Opinion of Judge Vesterdorf

Mr President,

Members of the Court,

Introduction

On a proposal by the First Chamber, the Court of First Instance, sitting in plenary session, decided on 16 November 1989 to designate an Advocate General in Joined Cases T-l/89 to T-4/89 and T-6/89 to T-15/89 (the polypropylene cases). I was subsequently designated by the President of the Court of First Instance to perform that function. Examination of the case-files and the impression given at the oral hearing show that the cases fully satisfy the criteria — namely the legal difficulty and the factual complexity of the case — for designating an Advocate General now laid down in the Court of First Instance's own rules of procedure.

The cases concern the Commission's decision of 23 April 1986 relating to a proceeding under Article 85 of the EEC Treaty (IV/31.149 —Polypropylene) (Official Journal 1986 L 230, p. 1). In that decision heavy fines were imposed on fifteen undertakings in the chemical industry for having, over periods of varying lengths from mid-1977 until at least November 1983, participated in an agreement and concerted practice whereby they formed a price cartel and introduced quota arrangements and other measures supporting the price cartel. In its decision the Commission further ordered the undertakings to bring the infringements to an end.

Fourteen of the fifteen undertakings thereupon brought proceedings claiming, with minor variations in the formulation of the individual applications, that the decision should be annulled or in the alternative the fines either cancelled or reduced. By order of the Court of First Instance of 25 September 1990, the fourteen cases were joined for the purposes of the oral procedure.

In addition to a number of objections of a procedural nature concerning the Commission's administrative handling of the cases, the actions brought raise a number of important questions on the interpretation of Article 85 of the EEC Treaty. These questions are dealt with in detail in all the pleadings and the way in which they are decided will be of considerable importance for the future application of Article 85. As the Court will recall, at the hearing the Commission even went so far as to express the view that Article 85 would have to be amended if the Court did not accept the Commission's fundamental arguments concerning the interpretation and application of Article 85. One of the most important points in the case is the question of the interpretation of the term concerted practice in Article 85, unless, of course, when everything is considered there are grounds for concluding that all of what the Commission alleges constitutes agreements within the meaning of Article 85. The question here is whether the Commission is correct in its view that a concerted practice within the meaning of Article 85 exists as soon as the concertation takes place or whether there must also be a practice within the meaning contended for by the applicants, in the sense that there must be a demonstrable attempt, by means of direct initiatives vis-à-vis the undertakings' customers, to put into effect on the market what has been concerted. Another major issue is the extent to which the much debated framework agreement can constitute a single agreement within the meaning of Article 85 if, on the one hand, there is no evidence to suggest that the measures which were subsequently taken were on the whole based on an agreement on future cooperation between the applicants, while, on the other hand, it may be assumed that a significant number of measures were put into effect, each one of which can be said to be based on agreements or constitute a concerted practice, which together form a pattern and which were systematized to a considerable extent. A third major problem is the question of each individual applicant's involvement. This question comprises two parts: on the one hand, there are the evidential problems and, on the other, there is the question of liability for cooperation or participation, or, as it was put in the course of the proceedings, the question of collective responsibility or of a collective infringement.

A number of the procedural objections recur in most of the cases and the main problems set out above are, with few exceptions, common to all the cases. The same is true of the question of penalties. I have thus decided to structure my Opinion in this way: I begin with a general section, in which I discuss the alleged procedural defects in the Commission's decision, the question of the interpretation and application of Article 85, and the general principles governing the question of proof of an infringement in cases of this type; in the second part, I examine each case individually with a view to establishing how the evidence stands in relation to the individual applicants; in the third part, I deal with the question of penalties.

The structural problems affecting the market in polypropylene in the years 1977 to 1983

As is apparent from many observations made in the pleadings and at the hearing, as well as from the Commission's decision, the present cases are marked by the very severe structural problems which beset the polypropylene market from the mid-1970s to the beginning of the 1980s. The problem was considerable overcapacity, which meant low profitability for producers and in many cases heavy losses in the production of polypropylene. As the Commission states in its decision (point 12), the very expensive production plant entailed such high fixed costs that reasonable profitability depended in large measure on obtaining a high level of plant utilization. Around 1977, after the expiry of a number of patents and in the light of very optimistic forecasts for growth in demand, seven new producers began to produce polypropylene in Western Europe causing production capacity to increase very significantly, but in the years which followed demand failed by far to match that increased capacity. In my view, it is not necessary for the purposes of the Court's decision to examine further the underlying reasons for the sudden overcapacity because, as will be seen later, I agree with the Commission that the question whether, how far or to what extent State aid was granted to some undertakings in the industry concerned must be immaterial.

When there arises in a market-oriented economy a situation of overcapacity in a particular sector, capacity must be reduced in the long term, all other things being equal. The Commission has indeed explained (in point 13 of the decision) that in 1982 discussions were held on this matter, which led to a proposal to reduce capacity. The Commission would not, however, give its approval to a crisis cartel suggested by the undertakings if there were to be unacceptable restrictions on competition such as price or quota-fixing.

If there is overcapacity on a market, the existing undertakings, confronted with new competitors entering the market, will normally either try to out-compete the newcomers, stop or reduce production themselves, or try to reach an arrangement with the new undertakings. It is hardly surprising that undertakings experiencing a situation such as that which prevailed on the polypropylene market in 1977 should consider steps to avoid devastating price competition, which could jeopardize the very heavy investments in plant they have made, by reaching a mutual arrangement in order to survive for the time being pending the advent of better times. In the present cases, a certain reduction in production capacity did take place, on a more or less voluntary basis, but there was no decisive reduction in capacity and the crisis in the industry lasted at least six years. Thus, the crisis lasted a long time, perhaps much longer than the undertakings involved had expected. This was probably due in particular to the direct and indirect effects of the second big oil price rise in 1979 and the subsequent sharp rise in the price of the raw material, propylene, together with the general economic downturn which resulted in reduced demand.

Those considerations prompt reflection on one of the fundamental problems of competition law, namely the question of the extent to which, and within which limits, undertakings which are normally in competition with one another may lawfully cooperate in order to defend their common interests, either in a systemized way through trade associations or similar organizations or in specific instances, when special problems arise. On the one hand, there is no doubt that undertakings, just like workers, enjoy the right of association and are entitled to defend their commercial interests against the State and others; on the other hand, it is equally unquestionable that there is an important public interest in ensuring that competition on the common market is not thereby distorted: see Article 3(f) of the EEC Treaty. A balance must therefore be struck between, on the one side, the interests of the undertakings involved, and, on the other side, the aim of ensuring that consumers and other parties on the market are not exposed to unreasonable prices or trading conditions.

In determining the possibilities open to undertakings for cooperation and identifying where the demarcation line between lawful discussions and unlawful agreements or practices lies there is thus a difficult course to be charted between Scylla and Charybdis. As Advocate General Sir Gordon Slynn has stated, the fact that an undertaking attends a meeting at which other undertakings reach agreements that will distort competition does not by itself amount to its having participated in an agreement or in a concerted practice. The undertaking's representative might have attended the meeting in the belief that there were no dubious subjects on the agenda, and, as stated, undertakings must be allowed a certain freedom to discuss common problems. However, any attempt to undermine one of the market's most essential functions, namely the free formation of prices, must be fiercely resisted, and any undertaking which, over a lengthy period of time, sends members of its staff to numerous meetings at which measures are discussed which, from the undertakings' point of view, may perhaps be comprehensible in view of the industry's difficulties but which are difficult to reconcile with Article 85(1), is, in my opinion, in a weak position.

As the Court of Justice held in the Ziichner judgment, the requirement of independence in Community competition law strictly precludefs] any direct or indirect contact between ... traders, the object or effect of which is to create conditions of competition which do not correspond to the normal conditions of the market in question, regard being had to the nature of the products or services offered, the size and number of the undertakings and the volume of the said market. On the other hand, however, competition law does not, of course, preclude contacts per se between undertakings.

I — General section

A — Issues arising from the administrative procedure

All the applicants have claimed with varying emphasis and in various combinations that the contested decision is vitiated by a number of procedural defects. In evaluating this claim it may be useful to look first at how the Commission conducted the administrative procedure in the present cases.

1. The course of the administrative procedure

As is apparent from the decision, on 13 and 14 October 1983 Commission officials carried out unannounced investigations pursuant to Article 14(3) of Regulation No 17/62 at the premises of ten producers of polypropylene supplying the Community market (Atochem, BASF AG, DSM NV, Hercules Chemical NV, Hoechst AG, Hüls AG, ICI PLC, Montedipe,* Shell International Chemicals Co. Ltd. and Solvay and Cie). Fines were subsequently imposed on those undertakings, which have brought actions against the Commission. In addition, an investigation was carried out at the premises of BP Chemie in Paris. BP is, however, not covered by the contested decision because the Commission did not consider that it had sufficient evidence against that undertaking (point 78, last sentence, of the decision). After the investigations, the Commission requested information from the abovementioned undertakings pursuant to Article 11 of Regulation No 17/62; five other undertakings (AMOCO, Chemie Linz AG, Petrokjemi AS, Petrofina SA and ANIC Spa) were also asked to provide information. Of those five undertakings, three have brought actions against the Commission; one undertaking, SAGA Petrokjemi AS & Co., has paid the fine imposed; finally, there is the undertaking AMOCO, against which the Commission did not consider it had enough evidence.

Commission officials subsequently carried out investigations pursuant to Article 14(2) of Regulation No 17/62 at the premises of ANIC and the selling agents of Chemie Linz in the United Kingdom and Germany. No investigations were carried out at the premises of Rhône-Poulenc, which is nevertheless covered by the decision.

On 30 April 1984, the Commission decided, upon its own initiative, to open the proceedings provided for in Article 3(1) of Regulation No 17 and in May 1984 it sent to the fifteen undertakings a written statement of objections pursuant to the regulation. This covered inter alia all the undertakings which have now brought proceedings, apart from ANIC and Rhône-Poulenc. The statement of objections was divided into a general part and a part specifically directed to each individual undertaking.

It appears from the files that all the undertakings involved at that time thereupon requested an oral hearing pursuant to Article 7 of Regulation No 99/63 (Regulation No 99/63/EEC of the Commission of 25 July 1963 on the hearings provided for in Article 19(1) and (2) of Council Regulation No 17).

In June 1984, the Commission granted the applicants access to files. This was done in a way which has given rise to some controversy in the proceedings and with which some of the applicants are not satisfied.

On 24 October 1984, a meeting took place preparatory to the oral hearing, which was to begin on 12 November 1984, between, on the one hand, the Commission official, Roland Monssard, whose task it was to conduct the hearing, and, on the other hand, the undertakings' legal advisers. It appears from a note produced by Shell that on that occasion the official in question made two statements, one of which is relied upon by some undertakings in support of their case, the other being relied upon by Commission. First, he is said to have stated that a representative of each individual undertaking had to be present at the hearing (in other words, that the matter should not be left exclusively to outside Counsel) and that, in his view, it was safer for the company's representative to be a member of the commercial service. The second statement concerned his own role. He explained that he was only expressing a personal view and that it was moreover not very important. He went on to explain that there was some disagreement between the Commissioner responsible and himself about his role in the conduct of competition cases, or, in other words, that there was disagreement over the interpretation of his terms of reference.

