Opinion of the Advocate-General
++++
Mr President,
Members of the Court,
1. In this action, the Federal Republic of Germany seeks pursuant to Article 173 of the EEC Treaty a declaration that Commission Decision C(92) 1783 Final of 23 September 1992, since published under number 92/491/EEC, on the clearance of the accounts presented by the Member States in respect of the expenditure for 1989 of the European Agricultural Guidance and Guarantee Fund (EAGGF), Guarantee Section, (1) is void on the ground that the Commission excluded from Community financing a sum of DM 432 000 representing aid for the processing of skimmed milk into casein and caseinates.
2. In the disputed decision, the Commission refused to charge that amount to the EAGGF, considering that the conditions laid down by the Community legislation for granting the aid were not satisfied.
3. The relevant provision, namely Regulation (EEC) No 756/70 of the Commission of 24 April 1970 on granting aid for skimmed milk processed into casein and caseinates, (2) provides in Article 3 that:
"(1) Producers of casein or caseinates shall not benefit from the aid unless they:
(a) ...
(b) submit to supervision by the competent intervention agency.
(2) ...
(3) The supervision mentioned in paragraph 1(b) shall consist of at least a regular supervision of the manufacturing concern and of the composition of the casein and caseinates.
(4) ..."
4. Let me state at the outset that that concept of "regular supervision" has not been defined by the Community legislation, although it is apparent from the Special Report of the Court of Auditors dated 11 November 1983 (3) that the majority of the Member States, considering that a permanent inspector was not justified, were of the view that having inspections on a regular basis satisfied the requirements of Article 3. That view is not criticized by the Commission.
5. The system applying in Germany at the relevant time was as follows: there were two stages to the supervision, one internal, the other official.
6. For the internal check, the manufacturers themselves took samples from each batch of casein or caseinates; the samples were analysed by their own laboratory, which recorded the results in a register. According to the German Government' s submissions at the hearing, aid was requested only in respect of the batches whose composition complied with the Community regulation.
7. As for the official check, agents of the competent body visited the manufacturing sites once a week to ensure that the internal analyses were carried out. Twice a month, they took samples of each sort of casein or caseinate, which were then analysed by an official laboratory. Lastly, every four to six months a final check was made at each manufacturing concern, at the end of which it was decided whether to grant aid.
8. According to the applicant, that system ensured regular supervision within the meaning of Article 3 of the regulation cited above, in so far as it "... guaranteed that samples from all the batches were examined". (4)
9. The Commission does not criticize the system of checking so much as its consequences for the grant of aid where the result of the official analyses was unsatisfactory, and thus conflicted with the internal analyses.
10. In my view it is useful, for a full understanding of the matter, to set out the example given by the Commission in its defence and taken from Annexes 1a and 2 to the application, namely that of the undertaking Lauingen for production in August 1988. During that period, five batches were manufactured, which were all given an internal analysis which confirmed their eligibility. The official supervisory body took samples from seven batches. One of those samples was "negative", that is to say not in compliance with the requirements. The Federal Republic of Germany disallowed aid for the corresponding batch. The Commission considers that such a method is contrary to Community law since "... although more than 14% of the official samples were negative (i.e. one in seven), aid was disallowed for only 2% of the total quantity (one batch out of 50)". (5)
11. In support of its application, the Federal Republic of Germany puts forward three pleas in law which I will consider in turn. It considers first that the supervision it has set up is consistent with the concept of "regular supervision". It then invokes the principle of the protection of legitimate expectations in the event that the Court considers that that supervision is insufficient. Finally it states that a sum of DM 24 365 is in any event due.
12. I will start by considering the first plea.
13. The German system is based principally on self-checking by the manufacturing concerns, which must analyse each batch of casein and caseinate manufactured, aid being requested only for those batches which comply with the requirements. By way of additional check, at irregular intervals the competent authorities take samples which are then analysed in official laboratories. The applicant thus considers that "it was guaranteed that for each batch a representative sample had been taken, analysed and recorded". (6)
14. Although that last statement is correct, it would be wrong to conclude from it that the authorities check each batch manufactured. Unless self-checking is considered to have the same value as checking by the authorities, the arrangements described above amount simply to a method of checking by sample.
15. Self-checking can give rise to no more than a simple presumption of regularity which can obviously be challenged by the authorities. If the official analysis disclosed a negative result, the manufacturer would either have to ask for the batch in question to be double-checked by reference to the reserve sample (7) or have to accept the extrapolation of that result to the entire output. If the manufacturer or the competent body were to resist the principle of extrapolation, an official analysis would have to be made of all the batches manufactured.
16. It is true that, as I have said, the Community legislation does not define what is to be understood by "regular supervision". However, the very concept of supervision necessarily presupposes action external to the manufacturer, hence by a public authority or a body designated by it.
