme by the grant of individual aid’. (16)
43. Contrary to what the Commission appears to be maintaining, I do not consider it possible to infer from those judgments that, in principle, a decision as to whether a national measure of general application is compatible with Article 87 EC is not open to challenge by the beneficiaries of that measure or by competing undertakings. In my view, all that can be inferred from those judgments is that, in such circumstances, the status of beneficiary or competitor may not of itself be sufficient to establish the locus standi of the applicant undertaking, in which case, in order to establish its locus standi , that undertaking will have to demonstrate that it is in a situation which makes it possible to differentiate it from the general category of economic operators affected by that measure, distinguishing it individually in accordance with Plaumann .
44. I should then point out that the Court has expressly held, from Codorniu onwards, (17) that the fact that a Community act is of general application does not mean that it cannot be of direct and individual concern to some traders. Referring specifically to Codorniu , the Court recently upheld a judgment by which the Court of First Instance had held to be admissible an action brought by a number of Belgian coordination centres against a Commission decision finding the tax regime applicable to those centres to be incompatible with Community law. (18)
45. Accordingly, the fact that, as in the present case, a Commission decision appears to be a measure of general application in relation to certain persons – since it authorises a tax regime that is applicable to a category of traders envisaged in a general and abstract manner – does not prevent it from being of direct and individual concern to some of those traders by reason of certain attributes which are peculiar to them.
46. As we have seen above, since Cofaz , the Court has held that, as regards actions brought against decisions on State aid, the conditions laid down in the Plaumann case-law have to be regarded as satisfied where the competing undertaking demonstrates that the measure taken by means of the contested decision is liable significantly to affect (adversely) its position on the market. In the light of the foregoing considerations, it seems to me that that rule must apply both where the contested decision concerns individual aid and where it entails a finding in relation to a general aid scheme or other measure of general application. (19) Moreover, that approach is expressly confirmed in paragraph 70 of the judgment in ARE , (20) referred to above.
47. It should also be pointed out that the harm of which BAA complains does not derive from the fact that its members are required to pay the AGL, but from the competitive disadvantage they allegedly suffer by reason of the fact that some competing producers are exempt. Accordingly, BAA does not rely on the damage to the interests of its members in their capacity as persons liable to pay the AGL, but in their capacity as undertakings subject to competitive pressure from producers that are exempt from the AGL. In other words, what BAA is challenging is not the application of the levy as such, but the element of aid which, it alleges, is inherent in the manner in which the scope of the levy has been delimited.
48. In those circumstances, it seems to me that the risk at which the Commission expresses concern – that, if BAA’s action were held to be admissible, the result would be that a potentially unlimited number of persons would have the right to contest decisions relating to general fiscal measures, thus ‘depriv[ing] the concept “individually concerned” in [the fourth paragraph of] Article 230 [EC] of its useful meaning’ (21) – has to be ruled out.
49. The Commission contends, lastly, that the selection of the three BAA members whose competitive position the Court of First Instance took into account appears to be arbitrary and unjustified. It seems to me that, on the one hand, that argument restates the Commission’s contention, criticised above, that where the contested measure is of general application, it is not sufficient for the purposes of establishing the admissibility of the action for the applicant to show that its market position has been significantly affected and that, on the other hand, it anticipates the Commission’s second plea against the judgment under appeal, to the effect that the assessment made by the Court of First Instance of the AGL’s impact on the competitive position of BAA’s members is inaccurate and incomplete. In any event, in that connection it is sufficient to point out that, as correctly stated in paragraph 47 of the judgment under appeal, it has been consistently held that an action brought by an association acting on behalf of one or more of its members which would themselves have locus standi to bring an admissible action is itself admissible. (22) Accordingly, it is for BAA to establish the requisite individual locus standi of one or more of its members and for the Court of First Instance to ascertain whether such proof has been provided. In those circumstances, the question whether or not BAA was justified in its selection of the information to submit to the Court of First Instance for appraisal will, in the final analysis, depend solely on the outcome of that appraisal.
50. For all of the reasons set out above, I do not consider that the first plea on which the Commission relies in support of its cross-appeal can be upheld.
51. I shall therefore move on to consider the Commission’s second plea against the judgment under appeal.
3. The error allegedly made by the Court of First Instance in assessing whether the competitive position of BAA members has been significantly affected
52. The Commission first of all contends that, for the purposes of assessing whether the market position of BAA members had been significantly affected, the Court of First Instance confined itself – contrary to the case-law and, in particular, to Cofaz – to establishing the existence of a competitive relationship between, on the one hand, the three undertakings subject to the AGL and, on the other, the undertakings which are exempt.
53. As I have already explained, the current position according to the case-law is that where an action challenging a decision adopted on the basis of Article 88(3) EC is brought by a rival undertaking, acting in circumstances outside the scope of those to which the Cook and Matra case-law applies, it will be declared admissible if the applicant proves that the measure taken by means of the contested decision is liable significantly to affect (adversely) its position on the market. Consequently, in circumstances of that nature, an undertaking cannot simply rely on its status as a competitor of the undertaking in receipt of State aid but must additionally show that its de facto circumstances distinguish it in a similar way to the addressee of the decision. (23)
54. In paragraphs 55 to 63 of the judgment under appeal, the Court of First Instance determined, in the course of examining the Commission’s plea of inadmissibility, whether BAA had ‘explained, by reference to relevant criteria, why the AGL is liable to have a significant adverse effect on the position of one or more of its members on the market for aggregates’. (24) The Court of First Instance first pointed out that the purpose of the AGL was to transfer some of the demand for virgin aggregates to other products (which were exempt from the levy) and that the United Kingdom authorities anticipated that the AGL would enable demand for virgin aggregates to be reduced by an average of approximately 8 to 9% per year. Secondly, the Court of First Instance noted that some of BAA’s member undertakings – such as Torrington Stone, Sherburn Stone Co. Ltd and Cloburn Quarry – were in direct competition with the producers of exempted materials, which had become competitive as a result of the introduction of the AGL. Thirdly, in paragraphs 59 to 61 of the judgment under appeal, and in paragraph 65 in relation to the exemption for exports, the Court of First Instance reviewed the individual position of each of those three undertakings. On the basis of the evidence examined, the Court of First Instance rejected the Commission’s plea of inadmissibility.
55. It is clear from the grounds for the judgment on this point that the Court of First Instance did not merely establish in a general way the existence of a competitive relationship between the products covered by the AGL and exempted products, but based its finding concerning the adverse effect on the market position of BAA members on a range of evidence, based in part on the nature and purpose of the AGL and in part on information provided by BAA. The latter evidence in particular shows both the increased competitive pressure on BAA members – which are undertakings of limited size – as a result of the application of the AGL, as well as the ensuing stock management problems, and also the impact of the exemption for exports on the prices applied on the national market by undertakings active on the export market.
56. Thus, contrary to the Commission’s contention, the Court of First Instance did not simply establish the existence of a competitive relationship between the taxed products and the exempted products. Nor do the precedents to which the Commission refers appear to display sufficient points of similarity with the present case to allow comparison with the judgment under appeal. (25)
57. The Commission also disputes the adequacy of the analysis which the Court of First Instance made of the situation of Torrington Stone, on the ground that no details were provided of the consequences ‘in terms of price, market share or profitability of the undertaking’ of the increased competitive pressure on Torrington Stone as a result of the application of the AGL. Criticism of a similar nature is levelled at the Court of First Instance’s appraisal of the situation of Sherburn Stone.
58. It may be sufficient in that connection to recall that the Court has recently explained that the adverse effect on the competitive situation of an applicant undertaking, capable of distinguishing it in accordance with the Plaumann and Cofaz case-law, need not necessarily ‘be inferred from factors such as a significant decline in turnover, appreciable financial losses or a significant reduction in market share following the grant of the aid in question’, (26) but may also consist in ‘the loss of an opportunity to make a profit or a less favourable development than would have been the case without such aid’. (27)
59. Lastly, the Commission contends that both the undertakings in receipt of the alleged aid and their competitor undertakings are potentially unlimited in number, and that the effects of the AGL on competition are sectoral in scope. In those circumstances, the conclusions reached by the Court of First Instance in the judgment under appeal suggest that an undertaking may demonstrate that it is significantly affected by an aid measure – and thus be recognised as individually concerned, within the meaning of Article 230 EC, by the decision authorising that aid – even if its circumstances do not differ in any way from those of many other undertakings.