That was ostensibly the reason, given in a letter of 30 October 1984 sent to the Commission by Shell's legal adviser, for Shell's declining to take part in the oral hearing before the Commission.

As a result of arguments advanced by the undertakings in their replies to the statement of objections, on 31 October 1984 the Commission sent to the legal advisers of the undertakings a bundle of documents. Among those documents were copies of price instructions given by the undertakings to their commercial services together with summaries of the documents. The lawyers of a number of undertakings refused to accept the conditions upon which the documentation was made available and returned it before the oral hearing. In particular, the lawyers were unhappy with the Commission's requirement that the documents were not to be shown to the commercial services of the undertakings.

The first session of the oral hearing took place from 12 to 20 November 1984. At that session, several undertakings refused to go into the material sent to them on 31 October 1984. They asserted that the Commission had altered its arguments and that at the very least they should have the opportunity to reply in writing; in addition, they claimed that they had not had sufficient time to acquaint themselves with the documents concerned before the hearing began.

In a joint letter of 28 November 1984 from BASF, DSM, Hercules, Hoechst, ICI, Chemie Linz, Montedipe, Petrofina and Solvay (with which Hüls associated itself by a letter of 4 December 1984), the lawyers of those undertakings were highly critical on two points: first, they claimed that the material that had been sent to them on 31 October 1984 introduced new issues, both legal and evidential. The undertakings then argued that they should have sufficient time to analyse the material and submit written observations and then should be given the opportunity to deal with the issues at a new oral hearing. In addition, the undertakings objected to the condition that their commercial services should not have the material disclosed to them. Secondly, the undertakings claimed that after the first round of the oral hearings there was an increasing and significant lack of clarity in the legal view adopted by the Commission and against which the undertakings had to defend themselves. In the undertakings' view, the origin of that uncertainty was the considerable lack of clarity in the statement of objections. Finally, they were of the opinion that the uncertainty was further increased at the oral hearing. It was alleged that various Commission officials had attempted to clarify the statement of objections orally and that those explanations were difficult to understand and appeared inconsistent in themselves and with each other.

On the first point, the undertakings claimed more specifically that from the text of the statement of objections and the material sent to them later it was impossible to establish which part of the documentary evidence was relied on to support the Commission's various findings: thus, it was not possible to establish on the basis of all the material exactly what the Commission alleged against each individual undertaking. As will be seen, that is a question which will recur later on in the case.

In addition, the undertakings raised the question of the meaning of the expression used in the subsequent letter of 31 October 1984, which referred to a representative sample of price instructions and so forth. In this connection, the undertakings expressed their fears about the use to which the Commission might decide to put the remaining documentary evidence that had not been sent. The undertakings considered that conduct to be contrary to the judgment of the Court in the AEG case. With regard to that first point, the undertakings insisted that the Commission should either identify all further allegedly similar price instructions on which the Commission intended to base its case and the documentation on which such a comparison was founded or confirm that it would not rely on such non-identified comparisons in connection with a decision. They further insisted that the Commission should remove the prohibition against showing the material in question to the undertakings' commercial services. Finally, they asked to be given the opportunity to comment on the new material both in writing and at a new oral hearing.

With regard to the second point, the undertakings raised the question of the extent to which it is necessary for the Commission to state precisely in each case what it regards as an agreement within the meaning of Article 85 and what it regards as a concerted practice. This question later developed into one of the main issues in the case. Even by that stage the undertakings were putting forward the view, subsequently to become their main argument, that the concept of a concerted practice has an objective element, a practice, whereas that is not the case where the concept of an agreement is concerned.

As was also to be asserted later, the undertakings took the view that the statement of objections was extremely unclear on that point and that the lack of clarity had become even more pronounced at the oral hearing; they therefore asked that the Commission clarify its points of view.

In response, on 29 March 1985 the Commission sent a new set of price instructions and tables to the undertakings, together with a summary of the evidence available for each price initiative for which documentation existed. However, it was essentially the same material which had been sent on 31 October 1984. In the same connection, the Commission lifted the prohibition against disclosing the documentation in question to the commercial services, gave the undertakings the opportunity to reply in writing and indicated that it was ready to hold another oral hearing. In addition, on the same day the Commission sent a letter to the undertakings responding to their arguments that the cartel that the Commission believed existed had not been precisely defined.

The Commission's view regarding the relationship between an agreement and a concerted practice, as notified to the undertakings in the letter of 29 March 1985, can be summarized as follows: over a long period the majority of polypropylene producers had agreed at regular and institutionalized meetings to set target prices, target volumes and target quotas and to take various measures to give effect to those plans. After the meetings the agreed plans were implemented by means of instructions given to the producers' commercial services. As detailed notes taken at the time of the meetings show, the degree of consensus attained in the matter of prices and volumes is a ground for finding that the prohibited cooperation between the participants in the meetings may be considered equivalent to an agreement or a number of agreements within the meaning of Article 85(1) having as its or their object or effect the restriction of competition. That finding applies even if, given the nature of the agreements, it was not possible to commit them to paper or to enforce them by legal means. In some cases, the arrangements resulting from the cooperation may have the character of both an agreement and a concerted practice, particularly when they are complex, comprehensive and subject to continual alterations. Examples of these can be found in the Fedetab case, in which the Court saw no reason to distinguish between the different forms of prohibited conduct. Even though some of the arrangements made by the producers who took part in the meetings do not, in the present cases, all present the characteristics of a detailed agreement, those producers nevertheless adopted measures with the common objective of coordinating their commercial policy. The precise form assumed by the prohibited cooperation is thus only of subsidiary importance; the producers participated in an unlawful cartel, the various elements of which can all be grouped together under the heading agreement and the heading concerted practices at the same time. Thus, even if it were to be assumed that the producers' anticompetitive arrangements as a whole should be regarded as a concerted practice rather than as an agreement which is possibly combined with a concerted practice, the prohibition laid down in Article 85 would be applicable not merely to their uniform conduct on prices and sales but also, and all the more, to the preparation stage and in relation to the previous meetings. When this is borne in mind, arguments designed to prove that the facts which the Commission has complained of constitute a concerted practice and not an agreement or vice versa appear to serve no purpose.

As regards ANIC and Rhône-Poulenc, which were not involved at the first stage, the Commission believed that it had obtained sufficient information concerning these undertakings and so, on 29 October 1984, a statement of objections matching the objections sent to the other undertakings was sent to both companies. The oral hearing for those two undertakings and ICI took place at the second series of meetings in July 1985. At this second round of meetings, which took place from 8 to 11 July 1985 and on 25 July 1985, every undertaking took part except Shell, which as stated, had not taken part in the first round either.

The draft minutes of the meetings were sent to the Advisory Committee on Restrictive Practices and Monopolies on 19 November 1985 and to the undertakings on 25 November 1985. The final version of the minutes, including the undertakings' amendments and additions, was sent to them on 8 July 1986.

As became apparent during the hearing before the Court, the Advisory Committee dealt with the case without being in possession of the final version of the minutes, whereas when the Commission dealt with the case it was in possession of the draft minutes as well as the undertakings' replies and requests for amendments.

In connection with the publication of the contested decision the Commission held a press conference and issued a press release. In the press bureaux reports of the press conference it is stated inter alia that one of Commissioner Sutherland's staff had said that the Commission was of the view that as a result of the cartel's activity the price of polypropylene had been artificially raised by between 15 and 30% in the period between 1977 and 1982 and by 40% in 1982.

That course of the procedure has given rise to a large number of objections of a procedural nature against the contested decision. The most important of them will be dealt with here, in the first part of my Opinion, while the views put forward only by individual undertakings will be dealt with in the second part in the sections concerning the individual undertakings.

Among the many objections which the undertakings, which have now become the applicants, have described as procedural objections, a number clearly relate to the administrative procedure, such as the issue of access to files, whereas others lie on the borderline separating what may normally be classified as procedural questions from issues of substantive law, for example DSM's claim that the Commission has not taken sufficient account of the in dubio pro reo principle. Again, in other cases it appears that some applicants consider it advantageous to redefine as procedural in nature issues which are clearly substantive. This may be the cause for some uncertainty in the applicants' definition of the issues arising. It is indeed clear that it is not possible, either as a matter of logic or of law, to draw a sharp line between questions which must be classified as procedural and questions of substantive law. On the other hand, it is equally clear that a question must be determined regardless of how it is defined.

2. The applicants' main arguments

As the Court of Justice has had occasion to find in a series of cases, the principle of audi alteram partem is an absolutely fundamental principle in the administrative law of the Community, including its competition law. The undertaking concerned must thus be allowed access to the documents and other evidence which the administration intends to put forward against the undertaking and must be given an opportunity to comment on both the material on which the decision is based and the legal arguments on which the decision is founded.

In the present cases, the applicants have put forward a number of factual arguments which all have the common feature of evincing the undertakings' conviction that both the course assumed by the procedure and the decision itself did not afford them a proper opportunity to prepare their defence. One applicant also challenges the way in which the Commission deals with competition cases.

The applicants put forward the following principal arguments:

The first principal argument raises fundamental issues concerning the Commission's activity in competition cases, which prompts me to embark first of all on a examination of the general nature of such cases.

To some extent, two other matters occupy a place apart, having been previously dealt with under an Article 91 procedure before the Court of Justice. These are the question of access to files in relation to the Hearing Officer's report and secondly the documentary evidence which formed the basis for the press conference held in connection with the publication of the decision. Moreover, many of the questions concerning procedural defects raised in the present cases have arisen previously in cases before the Court of Justice.

3. The nature of competition cases

In various connections the applicants have directed rather general criticism at the Commission's handling of competition cases. This question touches upon one of the major difficulties which arises in the handling of competition cases and which to some extent is manifest in this instance in connection with the handling of the procedural objections and to an even greater extent in connection with the handling of the issues of substantive law. I have in mind the tension which can clearly be felt — perhaps even more so in the present cases than in any previous competition case which has come before the Court of Justice — between the procedural framework of the cases, consisting of an administrative procedure followed by judicial review of legality, and the substance of the cases, which all broadly exhibit the characteristics of a criminal law case. In many instances, the parties' submissions can only be understood with the help of the terminology and concepts used in criminal law and procedure.

The Court of Justice has held that the procedure for dealing with cases before the Commission is an administrative procedure. I would refer here to the judgment of 15 July 1970 in Case 45/69, Boehringer Mannheim v Commission [1970] ECR 153, at paragraph 23, where it was held that the procedure before the Commission concerning the application of Article 85 of the EEC Treaty is an administrative procedure, even where it can lead to the imposition of fines. This means, as far as the present case is concerned, that the Commission's decision was not unlawful notwithstanding the fact that the members of the Commission themselves did not take part in the hearings but left this task to officials pursuant to Article 9(1) of Regulation No 99/63.