17. Thus self-checking, since it enables the manufacturer to anticipate to some extent the forthcoming aid, cannot be regarded as the "supervision" laid down in the regulation, so that the results of the internal analyses cease to be reliable where they do not correspond to those carried out by the supervisory authority.
18. I therefore consider that the first plea must be rejected.
19. I will now consider the second plea in which the German Government raises infringement of the principle of the protection of legitimate expectations on the ground that the Commission had previously accepted that the system was compatible with Community law.
20. The applicant relies on the Summary Report for 1974-1975, production of which was requested by the Court. The relevant information, so far as concerns the Federal Republic of Germany, is set out on page 43 as follows:
"In Germany the current system does not provide for the extrapolation of results, in the case of a negative analysis, to the whole of the period for which the sample is representative. Under national legislation the applicant undertakings are required to take samples from the daily output or from each lot of casein and caseinate, to enter the results in special laboratory records and to make samples available for a second opinion. Representatives of the Federal Office also take individual and mixed samples at irregular intervals; these are then analysed at official analytical laboratories to check that the quality criteria laid down in the regulations are being observed. The findings, together with the results of the analysis in the firm' s laboratories, provide the basis for the quality classification. If the results are negative, specific analyses are made. The individual results are then classified.
The negative results and the results with high quality classification can then be attributed only to the quantities from which the samples are taken."
21. In connection with the principle of the protection of legitimate expectations, I note the settled case-law of the Court, and in particular a judgment Italy v Commission (8) in which the Court stated that
"... the practice followed by the Italian authorities arises from an incorrect interpretation of Community law. In such a case the Commission is not obliged to charge expenditure incurred on that basis to the Fund unless the incorrect interpretation may be attributed to a Community institution." (9)
22. I will say at the outset that I have not found the arguments put forward by the applicant in support of this plea convincing.
23. The extract from the Report quoted above is set out as a brief summary of the system applying in the Federal Republic of Germany at the relevant time. That system gives weight to the internal checks, but subject to a limit: where the internal and the official analyses conflict.
24. Where the official analyses confirm the results of the internal analyses relating to the same batches, the other batches are considered to satisfy the criteria set out in the regulation.
25. Where on the other hand there is a conflict, the Summary Report indicates that "specific analyses are made", which, according to the Commission, unquestionably entails the obligation to carry out further investigations. The applicant does not really state how it considers those words should be interpreted, simply submitting "... that, where the results were negative, reserve samples, taken when the samples from each batch were taken, were tested in another laboratory, if the undertakings so requested". (10)
26. The Commission' s is the only interpretation compatible with the coherence of the system. It is only after specific and generalized administrative checks that it is no longer necessary to extrapolate the results: such a procedure checks not samples but the entire output during the relevant period.
27. That interpretation also follows from a comparison of the German system with that then applying in France, which the Commission criticized in the Summary Report on the ground that where the analysis was negative the result was not spread over total output but simply applied to the batch corresponding to the sample analysed. The Report accordingly took issue with the French method for the same reasons as those set out before the Court in relation to the German system.
28. That document suggests that in France there was only an official check carried out without notice once a fortnight, with no internal checking within the undertaking. (11) In any event, what value would an analysis by the undertaking itself have where the official check proves negative?
29. Thus where the official analysis discloses one negative batch, that casts doubt on the accuracy of the internal analyses so that it cannot be assumed that the batches not examined by the official body comply with the requirements. It is therefore the analysis of the entire output which is called in question.
30. As for the applicant' s argument that the undertakings could request a fresh examination by another laboratory, it may well be either that they did not avail themselves of that possibility or that double-checking disclosed another negative result, given that it has never been established or even argued by the Federal Republic of Germany that batches in respect of which aid was initially disallowed were subsequently found to satisfy the conditions laid down in the regulation. (12)
31. I accordingly consider that the second plea should also be rejected.
32. I will now examine the last plea in which the Federal Republic of Germany claims that DM 24 365 ° which breaks down into two sums of DM 6 668 and DM 17 697 ° should in any event be charged to the EAGGF.
33. It is stated in the applicant' s observations that the sum of DM 6 668 relates to a quantity of casein which was initially considered not to satisfy the quality criteria laid down in the regulation, although that finding was subsequently shown to be wrong. (13)
34. As for the sum of DM 17 697, it allegedly relates to errors concerning the number of negative samples since not only the main samples but also those taken for cross-checking were wrongly considered to be negative. (14)
35. The Commission had refused to charge the total of those sums to the EAGGF on the ground that the explanations, even though they may have been relevant, were provided after the deadline of 15 July 1991 imposed on the applicant in Decision C(91) 1270 Final of 28 June 1991 for submission of its observations.
