Important legal notice

61985C0182

European Court reports 1987 Page 03159

Opinion of the Advocate-General

Mr President, Members of the Court, 1 . In this reference for a preliminary ruling, the Amtsgericht Koeln ( Local Court, Cologne ) is asking the Court to interpret certain regulations concerning aid for skimmed-milk powder used as animal feed .

The questions submitted seek essentially to establish whether the aid system is subject to the method of analysis and corresponding tolerance margin provided for by the intervention system for the purpose of determining the presence of whey in skimmed-milk powder . The national court also seeks to ascertain whether, if it is established that the aforesaid method of analysis is applicable to the aid system and has led to the finding that the skimmed-milk powder is free of whey, the competent authorities are entitled to reclaim the aid if new and different tests reveal the presence of whey in the product .

2 . The main dispute arises from the delivery of a consignment of skimmed-milk powder by a German undertaking, Luetticke, to another German undertaking, Denkavit, which was to use it in the preparation of compound feedingstuffs eligible for Community aid .

The contract concluded between the parties stipulated that, if the milk supplied were ineligible for the aid under the rules then in force ( Regulation ( EEC ) No 986/68 of the Council of 15 July 1968 laying down general rules for granting aid for skimmed milk and skimmed-milk powder for use as feed - Official Journal, English Special Edition 1968 ( I ), p.*260 - and Commission Regulation ( EEC ) No 1725/79 of 26 July 1979 on the rules for granting aid to skimmed milk processed into compound feedingstuffs and skimmed-milk powder intended for feed for calves - Official Journal 1979, L 199, p . 1 - as subsequently amended ), the supplier was required to take the goods back, repay the price if already paid by the purchaser and bear the costs resulting from the analysis and transport of the product .

In the opposite situation - that is to say if the skimmed-milk powder were eligible for the aid but Denkavit none the less refused to accept the product and Luetticke took it back - the costs of transport and analysis were to be borne by the purchaser .

Under the terms of the aforesaid contract, Denkavit analysed a consignment received on 14 July 1983 . Depending on the method of analysis used, the whey present in the skimmed-milk powder varied between a maximum of 3% and a minimum - established by applying the method provided for under the intervention system - of 0.5 %. Denkavit then requested Luetticke to take the goods back . Luetticke did so but claimed reimbursement of the costs and, faced with Denkavit' s refusal, instituted proceedings against the latter before the Amtsgericht Koeln .

In those proceedings, the dispute between the parties turned on the question whether the consignment of milk was eligible for aid under Regulation No 1725/79 . Denkavit argued that any trace of whey in the skimmed-milk powder rendered it ineligible for aid . Although Commission Regulation ( EEC ) No 625/78 of 30 March 1978 on detailed rules of application for public storage of skimmed-milk powder ( Official Journal 1978, L 84, p . 19 ) provides for the possibility of offering the skimmed-milk powder for sale to the intervention agency provided that whey is not present in the product in excess of 2%, Regulation No 1725/79 does not provide for any tolerance margin . Accordingly, if it had agreed to accept the milk, the defendant would have exposed itself to the risk of having to repay the aid which it received; that risk would have existed for precisely 30 years, for that is the length of the limitation period to which the claim for repayment is subject, according to the competent German authority, namely the Bundesamt fuer Ernaehrung und Forstwirtschaft ( Federal Office for Food and Forestry ).

The plaintiff takes the opposite view . Luetticke considered that the argument that a product containing a minimum percentage of whey was ineligible for aid was both unfair and untenable . Regulation No 1725/79 does not provide for a method of analysis and, in those circumstances, it is clear that reference must be made, by analogy, to the method and tolerance margin laid down by Regulation No 625/78 . A different interpretation would be contrary to the principle of proportionality and, by depriving of aid products qualitatively suitable for public storage, it would frustrate one of the purposes for which those aids were established, namely to prevent the skimmed-milk powder surplus in its entirety from being offered for sale to intervention agencies .

