OPINION OF MR ADVOCATE-GENERAL ROEMER

DELIVERED ON 21 NOVEMBER 1967 ( 1 )

Mr President ,

Members of the Court,

The reference for a preliminary ruling on which I am asked to give my opinion has to do with the interpretation of certain provisions of the EEC Treaty.

The facts are as follows.

Three contracts were made in the course of the year 1963 between the parties to the main action, Brasserie de Haecht, a ‘société anonyme’ under Belgian law having its registered office at Boort-meeribeek, on the one hand, and the proprietors of a café in Esneux (Belgium) on the other. The agreement made therein was that the Brasserie was to lend equipment for use in the cafe, and two sums of money; in exchange the cafe proprietors undertook to obtain their supplies, both for their personal needs and for the purposes of their business, exclusively from the Brasserie. There was a condition that failure to observe the exclusive dealing clause was to entail rescission of the contract for which the borrowers would be liable, besides damages for breach of contract. According to the court making the reference, the obligations were to extend over a minimum period of 15 years, whereas in the course of this hearing the Brasserie de Haecht has claimed that under Belgian law this period cannot exceed 9 years. However, this minor point seems to be of little importance in our case. The crucial point is that the proprietors of the cafe appear to have failed in their obligation to obtain supplies exclusively from the Brasserie de Haecht (to be brief, let us call it the exclusive purchase condition). As a result the brewery brought an action before the Tribunal de Commerce, Liège, seeking the annulment of the various loan contracts against the defendants, the return of the borrowed equipment and payment of the damages agreed on.

The defendants in that case relied in their defence on the Community law on restrictive agreements: they claim that the contracts which they made are void because they may, within the meaning of Article 85 of the EEC Treaty, ‘affect trade between Member States’. In deciding whether this is so consideration must be given not exclusively to the three contracts in question, but also to the fact that similar agreements are in existence between the brewery and a large number of other customers, as well as between other Belgian breweries and their customers, preventing the customers from purchasing beer from other Member States under threat of severe penalty.

Since extracts from legal writings were given to support this interpretation, the Tribunal de Commerce considered that it should not be passed over in silence but that the proceedings should be suspended and the following question referred to us for a preliminary ruling, by a judgment of 8 May 1967:

‘In order to judge whether the contracts in question are prohibited by Article 85 (1) of the Treaty of Rome, is it necessary to take into account the economic context and the whole of the market, that is to say, in this case, the simultaneous existence of a large number of contracts of the same type imposed by a small number of Belgian breweries upon a very large proportion of liquor licensees or must consideration be limited to an examination of the effects on the market of the said agreements considered in isolation?’

The parties to the proceedings in the national courts and the Commission of the European Economic Community have presented written observations under Article 20 of the Statute of the Court of Justice of the EEC. In addition, the parties representatives explained their positions in detail, quoting academic opinion at times in their support. This has lent the reference an unexpected degree of significance, bearing in mind its comparatively modest origins in the national proceedings.

Let us now see how these viewpoints can be analysed, and what help in interpretation can foe sought from the Court of Justice in a case like the present on the implementation of the Community law regarding restrictions on competition.

Reply to the question referred

To begin with, I should like to remark that a situation seems to me perfectly possible in which a single long-term loan contract made between two firms in one Member State would fulfil the conditions in Article 85 (1) of the EEC Treaty, that is to say, it might have a noticeable effect on trade and competition between Member States, if the retailer is forbidden to purchase competitors' products. That may be the case where the product traded in is complex technical equipment, for which only one middleman with the requisite technical knowledge is available in the Member State concerned. If he is tied exclusively to one product, then foreign producers who are unable to seek out suitable middlemen in a short time might be prevented from having an effective turnover in their product in that Member State for a considerable period. General opinion holds that such cases only form rare exceptions in any case, and they certainly do not include — as the court making the reference agrees — contracts for the supply of beer containing exclusive sales clauses. That is why the question referred to us is whether, in considering contracts in the light of Article 85 of the EEC Treaty, it is necessary to take into account the simultaneous existence of a number of similar contracts made by a brewery to build up its distribution network or by several breweries in the same Member State.