60. These arguments are the same as those put forward in support of the first plea and, in the context of the second plea, they should in my view be construed as asking the Court to apply more stringent criteria in assessing the adverse effect on the market position of the applicant undertaking where the measure in question is of general application.
61. To some extent, the case-law appears to endorse that approach. As was pointed out by Advocate General Jacobs in his Opinion in ARE, ‘the case-law, although not consistently, appears also to impose a stricter test for establishing an effect on the applicant’s competitive position when the aid scheme is of a general character’. (28) By way of example, Advocate General Jacobs referred, inter alia, to Kahn Scheepvaart . Paragraph 72 of the judgment in ARE could also be construed to that effect. (29)
62. Notwithstanding the foregoing, I consider that the approach which the Commission is proposing should not be adopted, particularly in relation to the present case.
63. Generally speaking, it must first be pointed out that in defining the conditions which an undertaking must fulfil in order to have locus standi to bring an action against a decision authorising the grant of aid to competing undertakings, Cofaz alleviated the burden of proof which Plaumann placed on an applicant who is not the addressee of the contested act. From Cofaz onwards, proof that the applicant undertaking’s competitive position is significantly affected by the aid measure has been sufficient to confer on that undertaking locus standi to bring an action against the decision authorising that measure, regardless of the fact that an indeterminate number of other competitors may invoke a similar adverse effect. (30)
64. As regards more specifically the present case, it must be distinguished from Kahn Scheepvaart and ARE . In Kahn Scheepvaart , the contested decision had approved an aid scheme which the Netherlands Government was to implement by granting individual forms of aid. In those circumstances, as the Court pointed out, it was not possible to identify either the actual recipients of the measures authorised by the Commission or, consequently, their competitors. The Court therefore concluded that the adverse effect of which the applicant undertakings complained was merely potential. Similarly, in the case which gave rise to the judgment in ARE , the land acquisition scheme at issue simply identified the categories of potential beneficiary. In consequence, it was impossible to determine either the identity or number of the future purchasers, or, consequently, either the identity or number of their competitors. In the present case, however, the national measure at issue, while of general application, is defined, in terms of its scope, in such a way that it is possible to distinguish both the undertakings which are liable for the levy and those which are exempt from it and, according to BAA, enjoy an advantage as a result.
65. Moreover, in my view, we should not disregard the fact that the national measure at issue is specifically designed to transfer some of the demand for aggregates from virgin aggregates to alternative products. Thus, it is a measure which is intended to have a direct effect on the structure of the market for the product in question, hence on the competitive position of undertakings active on that market. (31) In Lenzing , a judgment which has been referred to on a number of occasions, the Court held, inter alia, that the ‘seriousness [of the adverse effect on the applicant’s market position] may vary according to a large number of factors such as, in particular, the structure of the market concerned or the nature of the aid in question’. And it seems to me that, among the factors inherent in the nature of the aid, account must also be taken of the effects which the national authorities are seeking to achieve by adopting the measure, regardless of the sectoral nature of such effects.
66. On the basis of all of the foregoing, I consider that the second plea on which the Commission relies in support of its cross-appeal against the judgment under appeal should be rejected.
67. I therefore propose that the Court should dismiss the cross-appeal in its entirety.
4. Concluding observations on the cross-appeal
68. I have still to consider the argument put forward by BAA in response to the cross-appeal. In that context, BAA claims that the fact that, in support of its action before the Court of First Instance, it alleged that the Commission was in breach of its duty to initiate the formal investigation procedure under Article 88(2) EC was of itself sufficient, on the basis of the case-law in Cook and Matra , to render the whole of the action admissible, including the pleas calling into question the merits of the contested decision. (32) That argument becomes significant only if, contrary to my proposal, the Court should find that BAA does not satisfy the conditions for admissibility established in Plaumann and Cofaz . Indeed, in that event, the Court would have to assess whether and, if so, to what extent, the action could be regarded as admissible on the basis of the criteria established in the Cook and Matra case-law.
69. Analysis of that argument gives me the opportunity to make a number of brief observations of a wider-ranging nature. Let me make clear at the outset that it is not my intention to enter into the debate concerning the appropriateness of reviewing current case-law on the conditions for admissibility of actions brought against the decisions which the Commission adopts in the context of the procedure for monitoring State aid. (33) I shall therefore confine myself to a few brief points.
70. As mentioned above, on the basis of the case-law, the conditions for admissibility which must be satisfied by a person challenging a decision adopted on the basis of Article 88(3) EC, where the procedure under Article 88(2) EC has not been initiated, differ according to whether the action is designed to protect the procedural rights which the applicant enjoys on the basis of the latter provision or to call into question the ‘merits of the decision appraising the aid as such’. In the former case, it is sufficient to prove the status of ‘party concerned’ within the meaning of Article 88(2) EC; but, in the latter case, it will be necessary to demonstrate that the contested decision concerns the applicant individually within the meaning of Plaumann . Moreover, when acting solely in its capacity as a concerned party within the meaning of Article 88(2) EC, the applicant is required expressly to raise in the action a plea alleging infringement of the procedural guarantees recognised by that provision, or breach of the Commission’s duty to initiate the formal investigation procedure. Furthermore, the Court rules out practically any possibility of remedying the failure to raise such pleas by reframing pleas which have actually been raised.
71. To my mind, the reasons underlying that particularly complex and rather formalistic case-law can be traced back to the need to ensure that where an action is brought by a person who is simply relying on his status as a concerned party within the meaning of Article 88(2) EC, the Community judicature’s review of the contested decision does not go beyond what is needed to ensure that the procedural rights conferred by that provision are complied with. That would be the case where, rather than merely determining whether the conditions justifying non-initiation of the formal investigation procedure were satisfied – namely, the absence of serious difficulties in classifying the measure as aid (34) and/or in assessing its compatibility with the common market – the Community judicature were to establish the existence of State aid (or of individual elements of State aid which the Commission had found to be lacking), or to find that the conditions relied upon by the Commission in declaring the aid compatible with the Treaty were not satisfied. In such circumstances, the applicant would, in fact, secure not only initiation of the formal investigation procedure, where appropriate, but the additional result that the Commission would be bound by those findings on the part of the Community judicature (35) and the substance of the decision to be adopted on conclusion of that procedure would, in part at least, be predetermined; that is to say, the applicant would have brought about the predetermination of a decision which it would not have been entitled to challenge solely by virtue of its status as a concerned party within the meaning of Article 88(2) EC.
72. However, it does not seem to me that, in order to meet that need for the scope of the judicial review of the contested act to be appropriate to the locus standi claimed by the applicant, it is necessary to compel the latter to comply with specific formal requirements when drafting the grounds for its action, failing which that action will be found to be inadmissible. It will be sufficient, for that purpose, to treat grounds that formally seek to establish the existence of State aid or the incompatibility of that aid with the common market as being limited, in substance, to claiming that there are serious difficulties concerning the classification of the measure or the assessment of compatibility, or, at least, to challenging the adequacy – for the purposes of ruling out such difficulties – of the grounds on which the contested decision is based.
73. In any event, it is my belief that, in the final analysis, the complexity of the case-law analysed above and the problems involved in actually applying it stem from the particular nature of decisions adopted under Article 88(3) EC.
74. Those decisions, which the Commission adopts without initiating the formal investigation procedure, are acts which are brought into being on conclusion of a brief review, which is conducted within a short time frame and, what is more, in the context of a dialogue exclusive to the Commission and the Member State concerned. The legislature’s intention was to enable the Commission, by means of such acts, to avoid considerable time being taken up by a full investigation in cases where it was already prima facie evident that there was no aid or that the aid was compatible with the common market.