36. However, at the stage of the rejoinder, after stating that, in so far as concerns the sum of DM 6 668, it had received details before that deadline, it accepted that as to that point the request was valid. (15)
37. The decision in question must therefore immediately be annulled to the extent of that amount.
38. To justify its refusal to pay the sum of DM 17 697, the Commission relies on Article 1 of Regulation (EEC) No 422/86, (16) which provides:
"Additional information may be forwarded to the Commission up to a deadline to be fixed by the Commission, taking into account in particular the amount of work necessary to supply the information concerned. In the case of failure to submit the aforementioned information within the period fixed, the Commission shall take its decision on the basis of those elements of information in its possession at the deadline, except in cases where the late submission of information is justified by exceptional circumstances."
39. The Federal Republic of Germany refers however to a letter from the Commission dated 1 August 1991 requesting it, without mentioning a time-limit, to provide further information. It considers that that letter demonstrates the Commission' s willingness to take into consideration observations made after 15 July 1991. Moreover, it claims that the method of calculation used by the Commission was not known until a meeting held on 23 January 1992.
40. I accept that argument. Although that letter is set out as a "summary of the main results of the bilateral meeting of 21 June 1991", it is there stated, in relation to the "implementation of the legislation on aid for casein":
"So that the EAGGF can more accurately assess the financial effects of that irregular procedure, the German authorities were requested to supplement the information in Annexes I and II to the letter of 20 March 1991 as follows ..."
41. Nowhere is there any mention of the time-limit initially imposed, which must have been known when that letter was sent, still less of its expiry. The Federal Republic of Germany was therefore entitled to consider that the Commission was prepared to take into consideration information provided to it thereafter.
42. In a decision concerning a case in which the Commission pleaded expiry of the time-limit, even though it had reserved its position until judgment was given in a pending case, the Court stated that:
"By merely reserving itself the right to review the question, however, the Commission did not signify an intention to close the matter as far as evidence of actual losses was concerned. A Member State was thus entitled to infer that the review could extend to any question left open by the judgments of the Court dealing with the financial years 1974 and 1975." (17)
43. Although it is not disputed that the Commission made clear its intention to close the procedure within a certain period, it subsequently sent a letter which, although set out as the summary of a meeting, nonetheless included an invitation to produce certain documents.
44. Moreover, the documents annexed to the pleadings, which come from both the applicant and the Commission, do not show conclusively that, before the decision setting the deadline, the method of calculation used by the Commission was known. Furthermore, the Commission does not claim that it was, simply considering that the point is irrelevant. (18)
45. I note that although the Commission initially stated that "Where the results are not satisfactory, the aid should be refused not just for the batch in question but for the entire output between two satisfactory results", (19) it subsequently considered that it was sufficient to apply the results to the entire output in the same proportions as the samples taken. That demonstrates the importance of ascertaining the method of calculation, since the consequences for the sum in dispute vary according to the method.
46. I accordingly consider that the decision should also be annulled on that point, the Commission having furthermore stated at the hearing that, had the request that that sum be charged to the Fund been submitted in due time, it would in fact have been granted.
47. I therefore propose that the Court:
° annul Commission Decision C(92) 1783 Final, published in the Official Journal of the European Communities under number 92/491/EEC, dated 23 September 1992, in so far as it disallows the charge to the EAGGF of the sum of DM 24 365 representing aid for the processing of skimmed milk into casein and caseinates,
° dismiss the remainder of the application, and
° order each party to bear its own costs.
(*) Original language: French.
(1) ° OJ L 298, p. 23.
(2) ° OJ, English Special Edition 1970 (I), p. 201.
(3) ° Special Report on aid to skimmed milk processed into casein and caseinates (OJ 1984 C 41, p. 1).
(4) ° Paragraph 15 of the application.
(5) ° Paragraph 13 of the defence.
(6) ° Paragraph 7 of the application.
(7) ° Two samples are taken from each batch, one for internal analysis and the other for administrative checking if necessary.
(8) ° Case 1251/79 [1981] ECR 205; see also Case 820/79 Belgium v Commission [1980] ECR 3537.
(9) ° Paragraph 17.
(10) ° Paragraph 12 of the reply; my emphasis.
(11) ° Page 42 of the Summary Report.
(12) ° In the third plea however the applicant requests that the sum of DM 6 668 be charged to the Fund precisely because further analyses showed that the product complied with the Community regulation.
(13) ° Paragraph 11 of the application.
(14) ° Ibidem, paragraph 12.
(15) ° Paragraph 18.
(16) ° Commission Regulation of 25 February 1986 amending Regulation (EEC) No 1723/72 on making up accounts for the European Agricultural Guidance and Guarantee Fund, Guarantee Section (OJ 1986 L 48, p. 31).
(17) ° Case 129/84 Italy v Commission [1986] ECR 309, at paragraph 41 of the judgment.
(18) ° Defence, paragraph 31.
(19) ° Annex 5 to the application. See also paragraph 19 of the defence.
Full & Egal Universal Law Academy