Furthermore, the Commission is not a tribunal in the sense in which that term is used in Article 6 of the European Convention for the Protection of Human Rights. This point has been made several times by the Court of Justice and it has not been gainsaid by the bodies set up under the Human Rights Convention — see the decision of the European Commission for the Protection of Human Rights of 9 February 1990, in which that body addressed itself to a question which formed an offshoot from one of the Pioneer cases. However, as the Court of Justice expressly emphasized in the said judgments, the Commission is bound to observe the procedural safeguards provided for by Community law. It follows, on the one hand, that no institutional claims, that is to say claims relating to the established system, can be advanced concerning the Commission's handling of competition cases but that the Commission — and this is self-evident — must in any event respect not only the written rules but also the unwritten principles laid down in the case-law of the Court of Justice. In the Pioneer cases, the Court of Justice considered Article 6(1) of the European Convention for the Protection of Human Rights in connection with an allegation that the decision in those cases was unlawful because the Commission exercised the function of both judge and prosecutor, whilst in the Fedetah cases it addressed the question in connection with alleged infringements of a number of procedural rules. The arguments of the undertakings were not accepted in any of the cases.

In view of the fact — in my view confirmed to some extent by the judgment of the Court of Human Rights in the Öztürk case — that the fines which may be imposed on undertakings pursuant to Article 15 of Regulation No 17/62 do in fact, notwithstanding what is stated in Article 15(4), have a criminal law character, it is vitally important that the Court should seek to bring about a state of legal affairs not susceptible of any justified criticism with reference to the European Convention for the Protection of Human Rights. At all events, within the framework formed by the existing body of rules and the judgments handed down hitherto it must therefore be sought to ensure that legal protection within the Community meets the standard otherwise regarded as reasonable in Europe.

In this connection it is also important to emphasize that the written and unwritten rules in any given legal order must be considered in their entirety when it is a matter of determining whether the private party or parties concerned have had sufficient opportunity to prepare their defence. If, for instance, there is in principle no unlimited access to files as regards all material not in the nature of internal working documents in the narrower sense of that term — which does not exist in the administrative law of the Community — the requirements regarding the other legal safeguards must be tightened considerably, for they will then have to sustain the fundamental principle that the private party must have cognizance of all material of importance for the resolution of the case and that no further material exists which might be relevant. Precisely that problem was discussed in connection with the applicants' concern that there might be some unknown material proving the effects of the cartel as alleged at the press conference and that that material might have been made the basis for the decision in the case and in particular for fixing the fines. It is thus necessary to examine carefully whether the decision rests on such a safe foundation that the existence of relevant material of which the undertakings have not been appraised can be practically excluded.

4. The Commission's internal working procedures

(a) The internal organization of the Commission's work

Shell has contended both in the procedure before the Commission and in the proceedings before the Court that the Commission disregarded essential legal safeguards during the procedure. Arguing that the Commission failed to fulfil its duty to act fairly, Shell states that the Commission's working procedures should be so organized, first, to ensure as reasonably as is practicable that that duty can be observed and, secondly, to afford the undertakings involved in cases before the Commission as well as the public a reasonable degree of confidence in the Commission as an impartial adjudicator. In this connection, Shell claims that in dealing with cases pursuant to Regulations Nos 17/62 and 99/63 the Commission has a duty to adopt an objective and impartial position with regard to all the evidence and to listen to arguments put forward by the applicants which may cast serious doubt upon the provisional views expressed in the statement of objections. Shell expresses the view that the said objective can only be attained if there is a functional separation between the investigative stage and the prosecutorial stage within the Commission. According to Shell, there was no such separation at the time when the present cases were being investigated and assessed, at any event not before the time when the statements of objections were sent out. Shell cites a number of concrete examples from the procedure, which, in its view, demonstrate such a biased and unbalanced assessment of some of the facts that it would be warrantable to conclude that the working procedure in Directorate-General IV was in itself capable of affecting the decisions to the applicants' detriment.

After explaining the reorganization of the Directorate-General which took place in 1984 and 1985 and which, in the defendant's opinion, was apt to allay the criticism expressed, the Commission states that in this regard there are no rules anywhere stipulating how the Commission should organize its internal working procedures. Furthermore, the Commission denies that it is possible to trace back the origin of any mistakes or errors of judgment to a specific way of organizing its work. In its view, the examples advanced by the applicants must therefore be dealt with as a whole, together with the substantive issues. Finally, the Commission points out that more than twenty of its staff worked on the case.

In my view, the first thing which must be said in response to those points is that as a rule it is not possible in any individual case to conclude from possible mistakes or possibly poor work that there is something generally wrong with the way in which the Commission's work is organized internally. Even administrative authorities organized in the best way conceivable, incorporating all possible procedural guarantees, can make mistakes. Conversely, organizations whose internal working arrangements perhaps leave something to be desired can perform outstanding work. I therefore agree with the Commission that any errors made in dealing with the present complex of cases do not warrant the conclusion that the internal organization of the Commission was arranged in a way that it may be assumed from the outset that it leads to mistakes of the sort the applicant considers it has demonstrated.

On the other hand, as suggested in the previous paragraph, Shell is right in pointing out that generally problems may arise if the same administrative authority has such wide-ranging powers that, in addition to investigating and prosecuting, it may also impose fines of such considerable amounts as in these cases. According to the case-law of the Court of Justice cited above, it is, however, quite clear that Article 6 of the Convention for the Protection of Human Rights can provide no specific legal underpinning for the call for the Commission's work to be organized in a particular way. It would appear that Shell, in referring to the normal requirements of good administrative practice, objectivity and impartiality, with which no one can disagree and which Shell rightly considers should form the guidelines for dealing with competition cases, is seeking to establish a principle that administrative working procedures should be organized in a particular way; however, such a principle cannot be derived from the Treaty, from the rules laid down pursuant thereto for the handling of competition cases, or from any other source of law. Shell's submissions concerning the Commission's internal working procedures should therefore be dismissed.

(b) The Hearing Officer

In the early 1980s the Commission found it expedient that one person, enjoying relative independence in his official duties, should in future take charge of oral hearings when they took place pursuant to Article 9(1) of Regulation No 99/63. The Commission therefore created the post of Hearing Officer with effect from 1 September 1982; Mr Roland Mussard was appointed to the post and it was he who took charge of the hearings in the present cases. The creation of the post was announced in the Eleventh Report on Competition Policy. In the Thirteenth Report on Competition Policy, the Commission published the Hearing Officer's terms of reference, and it was over the interpretation of the Hearing Officer's powers under those terms of reference that there was allegedly a disagreement between the Hearing Officer and the Commissioner concerned, to which reference is made above in Section 1.

The relevant provisions read as follows:

There is no doubt that the powers conferred on the Hearing Officer pursuant to the articles cited in the terms of reference are well suited to affording the Commissioner responsible for competition matters the best possible foundation for adopting a decision, because he does not receive just one picture through the usual administrative hierarchy but also receives a written or oral report from a person who conducted the hearing independently of the department otherwise responsible. Thus, the Commission has itself established legal safeguards which must be regarded as being broader than the safeguards that the case-law of the Coun of Justice requires to be observed in administrative procedures or that otherwise result from the written legal code.

Once the Commission has chosen to adopt such rules, which are in the nature of legal safeguards, it can at any rate be argued that it is bound to observe those rules, even if it was under no legal obligation to adopt them. That must be particularly so where the rules are published for the benefit of those concerned and it is expressly stated that the purpose of the rules is to strengthen the legal safeguards for undertakings caught in the Commission's spotlight. As regards the case-law of the Court on the administration's duty to observe its own rules, I would refer in particular to the judgment of 30 January 1974 in Case 148/73 Ĺouwage v Commission [1974] ECR 81, in which the Court stated (paragraph 12, at p. 89) that even if an internal service directive has not the character of a rule of law which the administration is always bound to observe, it nevertheless sets forth a code of conduct indicating the practice to be followed, from which the administration may not depart without giving the reasons which have led it to do so, since otherwise the principles of equality of treatment would be infringed. In Case 81/72 Commission v Council [1973] ECR 575, the Court took the same view, in that case in relation to the Council acting in its legislative capacity.

The most important provision in this regard is Article 6, according to which the Hearing Officer can refer directly to the Commissioner concerned and give him his observations at the time when the draft decision is sent to the Advisory Committee on Restrictive Practices and Dominant Positions. However, there is nothing in the present cases to indicate that the Hearing Officer was prevented from submitting his observations to the Commissioner concerned before the case went to the Advisory Committee. Furthermore, with reference to Article 5, there is nothing to indicate that the Hearing Officer was prevented from putting his views to the Director-General. Finally, no duty can be read into Article 7 of the terms of reference obliging the Commissioner concerned to forward the view of the Hearing Officer to the Commission. Even though it is, of course, unfortunate that the Hearing Officer and the Commissioner concerned were not in full agreement on how far the rules went, there is, in my view, nothing in the present cases to substantiate the contention that the Commission disregarded the provisions which it had itself adopted.

Consequently, the submissions made by a number of the applicants to the effect that the Commission improperly restricted the Hearing Officer's powers must be rejected.

5. Access to documents

(a) The Hearing Officer's report

As appears from the documents, when bringing its action ICI sought the production of this report in a separate procedure before the Court of Justice commenced pursuant to Article 91 of the Rules of Procedure. Other applicants also requested the Commission to produce the report but their request was rejected and they then claimed that the failure to produce the report was a procedural defect. In the proceedings under Article 91, ICI submitted that there was a substantial difference between the contested decision and certain statements concerning the case made by the Hearing Officer at the hearing. In response, the Commission contended that there was nothing in the Hearing Officer's terms of reference requiring his report to be disclosed to the undertakings. The Commission takes the view that it is an internal document whose disclosure would create difficulties for the Commission because a rule or practice of disclosure would entail a risk that the Hearing Officer and the staff engaged on the case would refrain from expressing themselves freely during the administrative procedure.

The Court of Justice dismissed ICI's request by an order of 11 December 1986, stating that the Hearing Officer's duties were only advisory in character and that the Commission was in no way bound to follow them. Consequently, in the Court's view, the report of the Hearing Officer did not constitute a decisive factor which the Court had to take into account in carrying out its judicial review.

In the case now before us, the question concerns the extent to which the non-production of the Hearing Officer's report can be regarded as a flaw in the decision. The reason for the claim for access to files is that the applicants believe that the Hearing Officer's report contains views which diverge from those set out in the decision.

As the Commission has stated, it is not, however, the Hearing Officer's report which is the subject of judicial review in the present cases. The contested decision does not fall to be reviewed in the light of the report. It is neither surprising nor unusual if, in a large administrative organization like Directorate-General IV, there are differing views on such a large complex of cases as that under review now, both as far as the facts are concerned and with regard to the legal issues, of which some are notoriously problematical. Once the internal debate in the Commission on how a case is to be decided, including its collegiate process, is over, the decision is adopted and it is the factual and legal basis of that decision which is to be reviewed by the Court. In my opinion, it would also be unreasonable for the Commission to be required to produce a document having the character of an internal working document if the securing of its production was intended solely to show that one of the Commission's staff held, or had held, views on the factual and legal aspects of the case which differed from the views finally adopted. It is also significant that in the eleven of the twelve Member States (excluding the United Kingdom) in which undertakings involved in national competition cases are in principle entitled by law to access to files, an exception is generally made for what are called internal working documents such as drafts, notes or reports. Such documents will typically express provisional views, including possibly earlier, subsequently abandoned, assessments of the case, whose disclosure to the party or parties involved is, in my view rightly, not considered necessary or otherwise appropriate. The only reasonable ground which might exist for producing such a document was mentioned by the Court of Justice in the order cited above, in which it referred to the order of 18 June 1986 in Joined Cases 142 and 156/84, British American Tobacco and Reynolds Industries, namely if such a document might be capable of throwing light on the question whether there had been any misuse of powers. As the Court of Justice stated, an examination of the Commission's internal working documents with a view to determining whether the Commission's decision was influenced by factors other than those indicated in the statement of the reasons on which the decision was based would constitute an exceptional measure of enquiry. It would presuppose that the circumstances surrounding the decision in question gave rise to serious doubts as to the real reasons for the decision and, in particular, to suspicions that those reasons were extraneous to the objectives of Community law.