75. Given the nature of those decisions, I wonder whether it would not be more appropriate for the review of such acts by the Community judicature to be confined in every case – hence regardless of the basis for the applicant’s locus standi – to verifying whether the conditions justifying non-initiation of the formal investigation procedure are satisfied , or to establishing that there are no serious doubts that the measure does not constitute aid or is, in any event, compatible with the common market. A ruling on the substantive issues – that is to say, a finding as to whether aid does or does not exist or whether aid is compatible with the common market – would thus, if the decision were to be annulled, be deferred until the examination of any action brought against the final decision subsequently adopted by the Commission on conclusion of the formal investigation procedure. That would make it possible to avert the possibility that, in making a definitive ruling on matters regarding which the Commission has really made only a prima facie finding on conclusion of a preliminary examination, the Community judicature exceeds the limits of the normal review of legality which it must carry out in the context of applications for annulment.
76. Turning now to BAA’s argument, referred to in point 68 above, it follows from the foregoing remarks that, whereas BAA is able to rely solely on its status as a concerned party within the meaning of Article 88(2) EC, the mere fact that the grounds relied upon in support of its action before the Court of First Instance include an allegation that the Commission was in breach of its duty to initiate the formal investigation procedure does not, contrary to BAA’s assertion, give it the right to obtain a ruling from the Court of First Instance on the classification of the AGL on the basis of Article 87(1) EC.
B – The appeal proper
77. According to the contested decision, the AGL cannot be classified as State aid since it did not entail any selective advantage. By the first, third and fourth pleas in law relied upon in support of the appeal, the judgment under appeal is impugned on the ground that it confirmed the Commission’s assessment on that point. I shall consider those pleas below (see Sections 2, 3 and 4). But before I embark on that analysis, I shall outline the case-law on selectivity (see Section 1 below).
78. The second plea in law in support of the appeal concerns the extent of the Court of First Instance’s review of the contested decision and will be analysed under Section 5 below.
79. Lastly, under Section 6 below, I shall consider the fifth and sixth pleas in law, which impute to the Court of First Instance errors of law in its assessment, respectively, of the Commission’s duty to initiate the formal investigation procedure and of the question whether the contested decision contains an adequate statement of reasons.
1. Outline of the case-law on the selectivity of aid
80. Under Article 87(1) EC, for a measure to be classified as aid, it must favour certain undertakings or the production of certain goods it must, therefore, be selective.
81. In accordance with the condition of selectivity, ‘general’ measures – which are designed to support, not specific activities or undertakings, but all economic operators active in the State’s territory – are excluded from the application of the provisions on State aid. In that connection, the case-law has made it clear that State action in favour of an indefinite number of beneficiaries, identified on the basis of a number of objective criteria, must be regarded as a system of aid constituting a selective measure if, owing to the criteria governing its application, it procures an advantage for certain undertakings or the production of certain goods, to the exclusion of others. (36) On the other hand, however, the case-law has also made it clear that even measures which seem to be general measures, in that they are not limited either to a particular sector or to a particular territory and are not applied to a restricted category of undertakings, may be caught by the prohibition laid down in Article 87(1) EC if their implementation is left to the discretion of the national authorities as regards, in particular, the choice of recipients, the amount and the conditions of the financial assistance. The Court has also held that aid may be selective even where it concerns an entire economic sector. (37)
82. More generally, it is clear from the case-law that the question whether the selectivity condition is satisfied must be assessed on a case-by-case basis, in order to ascertain whether or not, in the light of its nature, scope, method of implementation and effects, the measure in question involves advantages accruing exclusively to certain undertakings or certain sectors . (38) If it is established that an advantage of that nature exists, even action in the form of refraining from imposing a new tax on certain economic operators may constitute aid within the meaning of Article 87(1) EC. (39)
83. With particular reference to State measures of a fiscal nature, the case-law shows, however, that even measures which are selective, in that they differentiate between undertakings, may escape being classified as aid, if that differentiation is justified by the nature or structure of the tax regime of which they form part. (40) It follows, according to the Court, that, in order to determine whether or not a measure is selective for the purposes of applying Article 87(1) EC, ‘it is appropriate to examine whether, within the context of a particular legal system, that measure constitutes an advantage for certain undertakings by comparison with others which are in a comparable legal and factual situation’. (41)
84. It is in the light of the case-law referred to above, in particular, that it is necessary to determine whether the Court of First Instance erred in law in the various ways imputed to it by BAA in the context of its first, third and fourth pleas in law.
2. The first plea, alleging infringement of Article 87(1) EC
85. By its first plea, BAA claims that the judgment under appeal is vitiated by a number of errors of law in the application of Article 87(1) EC. BAA puts forward three main submissions in that regard.
86. First, the Court of First Instance is alleged to have applied a concept of aid that was not objective. BAA argues that, according to the case-law, the fact that a fiscal measure pursues general policy objectives does not preclude it from being classified as State aid. Rather, that fact is relevant only for the purposes of determining whether the nature and general scheme of the measure justify the distinctions drawn by the legislature in defining its scope. In the judgment under appeal, however, the Court of First Instance is alleged to have adopted a different approach, by attributing to the Member States a broad discretion in defining the scope of fiscal measures designed to achieve objectives related to environmental protection, and by ruling out the possibility that such measures can be selective, even where they differentiate between undertakings which are in comparable situations, in a manner that is not justified by environmental objectives.
87. Secondly, the selectivity assessment made by the Court of First Instance is alleged to be at odds with the assessment carried out in Adria-Wien Pipeline . (42)
88. Lastly, the Court of First Instance is alleged to have ruled out the possibility that the AGL was selective, without benefit of a precise definition of the sector to which it applies. That third main submission must, in my view, be declared inadmissible, since it seeks, in essence, to call into question the findings of fact made by the Court of First Instance.
89. I shall therefore proceed now to consider only the substance of the first and second submissions, which, to my mind, should be examined together.
90. Paragraphs 114 to 118, 120 and 121 of the judgment under appeal are of particular relevance for those purposes. The various steps in the line of reasoning followed by the Court of First Instance may be summarised as follows.
91. First, the Court of First Instance defines the environmental levy as an ‘ autonomous fiscal measure ’ (43) which is not part of a general tax regime and ‘is characterised by its environmental objective and its specific tax base’. (44) It follows from that definition that the fact that the environmental levy does not apply to activities similar to those which are taxed and which have a comparable impact on the environment cannot be viewed in the same way as a tax exemption which derogates from the system of burdens normally imposed on undertakings. (45)
92. The Court of First Instance then points out that in exercising their powers in relation to environmental policy, ‘it is open to the Member States ... to introduce sectoral environmental levies in order to attain those environmental objectives’. In particular, the Member States are free, ‘in balancing the various interests involved, to set their priorities as regards the protection of the environment and, as a result, to determine which goods or services they are to decide to subject to an environmental levy. (46)
93. In paragraph 117 of the judgment under appeal, the Court of First Instance concludes that, in that legal framework, ‘it is for the Commission, when assessing an environmental levy for the purposes of the Community rules on State aid, to take account of the environmental protection requirements referred to in Article 6 EC’. According to the Court of First Instance, that conclusion is justified in the light of Article 6 EC, which ‘provides that those requirements are to be integrated into the definition and implementation of, inter alia, arrangements which ensure that competition is not distorted within the internal market’.
94. In paragraphs 120 and 121 of the judgment under appeal, the Court of First Instance rejects BAA’s argument based on Adria-Wien Pipeline . It points out that, in that judgment, ‘the Court of Justice was called upon to consider, not the definition of the material scope of an environmental levy, as in the present case, but the partial exemption from payment of such a levy ... which was granted only to undertakings which were involved in the manufacture of goods’. (47) The Court of First Instance goes on to point out that ‘the distinction at issue thus did not relate to the type of product subject to the environmental levy in question, but to industrial consumers, depending on whether or not they carried on their activity in the primary and secondary sectors of the national economy’.