The Amtsgericht Koeln considered that the latter argument was not devoid of merit, particularly if it was true, as the plaintiff maintained, that the presence of minimal traces of whey might be due not to fraudulent admixtures but to other factors . The Amtsgericht therefore stayed the proceedings and, pursuant to Article 177 of the EEC Treaty, referred to the Court the following questions for a preliminary ruling :

"( 1 ) Is the intervention system laid down for the milk sector under basic Regulation No 804/68 on the common organization of the market in milk and milk products ( Official Journal, English Special Edition 1968 ( I ), p . 176 ) to be interpreted as meaning that skimmed-milk powder which meets the quality requirements for intervention storage in accordance with Article 5 of Regulation No 804/68, taken in conjunction with Regulation No 625/78, must therefore be regarded as being eligible for aid under Article 10 of Regulation No 804/68, taken in conjunction with Regulation No 1725/79? ( 2 ) If, upon application of the method of analysis provided for in Annex IV to Regulation No 625/78, it may be concluded, having regard to the tolerance margin specified in that provision, that whey is absent, does that mean that the skimmed-milk powder in question may also be regarded as free of whey for the purposes of the grant of aid under Regulation No 1725/79? ( 3 ) If Question 2 is answered in the affirmative : ( a ) does the fact that the method of analysis provided for under the intervention rules in Annex IV to Regulation No 625/78 allows for a tolerance margin of 2% mean that the same tolerance margin is to be applied to the results of analyses carried out by the Member States, using other methods not laid down by Community law, in the procedure for the grant of aid under Regulation No 1725/79? ( b ) if it may be concluded on the basis of the results of an analysis taking into account the relevant tolerance margin that whey is absent from a consignment of skimmed-milk powder, does that mean that the recipient of aid under Regulation No 1725/79 has an absolute defence to a claim for repayment of the aid even if the competent authority establishes on the basis of other findings ( for example following an inspection at the premises of the manufacturer of the skimmed-milk powder in question ) that whey powder has been added to the consignment of skimmed-milk powder? ( 4 ) Does Article 1 ( 2 ) of Regulation No 1725/79 contravene the principle of proportionality in Community law, in so far as skimmed-milk powder is not eligible for aid because whey is found to be present even though the same product must be held to be eligible for sale to the intervention agency on the basis of Annex IV to Regulation No 625/78?" In these proceedings, written observations were submitted by the parties to the main proceedings and by the Commission of the European Communities .

3 . Before I consider those questions, it is appropriate to give a brief outline of the relevant rules of Community law on the contested aid .

Article 10 of Regulation No 804/68 of the Council on the common organization of the market in milk and milk products provides for the grant of aid for skimmed-milk powder used as animal feed, on condition that it meets certain standards as to quality . That provision is substantially the same as Article 7 which governs the buying-in of skimmed-milk powder by intervention agencies . Article 7, however, requires the product to be "first-quality" but does not require it to be used as animal feed .

In Regulation No 986/68, which was adopted subsequently, the Council laid down general rules for granting the aid provided for by Article 10 of Regulation No 804/68 . Article 1 ( d ) of Regulation No 986/68, as amended by Council Regulation ( EEC ) No 876/77 ( Official Journal 1977, L 186, p . 24 ), provides that "skimmed-milk powder" means "powdered milk and buttermilk with a maximum fat content of 11% ...". Article 2 of Regulation No 986/68, as amended by Council Regulation ( EEC ) No 2128/84 ( Official Journal 1984, L 196, p . 6 ), provides that aid is to be granted for "skimmed-milk powder and buttermilk powder ... used in the manufacture of compound feedingstuffs ...". Finally, Article 1 of Commission Regulation ( EEC ) No 1725/79 provides that skimmed-milk powder intended for feed for calves qualifies for aid only if it corresponds to the aforesaid definition .

However, that definition must be viewed in conjunction with the definition of milk ( or buttermilk ) in Article 1 ( a ) of Regulation No 986/68, namely "the milk-yield of one or more cows, to which nothing has been added and which has, at the most, been only partially skimmed" ( emphasis added ). It follows that, for the purposes of the rules applicable in these proceedings and in particular Regulation No 1725/79, milk which has been prepared with ingredients other than those prescribed or with the admixture of substances not expressly permitted is ineligible for aid, even if its appearance and chemical composition are perfectly normal . Moreover, that principle was clearly laid down by the Court in its judgment of 21 September 1983 in Joined Cases 205 to 215/82 Deutsche Milchkontor v Germany (( 1983 )) ECR 2633, paragraph 12 of the decision, and reference must be made to it in order to resolve the problems before the Court .

4 . In its first question the Amtsgericht Koeln asks essentially whether milk intended for sale to the intervention agency can also be considered eligible for aid . Luetticke favours an affirmative answer . It maintains that the aid system is related to the intervention system because it enables the burdens which it imposes to be alleviated to an appreciable extent by diverting a proportion of the skimmed-milk powder available for public storage and using it for a less costly purpose . According to Denkavit and the Commission, however, the two systems apply to products which are inherently different, and it is therefore impossible to supplement one system with elements borrowed from the other .