In the present case, there is no indication of any misuse of powers. The Commission was thus justified in refusing to disclose the report. The submissions made in this regard must therefore, in my opinion, be rejected.

(b) Internal working documents relating to the press conference

In its order of 11 December 1986 the Court of Justice also ruled upon a request by ICI for access to internal working documents drawn up for the abovementioned press conference. As is clear from the order, ICI took the view — relying on various statements in the press — that the Commission had taken account of factors not mentioned in the statement of objections and on which the undertakings, including ICI, had not therefore had any opportunity to comment; nor were those factors apparent from the decision now being contested. The Commission's response is that good administration requires that its officers should be free to set down their internal deliberations in writing without them falling automatically into the hands of third parties. As I mentioned above, in connection with the production of the Hearing Officer's Report, the Court of Justice in the said order found first of all that access to the Commission's internal working papers was an extraordinary measure of enquiry to be used only when there were serious doubts as to the real reasons for a decision. The Court went on to state that ICI's assertions, based upon statements in the press, to the effect that the decision was based on reasons other than those set out in the decision, were not borne out by the articles to which ICI referred. The Court stated further that press statements and interviews with Commission officials cannot be equated with the Institution's position as set out in the contested decision. The Court accordingly took the view that at that stage of the proceedings there were no solid reasons for it to consider that the decision was based on factors other than those set out in the grounds for the decision.

As was emphasized at the hearing, the Court of Justice in its order considered the issue only as it stood at the time when it made the order. It thus expressly left open the possibility that new factors could emerge which might make the matter appear differently and indeed at the hearing before this Court ICI's representative requested it directly to reconsider the question.

In my view, it is not surprising that the applicants felt very unsure upon hearing figures of up to 40% mentioned as the effect of the alleged cartel when its effect is not quantified in the decision. According to the legal viewpoint which I shall advocate later in my Opinion, the effects of a cartel on the market are of significance only for the assessment of the fines. But even in that limited perspective, the undertakings' fear that the figures mentioned at the press conference concerning the effects of the alleged cartel were used as a basis for the assessment of the fines can be readily understood.

Neither in the written procedure nor in the oral procedure is there to be found any conclusive explanation for what happened. The Commission's explanation at the hearing may bring us the closest to the truth of the matter: the official who held the press conference may have misinterpreted the graphs that appear in the annexes to the decision. It does nothing to enhance the Commission's reputation if its spokesman at a press conference makes statements such as that in question for which there is no foundation in the decision.

It is, however, a characteristic feature of press conferences that what is said may easily sound more categorical than is actually intended, and there is often a real risk of ill-considered statements. Therefore, in my view, it needs more than such statements to substantiate a suspicion of impropriety. In this connection, it must be emphasized that the press release did not contain the figures mentioned. It should also be pointed out, as the Commission did in its pleadings in the Article 91 procedure, that Table 9 appended to the decision, which sets out information in the public domain, does actually show rises of the order mentioned. Against that background, the fact that the Commission official in question may have indicated that the said price rises reflected the effects of the cartel is perhaps not totally incomprehensible. Nor do I believe that the press cuttings put forward in evidence contain anything relevant to the case either. That Members of the Commission may have had differing views on the fining policy to be followed has as little significance for the outcome of the case as any speculation about the results of the lobbying in which, according to the newspaper reports, the undertakings engaged.

The applicants have not been able to point to any, let alone new, factors to substantiate their suspicion that the Commission relied on the figures in question as a basis for fixing the fines or otherwise used them in reaching the decision in the case. In my view, therefore, there are still no grounds for asserting that the material prepared for use at the press conference should have been produced during the proceedings. I therefore suggest that the applicants' arguments concerning this question should be rejected.

(c) Other questions concerning access to files and the presentation of evidence to the undertakings

As mentioned above, sound administration and sound administration of justice require persons and undertakings liable to fines to be given full opportunity to defend themselves. This means that those concerned should be apprised of all the relevant material.

In Community law, however, there is, as a general rule, no unconditional access to files; see in particular the judgment of the Court of Justice in VBVB and VBBB v Commission.

The Court of Justice has established a different method for safeguarding the rights of the defence: the administration must ensure that the undertaking concerned must have been enabled to express its views effectively on the documents used by the Commission to support its allegation of an infringement. In VBVB and VBBB v Commission, the Court did however imply that access to documents will be ordered only if there are concrete grounds for believing that documents have been used of which the undertaking had no cognizance: this is in accordance with the orders referred to under (a) and (b) above. The Court further explained that it was not particular documents which were decisive in themselves but the conclusions which the Commission has drawn from them. Consequently, in the Court's view, an undertaking can rightly assume that a document not mentioned in the statement of objections will not have been used in the decision. It follows that such documents may not be used as evidence in any subsequent legal proceedings.

In its Eleventh Report on Competition Policy, the Commission announced that it intended to give undertakings access to files in competition cases. It stated in particular (at p. 30)

In the Twelfth Report, the following rules were set out (pp. 40 and 41):

In the Thirteenth Report, the Commission further explained how it intended to grant access to files. It stated inter alia (p. 63) as follows :

In its Eighteenth Report on Competition Policy (1988), the Commission described at page 58 the experience gained:

As will be seen, there are therefore two competing systems, both having as their purpose to enable undertakings to prepare their defence. That circumstance has undoubtedly helped to create some of the problems raised by the applicants in connection with their arguments that not all the documents in the case were communicated to them.

Most of the applicants contend that a number of the documents on which the Commission's decision was based were not communicated to them.

Principally, three categories of document: are involved:

There is also a fourth category, namely the documents which, in the Commission's view, do not concern the particular applicant and with regard to which the Commission in its defence expressly states that those documents, in so far as they are not put forward in evidence against the applicant, as a document do not concern it.

With regard to the first category, a number of applicants first of all make a number of points on practical matters, namely that the space and copying facilities they were allowed when granted access to the records were inadequate for the purpose of providing a proper opportunity to examine the very extensive case-file which was presented to them. The Commission denies that there were problems in that respect. The applicants take the view, however, that owing to those practical difficulties alone the Commission should have indicated which documents it intended to use.

However, considering that the applicants are undertakings which can secure every conceivable form of expert assistance and did in fact do so, it cannot be validly maintained without further explanation on their part, that practical problems hindered them from using that opportunity to acquaint themselves with the material to which they were given access.

The applicants further argue, however, that the access to files which they were granted was not sufficient to permit the Commission subsequently to use in evidence against them the documents to which they were given access in order to acquaint themselves with them. In support of this contention, they refer to the case-law of the Court of Justice cited above, according to which documents which serve as the evidential basis for a decision are to be expressly mentioned in the statement of objections or, if need be, in supplementary statements addressed to the applicants. The Commission, on the other hand, considers that documents to which the applicants had access in order to acquaint themselves with them may be used against the undertakings.

The arguments put forward clearly show the schism which has arisen on this matter and on which the Court of Justice has not yet had an opportunity to rule.

It can, however, be inferred from the case-law of the Court of Justice that there can be no unconditional obligation on the Commission to send without request copies of documents if those documents are in any event identified in such a way in the statement of objections that the undertakings have the opportunity to ask for copies. On the other hand, there is nothing in the case-law to suggest that the Court of Justice would be disposed to alter its view of the way in which an undertaking should be confronted with a document before it can be used in evidence against it. In that regard, the AEG case is illustrative; the undertaking concerned had to be presumed to be acquainted with the documents found on its own premises.

In the present cases, the system introduced by the Commission for giving access to files was in place at least one year after it was announced in a publication which must be presumed to have come to the attention of those concerned. On the other hand, those who had access to the case-files could hardly be expected to realize that the Commission would rely on the access it had granted to them as support for a contention that it should not be necessary to refer to a document, at any rate by sending it out. All things considered and for those reasons, I consider that the proper course in the present cases would be to exclude from the individual cases documents which were neither sent to the undertaking nor mentioned in the statement of objections or material sent at a later date.

With regard to the second group of documents, consisting of documents which the undertakings were sent but which are not ostensibly identified in the decision, the matter is somewhat different. It is clear that the undertakings obtained copies of them and the statement of objections itself served as a warning that the Commission proposed to make use of them in any eventual decision. The applicants were thus able to comment on the probative value of those documents, which, according to the judgment of the Court of Justice in the AEG case, is the reason why the document should be mentioned in the statement of objections. That naturally presupposes that from the statement of objections the applicants were able to determine with reasonable certainty what the Commission was seeking to prove. As will be seen in the following section, my view is that in its statement of objections, taken together with the letters of 29 March 1985, the Commission made it so clear what it was seeking to prove that by that time the applicants should have realized how they should regard the individual documents and comment on them. I therefore take the view that there are no grounds for excluding those documents from the case.

The documents which, through inadvertence, were not sent, that is to say the third of the abovementioned categories, should, in view of what has been stated, be excluded from the case, unless they are mentioned so clearly in the statement of objections or in the letters of 29 March 1985 that the undertakings were in a position to ask for copies. The documents in question must, however, be examined in relation to the question how far they could be regarded as being of importance for the undertakings' defence (see below the comments concerning the fourth group of documents).

The founh category is less problematical in one respect since the Commission itself states categorically that the documents were not used against individual applicants. Therefore, our task here is simply to establish whether the document was actually used in the case against the undertaking in question. However, the fact that the Commission maintains that a number of the documents do not relate to or concern a particular undertaking gives rise to problems in two other respects.

In the first place, problems arise with regard to proving the very existence of a cartel, including in particular an agreement or framework agreement. Under the sequence adopted by the Commission for leading its evidence, it first sought to demonstrate the existence of the cartel itself and then the individual applicants' participation. The cartel is purported to be proved by the appendices to the general statement of objections, that is to say by the 101 pieces of documentary evidence, and the participation of individual undertakings by the annexes sent to each applicant. If, therefore, the Commission did not send all the annexes to all the undertakings, thus dividing the production of evidence into two phases — according to the questionable logic that it is possible to prove separately the existence of a cartel, which can hardly consist of anything else than the individual undertakings' participation — then this is the situation from which the Court must begin to assess the evidence. It must be examined whether the very existence of the cartel can be considered proved on the basis of the 101 documents without taking into account the annexes sent only to the individual undertakings in conjunction with the individual statements of objections. In the converse case, it could indeed be claimed that the documents were used against all the applicants.