95. The approach taken by the Court of First Instance, as it emerges from the paragraphs of the judgment under appeal which I have summarised above, is highly innovative as compared with the Community case-law on the application of the selectivity criterion and, more generally, of the concept of aid within the meaning of Article 87(1) EC.
96. It is, in fact, settled case-law that neither the fiscal nature nor the economic or social aim of the State measure at issue, nor the environmental protection objectives which it pursues (48) are sufficient to exclude it from the ambit of the prohibition laid down in Article 87 EC. In fact, as the Community judicature has pointed out on a number of occasions, Article 87 EC does not make a distinction between such State measures according to their aim or the reasons for their introduction, but defines them solely in terms of their effects. (49)
97. The objectives pursued by a measure therefore fall to be considered only at a point subsequent to its classification as State aid, that is to say, in the course of assessing whether it is compatible with the common market. While in some cases the Community judicature has taken into account, at the classification stage, the intentions of the national legislature and the objectives pursued by the measure adopted, it has done so solely for the purpose of determining whether the conditions establishing the existence of aid were present, (50) and not in order to exclude from the outset the possibility that the measure was covered by Article 87(1) EC.
98. By concluding without further examination that it is justifiable, because of the AGL’s environmental objectives, to accord different treatment to operators (or production sources) whose situations in terms of those objectives are comparable, the Court of First Instance is clearly seeking to move beyond that case-law. Such an approach actually excludes a priori the possibility that the non-imposition of the environmental levy on some operators may constitute a selective advantage under Article 87 EC, irrespective of any consideration of the competitive relationship that may exist between those operators and the operators liable to pay the levy, hence independently of an assessment of the effects of the measure in question. (51)
99. Moreover, in the part of the judgment under appeal which examines BAA’s allegations of inconsistencies in the definition of the AGL’s scope, (52) conclusions are reached which would not have been permissible if that examination had been based on the criteria of the nature and general structure of the tax regime at issue, in accordance with the case-law referred to in point 83 above. (53) In effect, starting from the premiss that the Member States are free, when defining the scope of an environmental levy, to balance the various interests at stake, (54) the Court of First Instance ultimately accepts that the existence of possible inconsistencies or differences in treatment may be justified even if based on objectives unconnected with the protection of the environment and, consequently, unrelated to the internal logic of the measure in question. (55)
100. Lastly, as regards the passage in the grounds of the judgment under appeal which rejects BAA’s argument based on Adria-Wien Pipeline , the analysis of the Court of First Instance focuses on the formal aspects of the measure in question, such as the legislative technique used by the national authorities. In terms of the impact on competition, however, there is no great difference between, on the one hand, the imposition of a general tax with an exemption for certain beneficiaries and, on the other hand, the imposition of a tax on certain taxable persons to the exclusion of others who are in a comparable situation. Here, too, it seems to me that the judgment under appeal departs from an approach focused on an analysis of the effects of the measure.
101. It does not seem to me that the conclusions of the Court of First Instance in the judgment under appeal, although limited to State measures which involve the imposition of burdens payable by a certain sector for environmental purposes, should be upheld by the Court of Justice.
102. Neither the competence enjoyed by the Member States in matters relating to taxation or the environment, nor the principle laid down by Article 6 EC of the integration of environmental protection requirements into the definition and implementation of Community policies, justifies the wholesale removal of public measures that could distort competition from the ambit of the supervisory power conferred on the Commission by the Treaty rules on State aid. In particular, it seems to me that compliance with Article 6 EC does not require the environmental objectives of a measure to be taken into account for the purposes of its classification on the basis of Article 87(1) EC, since the requirement that those objectives be integrated into the Community monitoring of State aid can easily be satisfied by taking them properly into consideration when it comes to assessing the measure’s compatibility with the common market on the basis of Article 87(3) EC.
103. For the reasons set out above, I consider that the first plea in support of the appeal, alleging infringement of Article 87(1) EC, is well founded and that the judgment under appeal should be set aside in so far as it held that the scope of the AGL could be justified solely in terms of the environmental objectives pursued by that levy.
3. The third plea, alleging errors of law in the assessment of the nature and the general scheme of the AGL
104. BAA puts forward three main submissions in support of the third plea.
105. First, BAA criticises the Court of First Instance for holding that the fact that the AGL is levied on a number of virgin materials for which there are no substitutes is justified on the basis of the ‘polluter pays’ principle, notwithstanding the finding in the contested decision that non-substitutability was a suitable reason for excluding from the scope of the levy certain uses of virgin materials which would otherwise have been subject to it. BAA concludes from this that, in paragraphs 135 and 136 of the judgment under appeal, the Court of First Instance substituted its own reasons for those of the Commission. According to BAA, the Court of First Instance also applied the ‘polluter pays’ principle ‘in a selective way’ and failed to give adequate reasons for its own conclusions. That submission seems to me to be entirely admissible, since – contrary to what the Commission and the United Kingdom maintain – it raises points of law.
106. In my view, BAA is not mistaken in considering that the ‘polluter pays’ principle plays no part in the description of the AGL’s objectives as set out in the contested decision, which makes express reference only to maximising the use of recycled aggregates or other substitute materials and promoting the rational use of virgin aggregates.
107. In DIR International Film (cited by BAA) after pointing out that, in reviewing the legality of acts pursuant to Article 230 EC, the Community judicature may not substitute its own reasons for those of the author of the contested act, the Court stated that ‘although, in proceedings for annulment, the Court of First Instance may be led to interpret the reasons given for the contested measure in a manner which differs from that of its author, and even, in certain circumstances, to reject the latter’s formal statement of reasons, it cannot do so where there is no material factor to justify such a course of action ’. (56) In the present case, contrary to the view expressed by the Court of First Instance in paragraph 124 of the judgment under appeal, it does not seem to me that point 31 of the contested decision – in which the Commission states that ‘[t]he environmental costs of aggregate extraction that the United Kingdom seeks to address through the AGL include noise, dust, damage to biodiversity and visual amenity’ – can be construed as a reference, albeit implicit, to the ‘polluter pays’ principle.
108. Even if the above considerations suggest that the Court of First Instance erred in law by substituting its own reasons for those of the contested decision, I none the less consider that the judgment under appeal ought not to be set aside on that point, because it is possible for the Court of Justice to substitute the reasons on which the Court of First Instance based its decision. As the Commission pointed out – correctly, in my view – BAA’s allegations concerning the lack of consistency in the way the scope of the AGL has been delimited, which it put forward at first instance and which the Court of First Instance rejected in paragraphs 135 and 136 of the judgment under appeal, were based on an incorrect premiss: that is to say, BAA predicated its allegations on the view that, in the contested decision, the Commission took the lack of substitute materials as the basis for its finding that the exclusion of virgin materials used for purposes other than as aggregates from the scope of the AGL was justified. Rather, it is clear from the grounds of the contested decision (57) that the exclusion of those materials was regarded as consistent with the sectoral scope of the AGL and with the intention of the United Kingdom legislature, which was to levy a tax exclusively on aggregates.
109. The remaining allegations made by BAA in the context of this first submission must, in my view, be rejected. The judgment under appeal appears to be adequately reasoned as regards the issue raised by BAA. Moreover, for the reasons already stated in point 108 above, and contrary to the view expressed by BAA, I do not consider it possible to ascribe to the Court of First Instance a ‘selective application’ of the ‘polluter pays’ principle.
110. Secondly, BAA challenges a number of passages in the judgment under appeal in which the Court of First Instance rejects the claims that taxing certain products derived from the extraction of materials which are not themselves taxed is incompatible with the AGL’s objectives. That criticism relates specifically to paragraphs 112 and 137 of the judgment under appeal.
111. In paragraph 112 of that judgment, the Court of First Instance interprets certain concepts used in the contested decision – in point 29, in particular – and concludes that, both in point 29 and throughout the contested decision, the Commission ‘used the expression “primary aggregates” essentially in order to designate aggregates subject to the AGL, and the expression “secondary aggregates” essentially in order to refer to the exempted aggregates specifically listed in the Act’. According to BAA, that construction of the contested decision is vitiated by an error of interpretation.