Let me say at once that the premise on which the latter contention is based is correct . As we have just seen, "first-quality" milk must be earmarked for sale to the intervention agency, although no such requirement is imposed in the case of milk eligible for aid . In addition, the intervention system expressly provides that the skimmed-milk powder must not contain buttermilk ( Article 1 ( 1 ) ( c ) of Regulation No 625/78 ) whilst under the aid system the presence of buttermilk in the product has been tolerated at least since 1975 ( Article 1 ( a ) of Regulation No 986/68, as amended by Regulation ( EEC ) No 472/75, Official Journal 1975, L 52, p . 22, particularly the first recital in the preamble ). Thirdly, the fat content, which under the intervention system may not exceed 1.25% ( Annex I to Regulation No 625/78 ), may under the aid system be as high as 11% ( Article 1 ( d ) of Regulation No 986/68, as amended by Regulation No 472/75 ). Finally, although under the intervention system the milk may not be more than one month old, no such limit is imposed by the aid system .

However, the consequences drawn from those facts by Denkavit and the Commission are excessive . Although it is true that a product which satisfies the requirements laid down for eligibility for aid will hardly ever be eligible for sale to the intervention agency, the opposite contention is surely incorrect . Since the characteristics which a product must exhibit in order to be bought in by the intervention agency ( high quality, absence of buttermilk, lower fat content, greater freshness ) are stricter, it is clear that a product considered eligible under the applicable system will, at least in principle, also be considered eligible under the other system . Does it follow, as Luetticke maintains, that the latter system is subject to the method of analysis and tolerance margin prescribed by the former system for detecting the presence of whey in skimmed-milk powder?

That remains to be seen . I would point out, to begin with, that the fact that the aid system does not provide for specific methods and tolerance margins lacks the significance ascribed to it by the Commission . Article 10 of Regulation No 1725/79 provides that "with a view to ensuring compliance with the provisions of Article 1 ( 2 ) and ( 4 )" the Member States are to "provide for supervision of the enterprises concerned" with regard to "the composition of the ... skimmed-milk powder ". In so far as it adopts the definition in Article 1 ( a ) and ( d ) of Regulation No 986/68 ( supra, Section 3, at the end ), Article 1 ( 2 ) of Regulation No 1725/79 requires that nothing should have been added to the skimmed-milk powder . Accordingly, one of the aims of supervision of the enterprises concerned will be to establish whether the skimmed-milk powder has been enriched with whey . Moreover, the Court has also come to a similar conclusion, although it is couched in more general terms . In its judgment in the aforesaid Milchkontor case, it held that Member States "must verify by means of appropriate controls that skimmed-milk powder complies with the relevant Community rules so as to ensure that Community aids are not paid in respect of products for which they ought not to be granted" ( paragraph 43 of the decision ).

The Commission disagrees . In its view, as is clear from the analysis report set out in Annex I to Regulation No 1725/79, it is unnecessary to test the product for the presence of whey and the test is carried out only at the request of the national authorities . However, that opinion must be rejected . The analysis report in fact provides for the indication of the percentage content of whey in the product only in so far as its detection is required by the Member States ( A 2 ( g )*). It is clear, however, that a text of that kind cannot be binding when it comes to interpreting the provisions of a regulation and, as I have just pointed out, the relevant provisions require the milk to be free of whey and impose on the Member States the obligation to ensure by means of appropriate controls that no trace of whey is present in the product .

Now that this crucial point has been clarified, the essence of the problem is to determine the scope of those controls . The aforesaid provisions and the case-law of the Court clearly demonstrate, it seems to me, that it is for the Member States to determine the scope of those controls and the procedures for applying them, and that also holds true for the tolerance margin . The fact that Regulation No 1725/79 makes no mention of the tolerance margin certainly does not mean, as Denkavit and the Commission maintain, that it is inapplicable to the aid system . The consequence that must be drawn from the regulation' s silence on that point is quite different : the existence and the extent of the tolerance margin depend on the type of analysis selected by the State in the exercise of the discretion conferred upon it by the relevant rules of Community law .

However, the scope of that discretion must be clearly understood . As is stated in point 8.3 of Annex IV to Regulation No 625/78 with regard to the method of analysis laid down for skimmed-milk powder intended for public storage, a tolerance margin needs to be fixed for technical reasons connected with the ever-present possibility of "errors in the method" or "natural variations in the composition of the sample", and certainly does not imply that whey may be added to the milk up to the limit of the tolerance margin . Hence there is a discretion, but it must be exercised within the limits of the objective - namely verification that the product is free of whey - which underlies the entire body of rules comprised in both the aid system and the intervention system . In other words, it would be unlawful to establish a margin taking account of factors which cannot be attributed to the inaccuracy of the method used for verifying that the product is free of whey .