Secondly, the procedure followed by the Commission raises difficulties in relation to the question of access to exonerating material upon which the applicants have insisted during the proceedings. According to what is stated above, the said documents may not be used against the applicants, but they argue that the failure to give them access to the documents is a procedural defect and point out that the documents could possibly have contained something to their advantage.

The Commission describes, for instance, the circumstances surrounding the lack of access to files as follows (BASF, defence, p. 61, para. 3.2):

Besides showing that the Commission has deprived itself of an opportunity of strengthening other evidence which is perhaps not that strong, that quotation shows that the Commission in fact believes that it is its task in a combination of cases like this to decide which documents may be of interest for which undertakings for the purposes of their defence. It is understandable that, in the conduct of the procedure, the Commission perhaps focused sharply on the material which can form the basis for the decision, but it is less understandable that it was unable to foresee that it would run into serious problems by refusing to grant access from the outset and as a matter of course to all the documents except for those covered by Article 20 of Regulation No 17/62.

As was emphasized at the hearing before the Court, the Commission, too, considers that the case must be judged on the basis of an overall assessment of the weight of evidence. The applicants' view that they should also have had access to the documents used only against other undertakings thus appears justified. In those circumstances, I believe that it should be held that all the undertakings concerned should in principle have access to all the documentary evidence in a complex of cases like this where it is particularly necessary to be able to arrive at a finding on the basis of an overall assessment of all the facts and circumstances of the case.

It is difficult to deduce from the case-law of the Court of Justice any clear indication of its attitude to this question. There is a hint, however, in the AEG case (paragraph 24, [1983] ECR 3192), in which it is stated that AEG was justified in submitting the view that the Commission could not use a document when part of it had not been communicated to the company and that it was not for the defendant to judge whether a document or a part thereof was or was not of use for the defence of the undertaking concerned. The Court of Justice therefore held the document to be inadmissible in its entirety. It is thus established that it is not the Commission's task to assess what the undertaking can use for its defence. At first sight those statements do not appear particularly congruent with the rest of the Court's case-law according to which, as we have ascertained above, there is no general access to files and the boundaries of a case are formed by the documents which the Commission has used as the basis for its decision. On the other hand, in the relevant case-law the Court has not in any event expressly addressed the question of access to files in relation to documentary evidence which exists in a case but which, in the Commission's view, does not concern another applicant.

In those circumstances, it must be warrantable to conclude that the case-law of the Court of Justice is at all events not inconsistent with the view that the applicants ought also to have had access to the documents used against other undertakings.

It is therefore plain that to exclude such documents from the proceedings would not be the appropriate step, the applicants' declared objective being precisely to use those documents if possible in order to prove that they did not take part in any particular economic relationship or relationships. Furthermore, it would also be unreasonable to exclude on that ground documents which might be suitable as evidence against other undertakings.

Incidentally, it ought to be mentioned that the applicants must, of course, be entitled to rely on documents which they did manage to obtain, even if the Commission does not think that they concern the undertaking in question (note, for example, Shell's use of the Solvay document of 6 September 1977, of which Shell took a copy when given access to the records (defence in the Shell case, p. 69)). The documents cannot in any way be regarded as procured unlawfully.

If in a particular case documents exist which have not been communicated to the applicants when they should have been and which might be important for their defence, there are, as far as I can see, two possible courses. Either the decision must be annulled, on the principle that the Court's task is merely one of reviewing legality, if after an assessment of the documents' contents it must be concluded that they might have been of real importance for applicants other than those whom, in the Commission's view, the documents concern. Or the Court must undertake the task of specifically assessing the importance for Case Y of evidence found only in Case X.

At all events, however, it is necessary that the documents might have been of real and specific importance. In the present cases, none of the documents we have seen and which were not communicated to all the undertakings, for example the Solvay document just mentioned, were likely to alter the picture in any important point. In the light of the fairly clear and convincing evidence before the Court concerning most of the Commission's objections, it seems to me, furthermore, improbable that the (few) documents mentioned which were not put forward in evidence might be capable of affecting the overall assessment of the evidence to such a degree as to lead to a different result.

6. Alteration of the legal assessment in the course of the procedure

Pursuant to Article 19(1) of Regulation No 17/62 (and see Articles 1 and 2 of Regulation No 99/63), the Commission must give undertakings against which it is considering taking a decision pursuant to Articles 85 and 86 written notice of the objections raised against them. This defines the subject-matter of the dispute and it is on this basis that the undertakings concerned have the opportunity of making known their views (see Article 4 of Regulation No 99/63).

In the present cases, several of the applicants claimed that the legal assessment of the case was changed in the course of the administrative procedure. That submission, which was put forward with particular force by the applicants Hoechst, Hüls and Chemie Linz, will be dealt with below.

The argument that the Commission altered its legal assessment of the case during the procedure can be summarized as follows: according to the applicants, throughout the administrative procedure the Commission indicated that there was a whole series of infringements of Article 85(1), whereas only in its decision did it describe the situation as a single agreement or, as it is called in point 81(3) of the decision, a framework agreement. The applicants who put forward this argument rely on the wording of both the statement of objections and the Commission's letter of 29 March 1985. The Commission's response is to point out that from the time that the statement of objections and the letter of 29 March 1985 were issued it had depicted the case in a way which corresponded with the view expressed in the decision. It also points out that, according to the case-law of the Court of Justice, it is not necessary for the statement of objections on the one hand and the decision on the other to be completely identical.

The applicants' more specific argument that the alleged alteration of the legal case presented reduced their capacity to defend themselves can be resumed as follows: a framework agreement is necessarily and by definition anterior to individual agreements. A framework agreement is a legal act by means of which the parties to the agreement establish rules in advance which will subsequently be observed and put into effect in individual cases. A framework agreement rests, at least in part, on different legal and factual premises than continuous conduct. Whereas evidence of a framework agreement releases the Commission from the need to produce irrefutable evidence of individual agreements or other anticompetitive arrangements, a continuing infringement in the legal sense is an ex post facto synthesis of individual actions into a single act. In contrast to a framework agreement, a continuous infringement presupposes, however, evidence of an unbroken chain of individual agreements. Framework agreement on the one hand and continuous infringement on the other are thus mutually distinguishable concepts, both as regards their factual ingredients and their legal consequences. In the applicants' view, what they describe as the Commission's change of position suggests that as a result of the undertakings' objections the Commission recognized that the evidence of a continuous infringement was insufficient and that is why it now alleges that there was a framework agreement so that it can still maintain that a cartel existed during the whole period from 1977 to 1983.

Thus, the applicants' view is that in the decision a substantial shift in foundation took place, since they now have to defend themselves against what is described as a single continuing agreement within the meaning of Article 85(1) and against an overall framework agreement which was manifested in a series of more detailed sub-agreements worked out from time to time, which in Article 1 of the decision is expressed as participating... in an agreement and concerted practice originating in mid-1977. This is compared by the German-language applicants with the first paragraph of the statement of objections in which it is stated in German that the decision concerns eine Vielzahl [a large number] of agreements and/or concerted practices; they point out that throughout the statement of objections agreements, concerted practices,

infringements, and so forth are mentioned in the plural.

In response to those arguments the Commission refers first of all to the case-law of the Court of Justice, according to which the decision must not necessarily be a replica of the statement of objections. According to those judgments, the Commission must in fact take into account the factors emerging from the administrative procedure in order either to abandon such objections as have been shown to be unfounded or to supplement and recast its arguments both in fact and in law in support of the objections which it maintains. As the Court of Justice has stated, that possibility is not inconsistent with Article 4 of Regulation No 99/63.

In this regard, the Commission points out that it had already contended at various places in the statement of objections forwarded to the applicants on 25 May 1984 that there was continuing and institutionalized cooperation (see points 128 and 132). In its letter of 29 March 1985 concerning agreements and concerted practices, the Commission stated, moreover, that it was not excluding the possibility that there was a core agreement between the four largest producers and that the agreements as regards the other participants were based on a scheme that was sufficiently detailed to amount to an agreement or agreements under Article 85. In the Commission's view, the clarification thus given was sufficiently explicit to enable a proper defence to be prepared. In the administrative procedure there was enough opportunity for an extremely thorough discussion of the true nature of the cartel and in its decision the Commission drew the conclusions from that debate, just as the applicants have made the most of their opportunities to defend themselves.

It is hardly surprising that it is in fact three of the German-language applicants who have in particular put forward the argument examined here, for in the German version the first point of the general statement of objections was formulated slightly differently than in the other languages used in the case. Whilst in the English version the statement of objections is stated to concern a complex of agreements and/or concerted practices, in the French version un ensemble d'accords et/ou de pratiques concertées, in the Italian version un complesso di accordi e/o di pratiche concordate and in the Dutch version een geheel van overeenkomsten en/of onderling afgestemde feitelijke gedragingen, the German version states, as I have mentioned, that it concerns eine Vielzahl von Vereinbarungen und/oder aufeinander abgestimmten Verhaltensweisen (my emphasis). While in its written observations the Commission did not expressly mention this, it maintained at the hearing, with specific reference to the word ensemble, that it had alleged from the outset that it was the totality of agreements and so forth which constituted the infringement of Article 85(1) and not the individual agreements.

Whereas in the other languages one may reasonably construe point 1 of the statement of objections as referring to a sum total or a complex of agreements in the sense contended by the Commission, it does not seem that this can be the case in German, for the word Vielzahl can mean only a (large) number of agreements and so forth, and not the sum total of such agreements. The first step in the examination of this argument must therefore consist in seeking to establish how far the other parts of the statement of observations and the Commission's letter of 29 March 1985 gave the applicants such a good indication of what the matter was about that the word Vielzahl can be ignored; that the applicants should thus have realized from the other reasons and matters adduced that it was not a Vielzahl but a complex that was meant. If this proves to be the case, the question which must then be examined is whether a comparison of the statement of objections and the Commission's letter of 29 March 1985 on the one hand and the decision on the other hand otherwise gives such a reasonably uniform picture that it can be said that the legal assessment of the objections was merely recast or supplemented or, whether, as the applicants maintain, the grounds relied upon are entirely new.

In the statement of objections it issued, the Commission described chronologically and in great detail what it believed had taken place, and indeed that factual part of the general statement of objections has given rise to only scattered criticism in the context now under examination.

In point 127, one reads in the German version ..., ob sämtliche Regelungen und Massnahmen,..., whereas the other versions refer to a complex, l'ensemble or use similar expressions. In point 128 — all the language versions correspond here — the Commission mentions ...the continuing collaboration between the parties in the framework of the meetings.... In the last sentence of point 132, the wording is as follows : Effectively the producers were aiming to control the market and a continuing and institutionalized cooperation at a high level was substituted for the normal play of competitive forces. The relevant contents of the letter of 29 March 1985 is summarized above.

From the extracts cited here, which are crucial, I am satisfied — with reservations in regard to the German versions — that it must have been clear to the undertakings that it was the continuing and institutionalized cooperation which, in the Commission's view, fell under Article 85, not the individual elements per se taken separately. In the decision, all the versions refer to an agreement etc. as well as to a whole complex of schemes ..., Overall plan, a whole complex of schemes and arrangements, consensus on a plan and framework agreement.