112. In point 29 of the contested decision, the Commission states, inter alia, that ‘the AGL will be levied only on virgin aggregate. It will not be levied on aggregates extracted as a by-product or waste from other processes (secondary aggregates), nor will it be levied on recycled aggregates’.
113. I must confess that I find it difficult to attribute to the term ‘secondary aggregates’, used in point 29 of the contested decision, a meaning other than the meaning which follows from the definition that precedes it: ‘aggregates extracted as a by-product or waste from other processes’. Consequently, it does not seem to me to be possible to agree with the Court of First Instance that, in point 29 of the contested decision, the Commission was merely stating that ‘the AGL will not be levied on secondary products or waste arising from primary extraction, when they are exempted by the Act, as amended’.
114. Moreover, as BAA – in my view, rightly – points out, the construction placed by the Court of First Instance on the term ‘secondary aggregates’ appears to be contradicted by point 32 of the contested decision – which plays a key role in the scheme of that decision, since it defines the structure and scope of the AGL – in which the Commission states that ‘the structure and the scope of the tax reflect the clear distinction between the extraction of virgin aggregates, bearing with it undesirable environmental consequences, and the production of secondary or recycled aggregates, which makes an important contribution to the treatment of rock, gravel and sand incidentally arising from excavations or from other works or treatments lawfully carried out for different purposes’.
115. I therefore consider that in paragraph 112 of the judgment under appeal, the Court of First Instance misconstrued the contested decision, which must be regarded as an error of law.
116. BAA also challenges paragraph 137 of the judgment under appeal, in which the Court of First Instance held to be justified the fact that a number of products derived from the extraction of untaxed materials are subject to the AGL. BAA maintains that in that regard the Court of First Instance substituted its own reasons for those of the contested decision; that it adopted reasons vitiated by errors of assessment; and that it distorted some of the evidence.
117. In that connection, it should be pointed out that in justifying the taxation of the by-products in question, the Court of First Instance did not simply refer to the ‘polluter pays’ principle; it also referred to the objective of achieving more rational extraction and treatment of aggregates, which is mentioned in the contested decision as one of the AGL’s objectives. Accordingly, I do not accept that, in this regard, the Court of First Instance can be accused of substituting its own reasons for those of the contested decision. As for the remaining claims, I need only point out that in so far as BAA disputes the factors specified by the Court of First Instance in paragraph 137 of the judgment under appeal (the impossibility of reducing the quantity of those by-products, the price difference and the United Kingdom’s letter of 19 February 2002), it is in essence seeking to obtain from the Court of Justice a reappraisal of the assessments of fact in the judgment under appeal and, on the other hand, that BAA has failed to prove that the Court of First Instance distorted the evidence before it.
118. On the basis of the foregoing, I consider that the second main submission in support of the third plea must be upheld in so far as it imputes to the Court of First Instance an error of interpretation in relation to the contested decision. For the rest, I consider that it must be rejected as being in part unfounded and in part inadmissible.
119. Thirdly, BAA submits that the Court of First Instance erred in considering that the non-taxation of a number of virgin aggregates – such as slate, shale and clay – was justified. In that regard, according to BAA, the Court of First Instance substituted its own reasons for those of the contested decision, by stating in paragraphs 130, 131, 133 and 134 of the judgment under appeal that that exemption was designed to encourage the use of such aggregates as substitutes for the virgin aggregates subject to the AGL. BAA maintains that, as well as being incorrect, the justification accepted by the Court of First Instance owes its origins to a distortion of the evidence submitted in the course of the proceedings (in particular, the United Kingdom’s letter of 19 February 2002); insufficient reasons are given for that justification; and it was accepted in breach of the rights of the defence, since BAA was not given an opportunity during the proceedings at first instance to present its own observations on that issue.
120. On that point, it is first necessary to recall that, in paragraph 130 of the judgment under appeal, the Court of First Instance pointed out that the exemption from the AGL of a number of materials – such as shale and low-quality slate, clay waste and ball clay – ‘allows their use as substitutes for those virgin aggregates subject to the environmental levy and may, as a result, contribute to a rationalisation of the extraction and use of the latter’. As the Commission rightly pointed out, the Court of First Instance based that conclusion on the finding of fact – not open to verification by the Court of Justice – that such materials have been ‘until now rarely used as aggregates by reason of their high transport costs’.
121. It should also be recalled that rationalising the extraction and use of virgin aggregates is one of the objectives of the AGL to which the contested decision refers. (58) Thus, contrary to BAA’s submission, the Court of First Instance did not in that regard substitute its own reasons for those of the contested decision.
122. Nor do I regard as well founded BAA’s assertion that the way in which the nature and general scheme of the AGL are defined in the contested decision is incompatible with an exemption designed to encourage the use of virgin materials, hitherto rarely used as aggregates, in place of other materials which have traditionally been used as aggregates. As the Court of First Instance concluded – in my view, correctly – that exemption may in fact contribute to the objective, mentioned in the contested decision, of rationalising the use of aggregates.
123. Lastly, as regards paragraph 131 of the judgment under appeal, it seems to me that the reference therein to the content of the United Kingdom’s letter of 19 February 2002 does not lend itself to the construction placed upon it by BAA , and that that paragraph in no way distorts the content of that letter.
124. I therefore consider that the third submission in support of the third plea must be rejected as unfounded.
125. On the basis of all of the foregoing considerations, I consider that the third plea in law in support of the appeal is well founded so far as concerns the submissions referred to in points 108 and 118 above, but otherwise either inadmissible or unfounded.
4. The fourth plea, alleging an error of law in the assessment of the exemption for exports
126. BAA maintains that in assessing whether the exemption of exports from the scope of the AGL constituted aid, the Court of First Instance permitted the Commission and the Member State concerned retroactively to improve upon the reasons given for the contested decision; infringed Articles 91 EC and 92 EC by classifying the AGL as an indirect tax; and failed to give adequate reasons for that classification.
127. In paragraph 148 of the judgment under appeal, the Court of First Instance pointed out that in the contested decision (point 33), the Commission justified the exemption for exports ‘by the fact that the United Kingdom authorities have no control over the use of aggregates outside their jurisdiction’. In paragraph 149, the Court of First Instance noted that, in the course of the proceedings before it, ‘the Commission and the intervener [had] amplified that reasoning’, the Commission by emphasising that the AGL constituted an indirect tax on consumption governed by the principle of taxation in the State of destination, and the United Kingdom by stating that the exemption was permitted under Article 91 EC. According to the Court of First Instance, that reasoning, which is based on the notion that the AGL is an indirect tax, had to be taken into consideration ‘in so far as ... it reflects the reasoning set out by the Commission in the contested decision and therefore does not fall to be considered as additional reasoning put forward after the decision was adopted’.
128. In paragraph 151 of the judgment under appeal, the Court of First Instance then noted that the AGL ‘applies to the commercial exploitation of aggregates, and is thus charged against products and not on the income of producers’, and, consequently, constitutes an indirect tax. In paragraph 153, the Court of First Instance concluded that in the case before it, ‘the exemption for exports cannot therefore be considered to confer a selective advantage on exporters since it is justified by the nature of the AGL as an indirect tax’. According to the Court of First Instance, in fact, it was open to the Member State concerned ‘to grant priority to considerations linked to the structure of the tax scheme concerned over the environmental objectives pursued’.
129. First of all, it seems to me to be necessary to reject BAA’s argument that the AGL must be regarded as a direct tax – and as such excluded from the scope of Article 91 EC – since it applies to the extraction process and not to the product as such. In fact, as explained elsewhere in the judgment under appeal (see paragraph 136), the AGL is set at a specific amount per tonne of taxable product commercially exploited in the national territory.