If we enquire whether general consideration of several restrictive agreements is required by Article 85, it must be admitted that it does not expressly provide the answer. Of course it seems obvious in the context of cartel law to examine contracts first of all in isolation, and indeed there is some indication in Article 85 that this procedure would be correct. This applies in paragraph (1) to the term ‘have as their object’ [‘bezwecken’] (here: the restriction of competition), which depends on the intention of the parties, on something over which they exercise control; it applies, too, to the negative exemption provided for in subparagraph (3) (b), which says that the undertakings concerned are not to be afforded the possibility of eliminating competition in respect of a substantial part of the products in question. (This is apparently directed at undertakings which are parties to a single contract). On the other hand paragraph (1) mentions only action which ‘may’ affect trade between Member States, and the possibility that the effect may be to restrict competition, in other words the terms used are such as to allow objective assessment and not to exclude consideration of factors outside the contract in question, that is, agreements of a similar nature as well. Nor does this exemption require the existence of any causal connexion between the contract being judged and the effects listed in Article 85, for naturally the causal factor is not absent from the individual contract if it is determined that the effects relevant for Article 85 only occur in conjunction with other contracts.

A deeper investigation as therefore called for if a solution is to be found to our problem.

This the parties have done in the proceedings by referring to our previous judgments on the cartel law of the EEC Treaty. Here the question at once arises whether this method will lead us to a more certain conclusion because of course both parties to the main action adopted it to support their conflicting interpretations of the law. The judgments in question are the Grundig case and the reference for a preliminary ruling in the Maschinenbau Ulm case but not — as the plaintiff in the main action thinks — the judgment in Joined Cases 8 to 1.1/66. It is true that the last contains a sentence which states that it must be expressly found ‘that the specific case in point contains all the constituent elements prescribed in Article 85 (1)’ (it requires ‘la constatation explicite que l'espèce, considerée dans son individualite, reunit toutes les conditions prevues par Particle 85, paragraphe 1’.) But, on the facts of the present case, it is quite clear that the above remark was not directed at the problem which we are discussing here, so that mat judgment should in any case be applied only with circumspection, even if general theories or statements on related problems make it relevant.

The plaintiff in the main action relies principally on the Grundig judgment, with the claim that there the Court of Justice placed the main emphasis on the purpose of the contract against which judgment was given, refusing to examine the actual consequences for the market (in respect of competition between the various products). It would not be right to conclude from this that the Court of Justice felt that it could not contemplate any other judgment, for, obviously, if the purpose of an agreement does not make it possible to arrive at a decision with regard to the law on cartels, then according to Article 85 regard must be had to its effects. Similarly, no conclusion can be drawn and applied to our case from the statement in the Grundig judgment that the whole distribution network must be examined (which was a reference to the need to take into account export restrictions imposed on other wholesalers). It would be wrong to conclude from that that Article 85 permits only a comprehensive appreciation of integrated networks with a multiplicity of agreements based on the same animus contrahendi, that is to say, which are inter-dependent, if only because in the Grundig case the question how to evaluate a great number of parallel agreements made by different undertakings which were not interrelated did not arise.

But we may find it more helpful to consider certain points made in the reference in the Maschinenbau Ulm case, although none of the legal guidelines there is completely unambiguous. That judgment contains indications — so the plaintiff in the main action finds — that the individual agreements should be considered in isolation; in the statement, for instance, that (it may be assumed that) ‘die fragliche Vereinbarung’ [the agreement in question] might impede the realization of a single market between Member States; when it is emphasized that it depends on ‘whether it (the agreement under consideration) is capable of bringing about a partitioning of the market in certain products’; when it is said that ‘the interference with competition referred to in Article 85 (1) must result from all or some of the clauses of the agreement itself’; or when it is held that ‘the competition in question must be understood within the actual context in which it would occur in the absence of the agreement in dispute’. On the other hand the judgment also contains statements, on which the defendants in the national proceedings lay their emphasis, seeming to call for a more farreaching examination. These occur in references to an ‘economic evaluation’ (‘Wiirdigung wirtschaftlicher Zusammenhange’) in assessing the effects of an agreement, to ‘a particular factual situation’, to ‘the economic context’ in which an agreement is to be applied; or in the statement that for the requirements to be fulfilled ‘it must be possible to foresee with a sufficient degree of probability on the basis of a set of objective factors of law or of fact that the agreement in question may have an influence, direct or indirect, actual or potential, on the pattern of trade between Member States’. The wider examination for our problem is supported, finally, by the requirement stated in the judgment that there should be taken into account ‘the isolated nature of the disputed agreement or, alternatively, its position in a series of agreements’. This seems to be irreconcilable with the argument which holds that agreements in restraint of trade should always be considered in isolation, each agreement taken separately, although it is not clear quite how one is to understand the phrase ‘position in a series of agreements’: does it refer only to the interrelated distributive channels of a supplier, or to a large number of coexisting agreements not related in any way?

This last point only becomes clear when viewed in the light of the spirit and purpose of the EEC Treaty's provisions concerning restriction on competition. The following considerations should be borne in mind.