The decision itself is not completely clear on the question as to how far it was directed at a prior agreement or at what some of the applicants call a continuous infringement, or, in other words, whether the Commission considers that the cooperation was such that it could be given the overall designation of a framework agreement or whether it considers that there was an agreement from the outset. This has brought forth strong criticism. However, the decision does give exactly the same general impression — although it is not perhaps a model of clarity — that it is the continuous and institutionalized cooperation as such which falls foul of Article 85(1).

With regard to the German-language versions of the statement of objections, the situation is perhaps slightly less certain but despite the somewhat imprecise formulations, it should nevertheless have been clear to the addresses of the statement of objections, considering who they were, that the Commission was not relying on the individual elements separately.

In the judgment of the Court of Justice in ACF Chemiefarma y Commission, the dicta cited above concerning the interpretation of Article 4 of Regulation No 99/63 are amplified to some extent, although it is not otherwise apparent from the judgment which differences between the statement of objections and the decision were in the question. In paragraph 94 of the judgment it is stated that Article 4 of Regulation No 99/63 is observed if the decision does not allege that the persons concerned have committed infringements other than those mentioned in the statement of objections and only takes into consideration facts on which the persons concerned have had the opportunity of making known their views.

In the van Landewyck case the Commission had mentioned in the second of two statements of objections only the first of the conditions for exemption provided for in Article 85(3), whereas in the decision it addressed itself to two of the other conditions in Article 85(3), and they were treated by the Court of Justice as two new objections (paragraph 70). In view of the fact that the applicants had already to a large extent given their views regarding all the conditions in connection with the notification, the fact that one of the conditions was expressly mentioned in the first statement of objections and the fact that the Commission had summarized in its decision the statements of the applicants regarding the said condition, the submission was rejected in so far as that condition was concerned because, as was stated inter alia, the two notifications had to be read as a whole. With regard to the second condition, the Court of Justice rejected the applicant's submission, on the ground that the content of that condition constituted the very basis of the second statement of objections — to which the applicant had replied — albeit in a context other than that referred to in the decision.

As I have stated, no-one is contending that the factual basis of the decision was different from that of the statement of objections. Consequently, the question is how far the Commission is entitled when recasting its arguments to alter the legal basis. The Chemiefarma judgment does not really appear to address that point. From the von Landewyck judgment probably all that can be concluded is that the addition of one or more legal factors of the same character as those on which the Commission has based its decision constitutes the addition of one or more objections in the sense of Article 4, the question of the alteration of the legal case not being addressed. The interesting thing about the judgment in von Landewyck, however, is that by specifically assessing the facts of the case the Court of Justice in fact examines how far the applicants actually had the opportunity to express their views on what later became part of the basis for the decision and how they availed themselves of it.

It would therefore seem necessary to examine specifically whether it may be considered that the applicants concerned in fact understood the statement of objections in a way consistent with the content of the decision. In its reply to the statement of objections, Hoechst states that the Commission alleges an agreement and/or concerted practice against the undertaking but the legal categorization is not discussed in relation to framework agreement or continuing infringement. Neither Hüls nor BASF make any statements which might point one way or the other. Chemie Linz, on the other hand, refers to point 1 in the statement of objections quoting the words eine Vielzahl von Vereinbarungen und/oder abgestimmten Verhaltensweisen, but without discussing the precise meaning of that expression. No clear deductions can therefore be made from that examination of the undertakings' replies.

The replies to the statements of objections show, however, that apart from quite sporadic legal discussions which do not relate to the present question, the applicants comment only on the facts held against them by the Commission and their replies show no indication at all of their being aware of the direction which their subsequent defence would take. Thus, there are no sufficient grounds for assuming that the way in which the statement of objections was formulated led the undertakings, when replying thereto, down a legal path entirely divergent from the path later followed by the Commission in its decision.

Despite there being no perfect congruence between the statement of objections and the decision, I therefore take the view, even as far as the German-language versions are concerned, that Article 4 of Regulation No 99/63 was not disregarded; on the contrary, the legal assessment of the case was properly clarified in the light of the replies to the statement of objections.

7. The minutes of the hearing before the Commission

A number of the applicants have alleged that the fact that the Advisory Committee had before it only the draft minutes prepared by the Commission when it delivered its opinion pursuant to Article 10 of Regulation No 17/62, and not the proposals submitted by the applicants for amendments of the minutes, constitutes a major procedural irregularity. Similar criticisms are put forward as far as the Members of the Commission are concerned when they adopted the decision.

The Commission has explained that the applicants' comments on the draft minutes were not available when the Advisory Committee delivered its opinion. However, the applicants' remarks were enclosed with the draft minutes when the case came before the Commission for a decision. For their part, the applicants do not dispute the Commission's explanation.

The Commission does not expressly deny that there was a procedural defect, but, with reference in particular to the judgment of the Court of Justice in Distillers Company v Commission, submits that what happened cannot be regarded as having affected the contents of the decision. The Commission points out that there is no period prescribed for sending out the draft minutes, just as, in the Commission's view, there are no rules stipulating to whom the draft and the approved minutes should be sent.

The Commission further points out that in the present case both the Commissioners and the Advisory Committee took the decision in full knowledge of the applicants' views. As regards the Advisory Committee, that view is borne out by the fact that representatives of all the Member States took part in the hearing before the Commission although Greece and Luxembourg attended only the 1984 meetings. According to the Commission, it is immaterial in this connection whether the same officials attended the hearings and participated in the Advisory Council's meetings. Finally, the Commission observes generally that the applicants have not claimed that the draft minutes did not contain a true record of the substance of the applicants' statements.

As Mr Advocate General Warner stated, it must be inferred from the content in particular of the provisions of Article 10 of Regulation No 17/62 and Article 9(4) of Regulation No 99/63 that both the Advisory Committee and the Members of the Commission should have to hand a final and approved version of the minutes of the hearing before the Commission when the Committee delivers its opinion or the Commissioners adopt the decision.

That view seems to find support in the judgments of the Court of Justice in Buckler & Co. v Commission and ICI v Commission.

It is true that the Court of Justice does not directly and explicitly address itself to the problem but rejects the objections raised in the cases cited by pointing out that, on essential points, there were between the final approved minutes and the draft minutes no discrepancies which were so great as to have been capable of misleading the Members of the Advisory Committee or the Commission with regard to the applicants' statements at the hearing before the Commission.

In the present cases, I think that it may be assumed from what we know, firstly, that the Members of the Commission did have the necessary bases for their decision.

Secondly, as far as the Advisory Committee is concerned, the question to be examined with reference to the case-law of the Court of Justice is how far in each individual case discrepancies can be found that might be regarded as having been capable of giving the Committee's members a mistaken impression of the statements of the individual applicants. Many of the applicants have made this general assertion without specifying where the discrepancies lay. In my opinion, that is not sufficient. The applicant concerned must be required to state expressly the essential points on which the final minutes differ from the draft minutes. Without a more detailed explanation from the applicants of the points on which, in their view, the provisional minutes are actually misleading, it is not possible for the Court to exercise its review function. In cases like this it cannot be the task of the Court to go through the provisional minutes and then the final version page by page with a view to determining what discrepancies there are so as then to consider what the applicants might have had reason to be dissatisfied with. Since none of the applicants have made their objections more precise, I consider it appropriate to dismiss these objections on the basis of the said procedural grounds.

B — The reasons on which the decision is stated to be based

1. The reasoning requirement in general

Most of the applicants have submitted that the Commission's decision is insufficiently reasoned. In this section a number of points which are essentially common to many or all of the applicants will be addressed. The question of the reasons given for the amount of the fines will be dealt with below in Part III.

According to Article 190 of the Treaty, acts of the Commission are to state the reasons on which they are based. The purpose of the reasoning requirement is not only to enable addressees of acts to ascertain whether the decision is materially correct; it must also serve as the basis for judicial review of the administration's decision. Furthermore, the requirement that administrative decisions should be fully reasoned may compel the administration to make clear for its own sake the reasons on which a decision is based.

As regards the duty to state reasons, the Court of Justice has held generally that the duty is to be regarded as fulfilled if the statement of reasons indicates clearly and coherently the considerations of fact and of law on which the decision is based. According to the judgment just mentioned, this also applies in the case of a decision imposing fines. The scope of the duty to state reasons depends largely on what has to be explained and must be considered in the light of the substantive law context. Thus, the fact that, for example, a wide discretionary power is available to the administration means that in the statement of reasons it must give a thorough account of the reasons on which the exercise of its discretion was based. The relative nature of the duty to state reasons is brought out especially clearly in the case of Usinor v Commission, in which the Court of Justice stated inter alia that the requirements to be satisfied by the statement of reasons depend on the circumstances of each case, in particular the content of the measure in question, the nature of the reasons given and the need for information of the undertaking to which the measure is addressed.

Apart from that, it may be difficult to be more specific about the requirements applicable in general to the content and scope of the statement of reasons. In a case concerning another field of law, namely freedom of movement for workers, the Court of Justice stated that where it is a question of securing the effective protection of a fundamental right conferred by the Treaty on Community workers, such persons must be able to assert that right under the best possible conditions and have the possibility of deciding, with full knowledge of the relevant facts, whether there is any point in their applying to the courts. In my opinion, the requirements relating to the statement of reasons in cases such as those now under examination, particularly in view of their vitally important nature, can hardly be less strict.

As stated above, the statement of reasons is the basis for judicial review under Article 173 (and see Article 168a). Those provisions provide for a review of legality, which the Court of Justice has carried out quite intensively in earlier competition cases. The fact that only a review of legality is involved is not altered by the acts forming the basis of the activity of the Court of First Instance. However, it is clear from the preamble to the Council's decision of 24 October 1988 that the very creation of the Court of First Instance as a court of both first and last instance for the examination of facts in the cases brought before it is an invitation to undertake an intensive review in order to ascertain whether the evidence on which the Commission relies in adopting a contested decision is sound.

That fact also has a secondary effect as regards the statement of reasons. Whatever the difficulties which will always be involved in giving expression to a finding when it is based on an overall assessment of an enormous volume of evidence, this must be done in the statement of reasons. This is a consequence of the rules applicable and the Commission must adhere to them.

In the area concerned here there is a particular danger of the argument shifting, so that it is attempted to some extent to bring questions which should normally be regarded as questions of substantive law within the ambit of the reasoning requirement. It is therefore important to stress that the requirement to state reasons, even though its scope is determined by the nature of the case, is purely procedural. So, if a statement of reasons is based on an incorrect legal view or on a wrong assessment of the evidence, this is not therefore a defect in the statement of reasons but, on the contrary, a defect in the legal and factual assessment on which, the decision in the case is based.

The question of the scope of the duty to state reasons has specifically given rise to dispute on the points dealt with in the section that follows.

2. The Commission's duty to comment in the decision on the evidence and arguments adduced before the decision is adopted

This question has been raised in a number of earlier cases. Briefly, applicants have taken the view, as they do in the present cases, that the Commission has neglected its duty to give adequate reasons for its decision if it does not address in the decision all, or at least the main, arguments and evidence which the undertakings have adduced during the administrative procedure. This view is asserted under various headings and in various contexts but the substance is the same.