130. The general prohibition, laid down in Article 92 EC, on granting remissions or repayments in respect of exports in the case of direct taxes is justified in so far as the latter are not reflected in production costs and, consequently, do not have an immediate impact on the price of the product; for that reason, it would be difficult to assess whether any remission or repayment granted in respect of exports corresponded to the domestic tax on that product or constituted a fiscal subsidy for exports. That problem does not arise, however, in the case of a tax which, like the AGL, is levied on the occasion of the product’s commercial exploitation. The fact – pointed out by the Court of First Instance in paragraph 136 of the judgment under appeal – that the amount of the levy corresponds approximately to the environmental costs linked to the extraction of the taxable products does not seem to me to be relevant, and does not alter the fact that, since it is applied to a fixed quantity of the product on the occasion of its commercial exploitation, that amount is liable to be directly reflected in the price of the product.
131. I therefore consider that the submission in support of the present plea, alleging infringement of Articles 91 EC and 92 EC, must be rejected as unfounded.
132. Similarly unfounded, in my view, is the submission that the judgment under appeal is not sufficiently reasoned. In paragraph 151 of that judgment, the Court of First Instance in fact explains, succinctly but clearly, the reasons why it considers that the AGL must be classified as an indirect tax, that is to say, because it ‘applies to the commercial exploitation of aggregates, and is thus charged against products and not on the income of producers’.
133. It remains for me to consider the submission concerning the retroactive improvement of the reasons given for the contested decision. That submission seems to me to be well founded.
134. In point 33 of the contested decision, the Commission states that the exemption for exports ‘is justified by the fact that aggregate in the United Kingdom may be exempted if it is used for exempt processes. Since the United Kingdom authorities have no control over the use of aggregate outside their jurisdiction, the exemption for exports is necessary in order to provide legal certainty to aggregate exporters and to avoid imposing an unequal treatment on exports of aggregate that would otherwise qualify for an exemption within the United Kingdom’.
135. That argument refers to the difference in treatment that would arise, were exports not exempt, between aggregates commercially exploited on the national market , which are not taxable if used for certain purposes, and exported aggregates , which would in any event be taxed, even if they were intended for use in exempt production processes. No reference is made, however, to the need to avoid creating a disadvantage for exported national aggregates as compared with aggregates commercially exploited in the State of destination , or to the objective of avoiding double taxation. In those circumstances, I find it difficult, candidly speaking, to identify in point 33 of the contested decision (as the Court of First Instance did in paragraph 150 of the judgment under appeal) even an implied reference to the rules laid down in Article 91 EC. In my view, rather than allowing the reasons for the contested decision to be amplified and improved upon retroactively, the Court of First Instance actually allowed the Commission to set out, on that point, reasons which are substantially different from those given in the contested decision.
136. For the reasons set out above, I consider that the fourth plea in law in support of the appeal should be upheld in so far as it alleges that the Court of First Instance erred in law in holding that the justification advanced by the Commission in the course of the proceedings at first instance, in relation to the exemption for exports, reflected the reasoning set out in the contested decision and, accordingly, did not constitute additional reasoning put forward after that decision was adopted.
5. The second plea in law, concerning the extent of the review carried out by the Court of First Instance
137. BAA maintains that the Court of First Instance erred in law by carrying out only a review of the assessments contained in the contested decision, rather than the comprehensive review which the case-law requires to be exercised in cases where it is necessary to verify that the Commission has properly applied the concept of aid within the meaning of Article 87(1) EC.
138. It is settled case-law that in an action for the annulment of a decision which the Commission has adopted under the procedure for monitoring State aid, the Community judicature must in principle carry out a comprehensive review when considering the classification of a State measure in order to establish whether it is caught by the prohibition under Article 87(1) EC. (59) However, the Community judicature carries out a limited review – confined to verifying that the rules governing the procedure and the statement of reasons have been complied with, that the facts are materially accurate and that there has been no manifest error of assessment and no misuse of powers – when it assesses the compatibility of the measure with the common market in the light of Article 87(3) EC, a provision which leaves a broad discretion to the Commission. (60)
139. As the Commission has rightly pointed out, however, even where the Community judicature is called upon to verify the proper application of the concept of aid, the scope of its review may be limited owing to the technical or complex nature of the factual assessments made in the contested act. (61) Thus, for example, case-law recognises that the Community judicature carries out only a limited review of the complex economic assessments made by the Commission when applying the private investor principle. (62)
140. In paragraph 118 of the judgment under appeal, in setting out the parameters for its review of the contested decision, the Court of First Instance states that, having regard to the broad ‘discretion which the Commission has in the application of Article 88(3) EC’, it must carry out only a limited review. A number of the subsequent paragraphs in the judgment confirm that the Court of First Instance followed such an approach (paragraphs 134, 139 and 171, in particular).
141. The Commission and the United Kingdom contend that, in paragraph 118 of the judgment under appeal, the Court of First Instance is actually referring to the scope of the review carried out by the Community judicature of decisions under Article 88(3) EC, and not the review which it carries out when assessing whether the Commission has properly applied the concept of aid for the purposes of Article 87(1) EC.
142. The argument contended for by the Commission and the United Kingdom does not seem to me to be persuasive.
143. First, it is clear from the case-law that if, on conclusion of the preliminary examination procedure, the Commission still harbours serious doubts as to whether the measure is in the nature of aid or whether it is compatible with the Treaty, it is under a duty to initiate the formal investigation procedure. (63)
144. Secondly, decisions under Article 88(3) EC include, as well as the procedural element, an examination of the measure in question in the light of Article 87(1) EC and, if that measure is classified as aid, in the light of Article 87(3) EC. I see no reason why the Community judicature’s review of the assessments made in the course of such an examination should not follow the rule, referred to on several occasions above, that when it is necessary to verify whether the Commission has properly applied the concept of aid, that review should normally take the form of a comprehensive review, whereas when the compatibility of the measure with the Treaty is being assessed, the review should be limited, in view of the broad discretion enjoyed by the Commission for the purposes of applying Article 87(3) EC. (64) Moreover, unlike BAA, I do not consider that a limited review is incompatible with the nature of decisions under Article 88(3) EC, as a limited review of that kind is justified by the margin of discretion that the Commission enjoys in relation to the assessments which it is required to make, quite apart from the procedural context in which such assessments fall to be made.
145. The present case does not seem to me to concern a situation in which the extreme technicality or particular complexity of assessments made by the Commission (65) justifies attributing to the Commission a margin of discretion and, accordingly, limiting the scope of the judicial review. Since that review involves verifying the proper application of the concept of aid, it must be comprehensive ‘in principle to the fullest possible extent’. (66) I therefore believe that, in paragraph 118 of the judgment under appeal, the Court of First Instance incorrectly defined the scope of the review to be carried out in relation to the contested decision and thus erred in law on that point.
146. The United Kingdom and the Commission further contend that the grounds relied upon by BAA before the Court of First Instance sought to demonstrate that the contested decision was vitiated by a series of manifest errors of assessment. In finding that there were no such errors, the Court of First Instance nevertheless thus adhered to the view advanced by BAA concerning the appropriate scope of the review. It should be pointed out in that connection that the fact that BAA described as manifest the errors which, in its view, vitiated the contested decision does not imply that its intention was to refrain from invoking errors that existed but could not be described as manifest, or that it was asking the Court of First Instance to review the decision less thoroughly than would normally have been the case. Furthermore, if the review of the contested decision was confined to an examination of the manifest errors of assessment simply because of the need to go no further than BAA had requested, the Court of First Instance would not have needed to make general statements of the kind set out in paragraph 118 of the judgment under appeal.
147. It follows from all of the above considerations that the Court of First Instance erred in law in paragraph 118 of the judgment under appeal by incorrectly defining the nature of the review that fell to be carried out in relation to the contested decision.
148. Such an error, which could undermine the entire assessment of the merits of the contested decision, means that the judgment under appeal must be set aside, notwithstanding the fact, pointed out by the United Kingdom, that in some paragraphs of that judgment, the Court of First Instance seems in fact to have gone beyond the bounds of a strictly limited review.