If those observations are correct, it clearly follows that the Member States are free to apply in the matter of aid the method specified in Regulation No 625/78 with the corresponding tolerance margins. The Commission refuses to resign itself to that conclusion and points out that the aforesaid method is designed to detect the presence of rennet whey in skimmed-milk powder, whereas so far as the grant of aid is concerned, Regulation No 1725/79 and Annex I thereto do not distinguish between rennet whey and acid whey. However, that argument is far from persuasive. For one thing, the Commission has weakened its case by acknowledging that "at present no objective method is recognized for detecting the presence of acid whey. Such a method exists only for detecting rennet whey, namely that described in Annex IV to Regulation No 625/78." It is sufficient to recall, however, that milk eligible for aid may exhibit the characteristics of a product eligible for sale to the intervention agency, from which it may be inferred that, at least where that is the case, the product is subject to the method of analysis prescribed by Community law. Moreover, the documents produced by the plaintiff and even by the defendant show that this is what normally happens in practice.

A final point. The intervention system and the aid system both form part of the common organization of the market in milk and milk products (judgment of 18 October 1979 in Case 5/79 Buys v Denkavit ((1979)) ECR 3203, paragraph 20 of the decision) and complement one another so as to satisfy the requirements of that common organization of the market in optimum terms. Thus, it is true, as Luetticke maintains, that the aid granted in respect of skimmed-milk powder intended for use as animal feed is designed to prevent the largest possible quantity of skimmed milk from being bought in by the intervention agency. That aim - which is the result of the existence of huge stocks of skimmed-milk powder and the high costs resulting from prolonged public storage - is expressly set out in the first two recitals in the preamble to Regulation No 876/77.

In other words there is no logical or technical reason for considering the two systems incompatible as regards the type and the quality of the products to which they apply. Admittedly, the higher value of skimmed-milk powder intended for sale to the intervention agency will render its use in the preparation of animal feed a less attractive option. But it is precisely the two disquieting phenomena to which I have referred - the skimmed-milk powder surplus and the high cost of public storage - which make it advisable to provide incentives for exercising that option.

The conclusion which I have arrived at and the arguments which I have used in support thereof also enable the other problems raised by the national court to be resolved without too many difficulties.

To begin with, there is the problem raised in Question No 2. The Amtsgericht seeks to ascertain whether, once it has been established by applying the Community method that no whey is present in the product, the latter is to be regarded as free of whey also for the purposes of the grant of aid. The parties' answers are entirely foreseeable. Luetticke favours an affirmative answer because the unity of the system makes it necessary to adopt the same method in all cases. Denkavit and the Commission favour a negative answer because the intrinsic difference between the products eligible for sale to the intervention agency and those eligible for aid precludes any possibility of applying in one sector the rules applicable in the other sector.

As for me, I can only refer to the principles which I emphasized earlier. Since under the aid system the power to choose the method for detecting the presence of whey is vested in the Member States, the solution of the problem depends on the method of analysis adopted by the competent national authorities in specific cases. If that method allows for a smaller tolerance margin than that specified in Regulation No 625/78, the product tested by means of the Community method will not be recognized as being automatically eligible for the aid provided for by Regulation No 1725/79. If, on the other hand, the national method coincides with the Community method, the fact that the milk satisfies the requirements prescribed for sale to the intervention agency will also render it eligible for the grant of aid.

The objection will be raised that any differences between the methods chosen by the various Member States may give rise to certain drawbacks. That is true but the Court has already taken that objection into account, albeit in different circumstances, in paragraph 24 of its decision in the Milchkontor case, where it held that "... if disparities in the legislation of Member States proved to be such as to compromise the equal treatment of producers and traders in different Member States or distort or impair the functioning of the common market, it would be for the competent Community institutions to adopt the provisions needed to remedy such disparities."

As my answer to the second question is essentially negative, I could dispense with the need to consider Question 3 which is in two parts. However, as they lend themselves to some useful observations, I am inclined to examine both parts of that question.

The Amtsgericht asks in the first place whether the tolerance margin allowed for by the method set out in Regulation No 625/78 also applies to any other methods selected by the Member States. Once again the answer is in the negative. Every method has its own tolerance margin which is determined, as I have shown, by technical reasons peculiar to it. It follows that, to apply to one method a tolerance margin calculated in relation to another method would be tantamount not only to disregarding the discretion conferred on the Member States (supra, Section 3) but also to ignoring the aforesaid reasons and, in the final analysis, the purpose for which those reasons are taken into account, namely to verify that the product is free of whey.