It requires two different questions to be addressed. First, there is the question as to what need not be included in the statement of reasons under the law applicable. Then, it must be determined which positive requirements may possibly be imposed in this regard on the statement of reasons in order for it to satisfy the general requirements laid down in the case-law of the Court of Justice.

In the Consten and Grundig case, the Court of Justice stated very generally that in non-judicial proceedings of that kind (consisting of a competition case before the Commission) the administration was not required to give reasons for its rejection of the parties' submissions. In ACF Chemiefarma, the Court amplified that statement, stating in effect (in paragraph 76 et seq., [1970] ECR at p. 689) that in order to fulfil its duty under Article 190 to state reasons the Commission was not required to discuss all the points of fact and of law dealt with by the parties in the course of the administrative procedure. The Court went on to hold that the statement of reasons was to be considered sufficient if it indicated clearly and coherently the considerations of fact and of law on which the Commission had acted so as to acquaint both the addressee of the decision and the Court with the Commission's reasoning. Finally, the Court held that the Commission had not disregarded essential procedural requirements by omitting factors which it rightly or wrongly considered irrelevant to the proceedings.

In the van Landewyck case (paragraphs 64 to 66, [1980] ECR 3244) it was said of a specific statement of reasons that it contained answers to some of the applicant's answers but did not constitute a detailed refutation of them. However, the statement of reasons contained a self-sustained argument setting out in general terms why the Commission arrived at a specific view. The Court of Justice accordingly held that the Commission could not be required to comment on all the issues of fact and law raised by every individual undertaking.

In the case of GVL v Commission it is stated (in paragraph 12, at p. 500) that in its decision the Commission is not obliged to discuss all the observations put forward by the undertakings in their reply to the statement of objections if the reasons stated in the decision are of themselves such as to justify the conclusions at which the Commission arrived in the course of the administrative procedure.

From the case-law of the Court of Justice it can be concluded first of all that it is the Commission which has the absolute power to determine the matters to be discussed in the case. If the Commission considers that the legal situation is X and the matter to be proved Y, it has no duty to discuss legal situation A and matter B in its decision. The Commission thus has complete control over the scope of the case and is in no way obliged to deal in detail with evidence or arguments which, rightly or wrongly, as the Court says, it regards as irrelevant in relation to the scope of the case as so defined. Nor is the Commission bound to discuss arguments which may indeed be relevant but may give a self-sustained account explaining why it arrived at a specific conclusion if its explanation is in itself capable of sustaining the conclusion reached.

Of course, it may be in the Commission's interest to comment to some extent on the arguments and evidence put forward during the administrative procedure so as to obviate the allegation, as made in this case, that from the outset it has stuck to a particular position from which it would not subsequently move. The fact that such allegations fit awkwardly with the argument that the Commission changed its view in the course of the proceedings, which is put forward at the same time, is another matter.

As Advocate General Sir Gordon Slynn pointed out in the Hasselblad case, there may, however, be cases where two versions of the facts should be set out and analysed in relation to each other. This might be necessary in order to show that all relevant material has been included and that what the parties have adduced cannot lead to any other conclusion. Thus, in my opinion, there would be a defect in the statement of reasons if the Commission were to ignore evidence presented by the undertakings on a matter to be proved which the Commission, in view of the contents of the decision, also regarded as relevant. However, the same can hardly be true of the legal arguments. Normally, there is no reason to discuss in a specific administrative act the solution of a theoretical legal problem. It must be sufficient for the Commission to adopt a position towards the case as it stands and to give reasons of fact and of law for that position. Moreover, the fact that the Commission only summarily refutes the undertakings' criticism of its assessment of the evidence as that assessment appears from the statement of objections, is likewise unimportant provided that the contents of the decision can otherwise justify the conclusion reached.

Most of the Commission's comments on the views and evidence put forward by the undertakings during the administrative procedure are to be found in section E of the decision entitled The Commission's assessment of the producers' factual arguments', that is to say points 70 to 77. Read in conjunction with the rest of the decision, points E (a), (b) and (c) provide an excellent account of why the Commission does not consider the undertakings' arguments to be tenable. Even if under the law applicable the Commission could be required to refute the undertakings' arguments, in my opinion, this was done quite adequately. Point (c) contains a discussion of Professor Albach's study of the German market. Apart from the fact that, as emerged at the hearing, Professor Albach's conclusions are neither undisputed nor indisputable, that section of the Commission's decision gives an excellent explanation of why the Commission believes it can maintain its view regardless of the results of the market survey. In submitting these arguments the applicants seem almost to be saying that the Commission should yield to unprovable facts and should otherwise follow the undertakings' views. It is possible that the Commission's views on this matter are not entirely unshakeable but the fact that the Commission did not allow itself to be convinced does not constitute a defect in the statement of reasons. Quite identical arguments are valid as regards section (d) concerning the audit of the undertakings' net selling prices conducted by a firm of accountants. In my view, therefore, the Commission adequately commented on the relevant documentary evidence which the applicants submitted to it.

3. The individualization of the decision

It is contended that the decision is framed in such a way that it was impossible for each individual applicant to ascertain what was alleged against it. In particular, Rhône-Poulenc maintains that it has been victimized on account of the way in which the decision is framed because that undertaking, which left the market in 1980, was unable, so it claims, to ascertain on what the Commission based its decision as far as it was concerned. Rhône-Poulenc therefore considers that it is being held responsible for something which others did later. In my view, the procedural aspect of this issue can be regarded as forming part of the question as to how far the requirement to state reasons may be regarded as having been fulfilled in relation to each applicant.

In the Suiker Unie judgment, the Court of Justice stated that there was no reason at all why the Commission should not make a single decision covering several infringements, even if some of the undertakings to which it was addressed were unconnected with some of those infringements, provided that the decision permitted each addressee to obtain a clear picture of the complaints made against it.

Without giving more detailed reasons, the Court went on to state that the Commission's decision fulfilled that requirement as far as the two undertakings raising the issue were concerned. In the van Landewyck cases, the Court confirmed that different administrative procedures can be joined together in one decision, but added nothing of relevance to the issue now under examination.

The case-law thus indicates, rightly in my view, that each individual addressee must simply be able to obtain a clear picture of the complaints which concern it. However (I almost said of course), the case-law gives no indication of the degree of precision required of the decision in order for a clear picture to be obtained from it.

In the present cases, the Commission gives, in Chapter A of the decision, a detailed account of the market circumstances that it considers material. That section does not give rise to any problems. In Chapter C, the Commission describes in points 15 to 68 the documentary evidence in its possession; first, the nature of that evidence (point 15) and then, with great meticulousness in my view, the original floor-price agreement, the regular meetings, the purpose of the meetings, the target-price system, price initiatives and their implementation, individual price initiatives, the alleged quota arrangements and the special position of the four major producers. In Chapter F, all the undertakings are specifically named and the involvement of each one is discussed. Points 79 to 94 contain the detailed legal assessment of the Commission, which is presented in such a way that if the Court does eventually uphold that assessment, it can, as a legal statement of reasons, fully sustain the conclusion reached. The Commission sets out its view of its international jurisdiction in point 95; in points 96 to 102 the Commission addresses the points of dispute that might arise as regards the liability of the individual undertakings under criminal law and, in point 103, the question of limitation. In points 104 and 105 the Commission specifies the period during which it considers each undertaking participated in the infringement.

So we have first of all a clear temporal delimitation from a general point of view. The Commission has also addressed itself in each case to the most important factors which can give rise to liability and has given a detailed description of what it considers to be the deed committed as well as a description of the intensity with which, in its view, each individual undertaking participated. In the light of the foregoing I do not consider that the applicants are correct in their argument concerning the statement of reasons in the decision.

4. The interpretation of Article 85 as a formal problem

The final matter which may be addressed in the course of this general examination of the case concerns the argument put forward by some of the applicants that the Commission's failure to distinguish in the decision between, on the one hand, an agreement within the meaning of Article 85 and, on the other, a concerted practice constitutes a disregard of essential legal safeguards which restricted the applicants' ability to prepare an effective defence against the Commission's decision.

That view is based upon an interpretation of Article 85 which the applicants put forward as the correct one, namely that it is necessary to be able to prove actual conduct on the market in order for the ingredients of a concerted practice to be present. Consequently, their view is that the nature of the proof for the two different types of infringement is different and that the matter to be proved is therefore different. If that is the case, the Commission must, in the applicants' view, be obliged to explain precisely what it regards as an agreement and what it considers to be a concerted practice. Where the Commission has not done so, there is, in the applicants' view, such a blurring of the picture that their ability to conduct an effective defence is reduced.

The applicants do not, however, dispute that the Commission was entitled to draw up its decision in the way in which it did if its view that concerted practices are not necessarily and by definition required to

manifest themselves on the market is right. For its part, the Commission would in all likelihood agree that if the applicants are right the decision ought to have been framed in such a way as to make plain what it regarded as a concerted practice and what was considered to be an agreement.

As mentioned above, it is, however, quite clear that the requirement to state reasons must be seen in the light of the legal view which the Commission actually sets out.

As will be seen, the reply to that question therefore depends on the interpretation of the term concerted practice in Article 85(1) and, in my view, the statement of reasons, which in itself explains the Commission's point of view remarkably well (see, in particular, point 87(3) and (4) of the decision), therefore satisfies the requirements of Article 190 of the Treaty. If the Commission's legal view is not upheld, the statement of reasons will as a result also be defective, but this would then be of no particular interest.

C — The concept of an undertaking as perpetrator of an act

In points 96 to 102 of the decision the Commission sets out its reasoning on the question of the significance to be attached to a series of reorganizations within the European petrochemical industry for the purpose of imputing responsibility for the infringements covered by the decision. The considerations it sets forth concern inter alia ANIC, Rhône-Poulenc and SAGA Petrokjemi.

In point 96, second paragraph, of the decision, it is stated inter alia that:

In points 97 to 100 of the decision the Commission then describes in detail the structural changes that took place with regard to Saga Petrokjemi. It appears, inter alia, that prior to 1982 Saga Petroleum held 56% and then 100% of the shares in Saga Petrokjemi but that the Commission saw no grounds for considering that the two undertakings were the same undertaking for the purpose of liability to fines and thus for imputing infringements committed by Saga Petrokjemi to Saga Petroleum. In 1983, Saga Petroleum was sold to Statoil and on 1 January 1984 Saga Petrokjemi ceased to exist as a separate legal entity when, as it is stated, it was absorbed into Statoil, that is to say, it merged with Statoil. It is further stated that the undertaking now forms a separate profit centre in the Statoil organization. It is then stated that Statoil now operates in its own name the thermoplastics business formerly represented by Saga Petrokjemi, whose marketing subsidiaries in Denmark and the United Kingdom are now styled as subsidiaries of Statoil but perform exactly the same function as before in relation to the sale and marketing of thermoplastics.

The Commission goes on to say that it cannot seriously be contested that it could have imposed a fine on SAGA Petrokjemi if that undertaking had continued in existence in its original form. In the Commission's view, the key question is whether following the merger and despite the changes in structure and legal form, the undertaking which committed the infringement is still in existence or whether it has been liquidated, a question which falls to be determined exclusively by reference to the rules of Community law. The Commission then defines the term undertaking as referring to any entity engaged in commercial activities, and in the case of corporate bodies, as possibly referring to a parent or to a subsidiary or to the unit formed by the parent and subsidiaries together.