149. For the reasons set out above, I consider that the second plea in law, concerning the scope of the judicial review carried out by the Court of First Instance, must be upheld.
6. The fifth and sixth pleas in law, alleging errors of law in the assessment of the Commission’s duty to initiate the formal investigation procedure and the adequacy of the statement of reasons for the contested decision
150. The arguments put forward by BAA in support of the fifth and sixth pleas are, in my view, clearly insufficient to prove that the alleged defects exist. On the one hand, the fact that the reasoning of the judgment under appeal is more detailed than that of the contested decision and that the Court of First Instance has made assessments that differ from those made by the Commission (although they reach the same conclusions), even if relevant, is not in any event sufficient to show that the Court of First Instance erred in concluding that the Commission was not under a duty to initiate the formal investigation procedure. Furthermore, the fact that, in relation to various issues to which the contested decision alludes, the Court of First Instance set out more detailed reasons – and, at certain points, reasons radically different from those adopted in the decision – is clearly immaterial for the purposes of demonstrating that the Court of First Instance erred in concluding, in the light of reasons adopted in the contested decision, that the latter actually contained an adequate statement of the reasons.
151. The fifth and sixth pleas in law must therefore be rejected.
V – Conclusion
152. In the light of all of the foregoing, I therefore propose that the Court should set aside the judgment under appeal.
153. Furthermore, in view of the implications of setting aside the judgment under appeal, as proposed, I consider it appropriate that the Court refer the case back to the Court of First Instance, in accordance with the first paragraph of Article 61 of the Statute of the Court of Justice, for re-examination of the action at first instance, and reserve the decision on the costs of the appeal proceedings.
(1) .
(2) – Case T‑210/02 British Aggregates Association v Commission [2006] ECR II‑2789.
(3) – C (2002) 1478 final, concerning State aid Case N 863/01 – United Kingdom/Tax on aggregates.
(4) – Case 25/62 [1963] ECR 95.
(5) – Case 169/84 Cofaz and Others v Commission [1986] ECR 391.
(6) – Regulation No 17 of the Council of 6 February 1962: First Regulation implementing Articles 85 and 86 of the Treaty (OJ, English Special Edition 1959-1962, p. 87). In that case-law, the Court had established the principle according to which, if a regulation accords the applicant undertakings procedural guarantees allowing them to ask the Commission to establish that Community law has been infringed, those undertakings must be able to bring an action to protect their legitimate interests. Although implicitly recognising that the Treaty provisions on State aid do not accord procedural guarantees comparable to those for which Regulation No 17 provides, the Court none the less held that ‘Article [88(2) EC] recognises in general terms that the undertakings concerned are entitled to submit their comments to the Commission’.
(7) – Paragraph 25.
(8) – Case C‑225/91 Matra v Commission [1993] ECR I‑3203.
(9) – Case C‑198/91 Cook v Commission [1993] ECR I‑2487.
(10) – Case C‑78/03 P [2005] ECR I‑10737.
(11) – Paragraph 54 of the judgment under appeal.
(12) – Joined Cases 67/85, 68/85 and 70/85 [1988] ECR 219.
(13) – Case T‑86/96 [1999] ECR II‑179.
(14) – See paragraph 15 of the judgment in Kwekerij van der Kooy and Others v Commission , and paragraphs 45 and 46 of the judgment in Arbeitsgemeinschaft Deutscher Luftfahrt-Unternehmen and Hapag-Lloyd v Commission . In the first of those judgments, the Court of Justice rejected the action brought by a number of Dutch horticulturalists against the Commission’s decision declaring to be incompatible with the common market the preferential tariff applied in the Netherlands to the supply of natural gas for horticulture in heated greenhouses. In the second judgment, the Court of First Instance declared to be inadmissible on the ground of lack of locus standi the action brought against the Commission’s decision not to authorise the extension of tax provisions setting in place a mechanism for the special depreciation treatment of the cost of purchasing specific categories of merchant ships, fishing boats and aircraft.
(15) – Case T‑398/94 [1996] ECR II‑477.
(16) – See paragraph 41.
(17) – Case C‑309/89 Cordorniu v Council [1994] ECR I‑1853, paragraph 19.
(18) – Joined Cases C‑182/03 and C‑217/03 Belgium and Forum 187 v Commission [2006] ECR I‑5479, paragraph 58 et seq.
(19) – The actual contested decision in the Cofaz case, concerning a tariff system for natural gas in the Netherlands, related to a measure of general application, even though the advantage linked to the application of that measure affected just four operators because of the attributes of the relevant market.
(20) – At that paragraph, the Court states that ‘ARE, which is an association set up to promote the collective interests of a category of persons, can only be regarded as being individually concerned within the meaning of the Plaumann v Commission case-law to the extent to which the position of its members in the market is substantially affected by the aid scheme covered by the contested decision’.
(21) – In its judgment in Case T‑188/95 Waterleiding Maatschappij v Commission [1998] ECR II‑3713, the Court of First Instance refers to that risk when it rejects the argument whereby the applicant seeks to link the damage to its own interests simply to its status as a person liable for the tax levied in the context of the tax regime authorised by the Commission.
(22) – See Joined Cases T‑447/93 to T‑449/93 AITEC and Others v Commission [1995] ECR II‑1971, paragraph 60, and Case T‑266/94 Skibsværftsforeningen and Others v Commission [1996] ECR II‑1399, paragraph 50.
(23) – See, among others, the order of 21 February 2006 in Case C‑367/04 P Deutsche Post and DHL Express v Commission (not published in the ECR), paragraph 41; Case C‑106/98 P Comité d’entreprise de la Société française de production and Others v Commission [2000] ECR I‑3659, paragraph 41; and Case C-525/04 P Spain v Lenzing [2007] ECR I‑0000, paragraph 33.
(24) – Paragraph 54.
(25) – In the abovementioned judgment in Comité d'entreprise de la Société française de production and Others v Commission , which upheld the inadmissibility order handed down by the Court of First Instance, the action before the Court of First Instance was brought by a number of trade union organisations and not by competitors of the undertaking in receipt of the aid. In the order of the Court of First Instance of 27 May 2004 in Case T‑358/02 Deutsche Post and DHL v Commission [2004] ECR II‑1565, in which the Court of First Instance rejected the action brought by a number of competitor undertakings of the undertaking in receipt of the aid, the applicants failed to provide evidence of the impact of the aid on their market position, and simply relied on their status as competitors. Finally, in the judgment in Case T‑117/04 Werkgroep Commerciële Jachthavens Zuidelijke Randmeren and Others v Commission [2006] ECR II‑3861, in which the Court of First Instance also dismissed the action, the applicants made a number of general claims concerning the impact of the aid measure on their profitability, but those claims were shown to be inaccurate by evidence furnished by the Commission and the Member State concerned.
(26) – See Spain v Lenzing , paragraph 34.
(27) – Ibid., paragraph 35.
(28) – Point 110 of the Opinion.
(29) – In assessing whether the market position of the applicant’s members could be regarded as being substantially affected by the aid scheme covered by the contested decision, the Court noted at paragraph 72 that, ‘[e]ven if ... certain of ARE’s members are economic operators who may be regarded as direct competitors of beneficiaries of the aid provided for by the Compensation Law and, therefore, their competitive position is necessarily affected by the contested decision, it does not follow that their position in the market could be substantially affected by the grant of such aid since it appears to be accepted … that all farmers in the European Union may be regarded as competitors of the beneficiaries of the land acquisition scheme’.
(30) – To that effect see, most recently, the abovementioned judgment in Lenzing , in which the Court considered sufficient to prove a substantial effect on the applicant’s market position evidence pertaining to the structure of the market and the policy on price applied by the undertaking in receipt of the aid, which did not relate to the specific circumstances of the applicant.
(31) – To that effect, see, for example, Waterleiding Maatschappij v Commission , paragraph 80. See also, and more recently, Case T‑146/03 Asociación de Estaciones de Servicio de Madrid and Federación Catalana de Estaciones de Servicio v Commission [2006] ECR II‑98, paragraph 52.