As we know, the purpose or the Community cartel laws is to prevent the establishment of the Common Market (which is brought about by both the application and the abolition of measures based on national sovereignty) from being impeded or delayed by agreements on competition which affect international trade and introduce private restrictions on trade when national ones are removed. To ensure that this end is fully achieved, and that no loopholes remain in the system, it may prove necessary to interpret the provisions on restrictive agreements generously, and this is one case where that should be done. The agreements which we are considering do not seem, when viewed in isolation, to be prejudicial to the Common Market in any way (and for that reason not included in the group exemption in Regulation No 67/67). It can also be stated that this assessment is hardly altered if the entire distribution network of one brewery is examined as a whole, since where many such distribution networks exist (in Belgium they number some 300) it would be difficult for a single one of them to have an appreciable effect on trade between countries. On the other hand it is possible for such an effect to occur as the result of the combined operation of all the internal beer distribution agreements in a Member State. If one takes into account their duration and the effective sanctions with which their contracts are provided, then two points are clear: first, that beer is normally sold through ‘tied’ houses (so that other sales are negligible), and second, that owing to the legal or factual circumstances it is difficult to establish more licensed premises. These show that the national market is widely insulated and that trade with other countries is thereby hindered, because foreign producers encounter considerable difficulties in trying to enter such a market. If one is compelled in such a situation, which, while admittedly rather hypothetical, is not inconceivable, to consider individual agreements in isolation and thus find that Article 85 of the Treaty does not apply then a dangerous loophole is revealed in the cartel laws, a loophole which the national laws on the subject may not cover (because they do not normally concern relations between countries), and one which cannot be filled by reference to Article 86 of the Treaty either, since that applies only to the abuse of a dominant position in the market, that is, where there is at least an oligopoly, not merely the adoption of a similar attitude by a number of separate enterprises.

I am therefore inclined to think that the statements in the Maschinenbau Ulm judgment and consideration of the purpose of the Treaty provisions on restrictive agreements should be interpreted in such a way as to allow the existence of similar agreements by other undertakings to be taken into account in judging restrictive trade agreements with exclusive purchase clauses. This opinion will, moreover, be reinforced when one takes a look at the national laws on cartels, in particular the German law prohibiting restraints on competition. I quote § 18 of that law as the provision most relevant to the present difficulty. It runs:

‘The competent authority in matters of restrictive practices may, with immediate effect or as from a later date which it shall determine, declare of no effect contracts concluded between undertakings relating to products …, in so far as those contracts … restrict the possibilities of one of the contracting parties to obtain supply of other products … from third parties … and thus have the effect of unjustly restricting the access of other undertakings to a market or in so far as the extent of such restrictions has the effect of seriously harming competition on the market of those products or of other products … The term “unjust” in the sense used above shall not apply to a restriction which is insignificant in relation to the possibilities of offer and demand which exist for the other undertakings.’

One point is worth noting here: the extract is to be found in the section entitled ‘Other agreements’ in the German cartel law, not in the section on domination of the market by undertakings (§ 22 et seq.), a position which must not be abused: It is also noteworthy that all the commentaries on this provision stress that it is not sufficient to examine a single contract under § 18: all the relationships binding retailers and suppliers must be taken into account (which appears to eliminate the difficulty in establishing the causal link between the individual agreement and the effects in question). Whilst the national cartel laws cannot be taken as binding on us, the above example is worth bearing in mind.

However, our final solution has not yet been firmly established. One or two objections remain to be dealt with; they highlight some pitfalls in the answer arrived at so far and cannot be disposed of with a word. It has been alleged that the result of taking into account the beer distribution agreements made by all the breweries in a Member State for the purposes of Article 85 (1) would be unreasonable uncertainty so far as the law is concerned, because it is impossible for the parties to any agreements to know the number and content of all the other agreements (including perhaps foreign ones) and so foresee the legal consequences of their own agreement. The degree of uncertainty is heightened by the fact that the restrictive agreements made by the breweries are subject to continual change: their increase risks encroaching on the territory of the cartel laws, whilst their reduction removes them from it. This uncertainty also makes it difficult to limit the spheres of application of the national cartel rules, on the one hand, and the Community cartel laws on the other, since only when the number of exclusive dealing agreements in a Member State has reached a certain level can they present a threat to trade and thus bring the Community laws into play.

One cannot ignore the difficulties which this presents, bur neither should they be overestimated.

First, it should be said that there must always remain a certain degree of uncertainty with regard to the cartel laws which cannot be eliminated, for they depend greatly on the market conditions which are, of course, constantly changing.