If a Member State chooses to apply the Community method, then the opposite holds true. In that case, the exclusion or reduction of the applicable tolerance margin would raise, in relation to a product in respect of which the requirements are less strict, the standards tolerated under the intervention system, notwithstanding the fact that the purpose of the analysis is the same. Accordingly, a Member State which allows for a tolerance margin of less than 2% would lay itself open to the charge that its conduct is discriminatory to say the least.

In Question 3 (b) the Amtsgericht seeks a ruling from the Court on the lawfulness of a Member State's claim for repayment of the aid where inspections carried out following an initial favourable analysis by means of the Community method reveal that whey powder has been added to consignments of skimmed-milk powder. My impression is that this question exposes the mainspring of the dispute. Denkavit refused to accept the product supplied to it particularly for fear that the results of its own or other possible analyses might impel the German authorities to institute proceedings for recovery against it.

On that point the parties' positions coincide. The principle of legal certainty and the intolerable risks to which undertakings would be exposed if the recovery of the aid were allowed justify in their view a negative answer. However, that argument is unacceptable. As we have seen, the relevant provisions of Community law (Article 1 (a) of Regulation No 986/68 and Article 1 (2) of Regulation No 1725/79) leave no doubt whatsoever that skimmed-milk powder to which whey has been added is for that very reason excluded from the scope of the aid system. A claim for the recovery of what should not have been granted in the first place would therefore appear to be quite legitimate and in no way contrary to the rule of legal certainty.

That is not all, however. The Court has consistently held that in the absence of applicable provisions of Community law, the recovery of sums unduly paid by way of aid is governed by the law of the State concerned, subject only to the application of conditions which are no different from those laid down for the recovery of aids granted under national law and to the protection of the interests of the Community in its entirety (see the Milchkontor judgment, paragraph 19 et seq. of the decision, and the judgments referred to by Mr Advocate General VerLoren Van Themaat in his Opinion on that case). In conclusion, in this case the solution of the problems surrounding the claim for repayment and the protection of the debtor is a matter for German law.

The final question seeks to ascertain whether Article 1 (2) of Regulation No 1725/79 is contrary to the principle of proportionality, inasmuch as it precludes the grant of aid in respect of a product that may be offered for sale to the intervention agency.

In my view, in this case too the answer can only be in the negative. We know that the intervention system is no less strict than the aid system in prohibiting the admixture of whey to skimmed-milk powder. Hence, far from jeopardizing the principle of proportionality, the differences in treatment to which the aforesaid provision may give rise are justified by the supposedly higher degree of accuracy of the method which the Member State concerned has selected for the purposes of the grant of aid.

In the light of all the aforesaid considerations, I suggest that the Court answer the questions referred to it by the Amtsgericht Koeln, by order of 7 December 1984 in the proceedings between Luetticke and Denkavit, in the following terms:

(1) Regulation No 804/68, taken in conjunction with Regulations No 625/78 and No 1725/79, must be interpreted as meaning that skimmed-milk powder which meets the quality requirements for intervention and public storage may be considered eligible for aid provided the Member States have not decided to verify that the product is free of whey by applying a method of analysis other than that prescribed by the rules of Community law and provided the method which they use is observed in every respect.

(2) Verification that no whey is present in the skimmed-milk powder by means of the method of analysis described in Annex IV to Regulation No 625/78, having regard to the tolerance margin laid down therein, does not mean that the skimmed-milk powder in question may also be regarded as free of whey for the purposes of the grant of aid under Regulation No 1725/79, if the Member States have selected different methods of analysis with smaller tolerance margins.

(3) (a) The fact that the method of analysis described in Annex IV to Regulation No 625/78 in connection with the intervention system allows for a tolerance margin of 2% does not mean that a similar tolerance margin is applicable to the results of analyses carried out by the Member States for the purposes of the grant of aid, where the methods applied by those States differ from the Community method and may provide for smaller tolerance margins.

(b) If it may be concluded on the basis of an analysis carried out by means of the method described in Annex IV to Regulation No 625/78 that no whey is present in a consignment of skimmed-milk powder, that does not preclude the aid granted from being reclaimed by the competent national authorities where it is established on the basis of subsequent, more accurate findings that whey powder has been added to the consignment in question.

4. In so far as it provides that skimmed-milk powder is ineligible for aid where whey has been added to the product, Article 1 (2) of Regulation No 1725/79 does not contravene the principle of proportionality because the intervention system also requires the product to be free of whey and any differences in treatment can arise only from the different degree of accuracy of the method of analysis adopted for the purposes of the grant of aid.

(*) Translated from the Italian.