(32) – I should, by the way, point out that, again in response to the cross-appeal, BAA alleges that the Commission has misconstrued paragraph 54 of the judgment under appeal, which reads as follows: ‘[b]y this action, the applicant does not merely seek to challenge the Commission’s refusal to initiate the formal investigation procedure, but also calls into question the merits of the contested decision. Accordingly, it is necessary to consider whether it has explained, by reference to the relevant criteria, why the AGL is liable to have a significant effect on the position of one or more of its members on the market for aggregates’. Unlike the Commission, BAA considers that the Court of First Instance did not intend, in that paragraph, to exclude the applicability of the Cook and Matra case-law in this case. Unlike BAA, I consider that the reading of paragraph 54 of the judgment under appeal which the Commission proposes is correct, and that the Court of First Instance has in effect ruled out the possibility of BAA relying on the conditions of admissibility which the judgments in Cook and Matra envisage, particularly in relation to the grounds for the action which seek to call into question the merits of the contested decision.
(33) – See, in that connection, albeit taking different approaches, the Opinion of Advocate General Jacobs in the ARE case and Advocate General Bot’s Opinion of 6 March 2008 in Case C-75/05 P Germany v Kronofrance (not yet published in the ECR).
(34) – The admissibility of an action, based on the judgments in Cook and Matra , against a decision under Article 88(3) EC ruling out the existence of aid , which has been specifically recognised by the Court of First Instance (see Case T‑11/95 BP Chemicals v Commission [1998] ECR II‑3235, paragraphs 165 and 166, and Joined Cases T‑346/99 to T‑348/99 Diputación Foral de Álava v Commission [2002] ECR II‑4259, paragraphs 41 and 75 to 79) does not seem to me to be able to be called into question, if it is accepted that the Commission is also required to initiate the procedure under Article 88(2) EC in circumstances in which it experiences serious difficulties in classifying the measure in question as aid in the course of the preliminary examination.
(35) – See, by analogy, Joined Cases C‑442/03 P and C‑471/03 P P & O European Ferries (Vizcaya) and Diputación Foral de Vizcaya v Commission [2006] ECR I‑4845, paragraph 41 et seq.
(36) – See Case T‑55/99 CETM v Commission [2000] ECR II‑3207, paragraph 40.
(37) – See, in particular, Case C‑75/97 Belgium v Commission [1999] ECR I‑3671, paragraph 33, and Case C‑148/04 Unicredito Italiano [2005] ECR I‑11137, paragraph 45.
(38) – See Case C-241/94 France v Commission [1996] ECR I-4551, paragraph 24; Case C‑200/97 Ecotrade [1998] ECR I-7907, paragraphs 40 and 41; and Case C‑75/97 Belgium v Commission , paragraph 26.
(39) – See Case C‑53/00 Ferring [2001] ECR I‑9067, paragraphs 18 to 20 and 22.
(40) – See, to that effect, Case 173/73 Italy v Commission [1974] ECR 709, paragraph 33; Unicredito Italiano , paragraph 51; and Case C‑88/03 Portugal v Commission [2006] ECR I‑7115, paragraph 52. In that connection, it is worth pointing out that this approach was also adopted in Case C‑308/01 GIL Insurance and Others [2004] ECR I‑4777, paragraphs 65 to 78, in which the Court assessed whether the introduction in the United Kingdom of a higher rate of insurance premium tax for certain categories of insurance contract was justified by the nature and structure of the national system for taxing insurance premiums, irrespective, therefore of whether there was a selective advantage for operators subject to the standard rate.
(41) – See Portugal v Commission , paragraph 56.
(42) – Case C‑143/99 Adria-Wien Pipeline and Wietersdorfer & Peggauer Zementwerke [2001] ECR I‑8365.
(43) – Paragraph 114 of the judgment under appeal. Emphasis added.
(44) – Paragraph 114.
(45) – Paragraph 116.
(46) – Paragraph 115.
(47) – The exemption was accorded in the context of the 1996 Strukturanpassungsgesetz (Austrian Law on structural adjustment) in relation to the consumption of natural gas and electricity by undertakings. The Court found that conferring advantages on undertakings whose principal activity was the manufacture of goods was not justified by the nature or general structure of the taxation system introduced under the Strukturanpassungsgesetz.
(48) – See, among others, Case C‑409/00 Spain v Commission [2003] ECR I‑1487, paragraphs 53 and 54.
(49) – See Case 173/73 Italy v Commission ; Case C-56/93 Belgium v Commission [1996] ECR I-723, paragraph 79; and Case C‑241/94 France v Commission [1996] ECR I‑4551, paragraph 20.
(50) – See, for example, Case C‑189/91 Kirsammer-Hack [1993] ECR I‑6185, paragraphs 17 and 18, with reference to the condition for the use of State resources, or the judgment in Ferring , concerning compensation for services of general economic interest, with regard to the existence of an advantage, or the judgment in Adria-Wien Pipeline , paragraph 52, in particular, and, more generally, the case-law cited in point 83 above, with reference to the criterion of selectivity.
(51) – See, for example, paragraphs 128 and 130 of the judgment under appeal.
(52) – Paragraph 123 et seq. of the judgment under appeal.
(53) – But it seems to me that the consequence which flows from the definition of the environmental levy as an autonomous fiscal measure, that is to say, the possibility of justifying any differences in the treatment of undertakings or productive activities by reference to the nature and structure of the measure interpreted as a system , accords with that case-law.
(54) – See paragraph 114 of the judgment under appeal.
(55) – That conclusion is to be found at paragraph 128 of the judgment under appeal, in which the Court of First Instance states that ‘the decision to impose an environmental levy only in the aggregates sector ... even if based on a desire to maintain the international competitiveness of certain sectors, does not therefore undermine the AGL’s consistency with the environmental objectives pursued’.
(56) – Case C‑164/98 P DIR International Film and Others v Commission [2000] ECR I‑447, paragraphs 38 and 42.
(57) – In the appeal, BAA mentions points 10 and 16 of the contested decision which form part of the statement of facts.
(58) – Paragraph 31 of the contested decision.
(59) – See, for example, Case C‑83/98 P France v Ladbroke Racing and Commission [2000] ECR I‑3271, paragraph 25.
(60) – See, among others, Case C‑301/87 France v Commission [1990] ECR I‑307, paragraph 49.
(61) – See France v Ladbroke Racing and Commission , paragraph 25. In his Opinion in Case C‑222/04 Cassa di Risparmio di Firenze and Others [2006] ECR I‑289, points 109 to 112, Advocate General Jacobs took the opposite view as regards conferring on the Commission wide discretion in relation to the complex economic appraisals required for the purposes of classifying the national measure at issue in accordance with Article 87(1) EC.
(62) – See Case C‑56/93 Belgium v Commission , paragraph 11, and Joined Cases T‑126/96 and T‑127/96 BFM and EFIM v Commission [1998] ECR II‑3437, paragraph 81.
(63) – See, among others, Matra v Commission , paragraph 33.
(64) – As regards the quotation from the judgment in Matra v Commission , at paragraph 118 of the judgment under appeal, I shall merely point out that the paragraph of that judgment which is cited, namely ‘having regard to the broad discretion which the Commission has in the application of Article [88(3) EC]’ (paragraph 24) is specifically based on a precedent, Case C‑303/88 Italy v Commission [1991] ECR I-1433, paragraph 34, in which that margin of discretion was recognised in relation to Article 87(3) and not Article 88(3) EC. The same precedent is, furthermore, cited, in a similar context, at point 12 of the Opinion of Advocate General Van Gerven, and, on that occasion, the reference is properly made to Article 87(3) EC. In those circumstances, it does not seem to me to be possible to rule out that paragraph 24 of the judgment in Matra v Commission may contain a material error.
(65) – See the Opinion of Advocate General Cosmas in France v Ladbroke Racing and Commission , point 15.
(66) – See France v Ladbroke Racing and Commission , paragraph 25.