Next it should be borne in mind mat as regards certain serious legal consequences the uncertainty feared does not exist, even on a comprehensive objective examination, namely — as in tort and criminal law — to the extent to which there is also a subjective element in the activity giving rise to liability.

In fact the problem only arises with regard to the civil law consequences of nullity which, in the scheme of Article 85, results with retroactive effect in the absence of the conditions for exemption (whereas the German cartel laws are more elastic: § 18 provides for a declaration of nullity ex nunc or with effect from a future date). Meantime it is certain that this phenomenon (nullity in civil law owing to objective breaches of the law) is also to be found in economic law and that the problem would not be greatly reduced if one had to consider — the only possible alternative — the application of Article 86 of the Treaty in circumstances such as the present, for it is widely held that even where Article 86 is contravened the legal transactions concerned are void in civil law.

However, certain points made by the Commission cause the problem to lose some of its intensity. It claims that it is not true that the contracting parties have no possibility of foreseeing the possible legal consequences. There is in fact no need for them to know every detail of the various agreements, only to have an idea of the order of magnitude of their overall effects on trade and competition. It is perfectly possible to get such an idea — at least for the larger producers in a Member State, which will certainly be carrying out extensive market research. Once they realize that the sum of the agreements might approach the acceptable limits (a possibility more theoretical than probable in the field of beer distribution contracts, as I have said) then they would be wise to apply to the Commission in time for a possible exemption under Article 85 (3). The Commission, for its part, can be relied upon not only to be generous in applying this provision (for it knows that beer distribution agreements, with their various dements, are usually harmless, and are indeed the necessary agents of an efficient sales network); it will also, if it sees no possibility of exemption, say so in due course in the administrative proceedings and thereby reduce the uncertainty which we have discussed. It can also be relied upon in such a case not to proceed merely on a chronological basis (that is, dealing first with the most recently concluded contracts) but will (as in German cartel law) weigh against each other the interests of those involved, which means that it will tackle first those agreements which have the most serious effect on trade between countries. In view of this, it is precisely the parties which have difficulty in assessing the market (small breweries and retailers) which need least fear that their contracts will be affected by nullity under the cartel laws.

On the whole, men, I think that in spate of the difficulties feared, which have certainly not been fully overcome yet, Article 85 (1) can justly be interpreted so as to allow the existence of similar contracts made by other parties to be taken into account in judging national distribution agreements with exclusive sales clauses.

If one is basically in agreement with this conclusion, then there remains still a word to say in this effort at interpretation on the factors which will have a part to play in the general assessment. It will take into account the number of existing agreements, their duration, the volume of goods affected, their proportion to the volume of goods in free supply, the influence of the agreements on foreign products, the frequency of the agreements in particular geographical areas, a comparison with trade other-wise than through cafes, and its prospects, as well as the opportunity of establishing other cafes or buying up whole sales networks. Lastly it must take into account the size of the breweries making the agreements and the volume of trade affected by a single contract, because even should it appear that the limit has been reached beyond which trade will be affected (which for an agreement binding roughly 50 % of all Belgian licensed premises, bearing in mind the actual pr ogress of beer imports into Belgium, scarcely seems to be the case,) small breweries (like the plaintiff, whose sales do not even amount to 2 % of Belgian-produced been) and small retailers (like the two defendants with an annual turnover of 42 kl) would hardly be likely to find their contracts annulled.

Of course it will not be easy for a national court to assess all these factors. But here too, it is only a matter of assessing degree; for even if the national court gained the impression that the requirements for an application of Article 85 (1) were present it could not simply proceed to declare the beer distribution agreements void. According to Regulation No 17, Articles 4 and 6, the Commission can declare such contracts exempt even without being requested to do so, and this will have retroactive effect to the date on which they were made. In order to avoid conflicting decisions by the Commission and a national court it is advisable in such a case to suspend the proceedings in court and suggest to the parties that they obtain a ruling on the facts of the case with regard to the law on cartels from the Commission, which has the proper means of conducting such an investigation always at its disposal.

Summary

Taking all these considerations together, we obtain the following answer to the question referred:

When national beer distribution agreements with exclusive purchase conditions are being judged for infringement of the cartel laws and fall outside the ambit of Articles 85 (1) of the EEC Treaty in their character of individual agreements, regard must be had to whether similar agreements have been concluded with retailers by a supplier or other suppliers. In the process there must be taken into consideration, in particular, the number of such agreements, their duration, the volume of goods affected, how they compare with ‘free’ trade, and opportunities for opening new trade outlets. Before such a contract can be declared void, the court proceedings should be suspended and the opportunity taken to obtain from the Commission a decision in relation to the law on cartels under Article 6 of Regulation No 17.

( 1 ) Translated from the German.