The Commission takes the view that, although it was subsumed into a larger group of undertakings, the undertaking which committed the infringement, Saga Petrokjemi, continued in existence. According to the Commission, the determining factor is whether there is an economic and functional continuity between the original undertaking and its successor, even if the successor did not continue the unlawful activities. With regard to functional and economic continuity, the Commission refers to the fact that the successor, Statoil, had not dissolved the business of Saga Petrokjemi or liquidated its assets but, on the contrary, had continued the economic activities and retained Saga Petrokjemi's operating plant and marketing output. It also refers to announcements in the trade press emphasizing the continuity of activity, management and employment, and to the fact that the senior officer of SAGA Petrokjemi who took part in the bosses meetings' continued in his previous position and was later promoted to President of Statoil's petrochemicals and plastics operations. The Commission accordingly concludes that the undertaking which committed the infringement remained separately identifiable in economic terms and therefore responsibility for infringement of the law was not extinguished upon the merger. Responsibility for the infringement accordingly rests with the undertaking which took over the undertaking which committed the infringement.

In the decision, the Commission distinguishes between, on the one hand, the Norwegian case and, on the other, ANIC and Rhône-Poulenc's situation, inasmuch as it is stated that in the latter two cases the undertakings which committed the infringement remained in being as separate entities, even though they had disposed of their polypropylene business to other producers (see the first paragraph of point 101 of the decision).

Rhône-Poulenc has not objected to being held responsible for any infringements committed in the period before that undertaking sold its polypropylene business. ANIC, however, has submitted that the Commission was wrong to impose a fine on it for acts committed before the transfer to Montepolimeri of ANIC's polypropylene business. ANIC considers that this constituted unequal treatment to the detriment of ANIC in relation to the Scandinavian undertakings and, moreover, in relation to the Commission's previous practice, according to which liability to fines followed the business transferred. At the start of the case, ANIC further claimed that there was a discrepancy between the Commission's view on the ANIC-Montepolimeri relationship and its opinion of the connection between ANIC and SIR, since ANIC was of the opinion that the Commission had imputed to ANIC possible infringements committed by SIR prior to ANIC's takeover of SIR's polypropylene business.

As far as the SIRANIC relationship is concerned, there was, as became apparent during the course of the proceedings, in part a misunderstanding, because at the hearing the Commission explained that it had not intended to let ANIC bear the consequences of the acts committed by SIR. The misunderstanding seems to have arisen because the Commission largely relied on documentary evidence in which SIR and ANIC were mentioned together and since in 1982 ANIC in fact took over SIR's polypropylene business through its subsidiary SIL S. p. A., which was later taken over by Enoxy Chimica S. p. A., ANIC believed that the Commission had imputed to it possible infringements committed by SIR. The Commission denies, however, that that was the case and does not dispute ANIC's explanation that prior to the takeover there was no connection between SIR and ANIC or that on SIR's production plant no production of any significance on ANIC's account ever took place.

It can therefore be concluded that the Commission did not intend to make any liability to a fine follow the transfer of SIR's polypropylene business to ANIC. ANIC, however, remains of the opinion that the Commission fixed its fine without taking proper account of SIR's market share in the relevant period. ANIC also considers that the Commission wrongly omitted to attach evidential value in ANIC's favour to the fact that ANIC and SIR are mentioned together in a number of pieces of documentary evidence, because, it is said, the joint designation SIR/ANIC can equally well mean that the infringement was wholly or partly committed by SIR and not by ANIC. The Commission contests both of ANIC's points. The question is one of evidence and will be dealt with below in the section concerning ANIC.

With regard to SIR, ANIC then raised the question why no legal action was taken by the Commission against that undertaking if the Commission now maintains that it did not wish to hold ANIC responsible for SIR's earlier activities. The Commission explains that legal proceedings were not initiated because SIR had gone into liquidation. In response, ANIC stated that SIR continued to exist because the undertaking was taken over by a consortium of, inter alia, banks, and was now part of the Montedison group. However, from the evidence available there still appears to be some doubt as to what has really happened to SIR.

According to what the Commission told the Court, it seems clear, however, that it took the view that if legal action was to be taken against anyone on account of SIR's infringements, it was not to be against ANIC because of that company's takeover of SIR's polypropylene business, but rather against the remaining part of SIR. The fact that the Commission, apparently on grounds of expediency, chose not to try to hold the liquidated undertaking liable cannot, I believe, be turned to account in asserting that the Commission was inconsistent in its attitude towards the different situations, even if the failure to bring legal proceedings was due to a misjudgment of the prospects of imputing liability. The Commission's attitude to the imputation of liability is thus reasonably clear within the framework of the present case and manifests itself in the ANIC-Montepolimeri relationship, in which the problem raised is real.

Consequently, the issue to be addressed is how to treat liability incurred but not yet penalized at the time of a transfer or other change in ownership of the entity with regard to which an infringement is committed. In other words, it is a matter of determining what is to happen to the undertaking qua perpetrator following restructuring involving a change of ownership. From the point of view of the transferor, the crucial point is, of course, whether it is possible to transfer an undertaking with contingent liability under the competition rules, and, from the transferee's point of view, whether a transferee which has not acted unlawfully itself should have to face the imposition of a fine arising from an economic activity it took over after the unlawful conduct took place. On this point, the first thing which should be observed is that while the unlawful acts themselves can be imputed to the undertaking on an objective basis, with the consequence that, regardless of the subjective situation of the person who acted on behalf of the undertaking, it can be ordered to bring the infringements to an end, the imposition of fines under Article 15 of Regulation No 17/62 always requires intentional or negligent conduct (see Article 15(2)).

Secondly, this question is to that extent different from the question as to which person or entity, qua addressee of the fine, is liable to the fines. It would be pointless to impose a fine on a division of an undertaking if the fine could not be enforced against that entity. It will thus always be necessary to identify the legal person, or, depending on the circumstances, the natural person against whom the fine can be levied (see, on this point, the second paragraph of point 101 of the decision), which in itself can give rise to problems.

As regards the imputation of liability, ANIC has further stated that in assessing ANIC's situation the Commission used a definition of the concept of undertaking different from that used with regard to the Norwegian undertakings. In the Norwegian case the Commission treated the undertaking as an economic-functional entity and not as a legal person. ANIC does not, however, continue to exist as an undertaking in the polypropylene sector but rather as a legal entity which owns a number of other economic-functional entities which are not engaged in the polypropylene business. The distinction on which the Commission relies, namely the transferor's continued existence after the transfer, has, according to ANIC, absurd and arbitrary consequences, since it allows the transferor's liability to depend solely on whether he is running another commercial undertaking and how that is organized. ANIC maintains that the fact that the Commission's view is internally inconsistent becomes apparent when a comparison is made between ANIC's situation and the relationship in the peroxygen case between PCUK and Atochem, to which the Commission refers in point 101 of the decision. If the Commission's view is upheld, it would have sufficed, in order for ANIC to escape liability, that the activities of the company's other sectors be transferred to other companies in the ENI group. The Commission's distinction is thus unfounded. It is, according to ANIC, difficult to see how the existence or nonexistence of the legal packaging can justify the Commission's acting in one case against the undertaking and in another case against the legal packaging. Regardless of whether the Commission wishes to treat the undertaking or the legal emballage as the decisive factor for imputing liability, it must follow its view consistently and independently of a subsequent finding that the entity which the Commission regards as the crucial one continues to exist. ANIC claims that upon the transfer of its polypropylene business a true transfer of undertaking took place, with all its tangible and non-tangible assets including plant and polypropylene know-how being transferred to Montepolimeri. The polypropylene business constituted an economic entity in itself within ANIC and that economic entity was transferred. As a company ANIC can be regarded as a group of different undertakings. The fact that one of those undertakings was transferred while the others remained within ANIC does not appear to differentiate ANIC's situation sufficiently from the Norwegian undertakings.

On this point, the Commission has referred to the fact that there is no discrepancy between its treatment of the Norwegian undertakings and ANIC. While the Norwegian undertaking continued to exist as part of Statoil with essentially unaltered economic and functional characteristics, ANIC remained the same undertaking before and after the transfer of its polypropylene production facilities. In the case of the Norwegian undertakings, the Commission did not proceed on the basis that the term undertaking is synonymous with an area of production or activity. According to the Commission, the term undertaking is, on the contrary, a complex concept which comprises personal and material elements making up the exercise of a specific economic activity. The views of competitors and customers can help to identify the undertaking. ANIC thus remained ANIC in the eyes of its customers and competitors after the transfer of its polypropylene business. They could see that ANIC was no longer involved in the polypropylene sector, but they could also see that ANIC continued in business as an undertaking. ANIC was not made up of many undertakings, that is to say one per area of production. As an undertaking ANIC has a single object which was not altered by the transfer of the polypropylene business and therefore ANIC must be regarded as having continued to exist as an undertaking.

The problem of identifying the guilty party or, as the Commission later put it in the PVC and LdPE cases, of undertaking identity, can arise both in relation to parent and subsidiary companies and in relation to transfers, changes in ownership in general or other forms of reorganization. In parent company/subsidiary relationships an undertaking can, in the connection relevant here, be the parent company, the subsidiary company or the economic entity made up of the parent and subsidiary company together. This form of problem has come to the fore many times in the case-law of the Court of Justice and is examined thoroughly in academic legal writing. The same question in the case of transfers has led to only a few decisions.

In the Suiker Unie judgment there arose the question of imputing responsibility to Coöperatieve Vereniging Suiker Unie VA, which commenced business on 2 January 1971. It disputed that it could be held responsible for infringements committed before that date. The specific situation was this: in 1966 four cooperatives had formed a coordinating body on a provisional basis with the cooperatives themselves as members, but in 1970 a cooperative society was formed in which members of the original four cooperatives were direct participants. The coordinating body, which had participated in the infringements, was then dissolved. Suiker Unie now maintained that it could not be held responsible for the coordinating body's acts since it had not taken over any assets from that body but could only be regarded as the legal and economic successor of the four original cooperatives. On this point, the Court stated (in paragraphs 84 to 88, [1975] ECR 1926) that, as Suiker Unie had assumed all the rights and liabilities of the four cooperatives of the old association, it had to be treated as the economic successor both of the old association and of its members, which had intended to confer that role on Suiker Unie. The Court went on to point out that it was not denied that the name Suiker Unie still covered the same undertakings, which were run for the most part by the same persons and had their registered offices at the same address, and that it was not even claimed that Suiker Unie's conduct on the market differed from that of the former association. The Court accordingly concluded that the main feature of the conduct of Suiker Unie and its predecessor was its obvious continuity, which means that the whole of this behaviour must be attributed to Suiker Unie. In his Opinion, Mr Advocate General Mayras examined the issue very thoroughly (pp. 2078-2079). He stated inter alia that, under the competition rules, fines are imposed on undertakings in their capacity as economic entities and it is the economic facts which must be made to prevail. He went on to allude to the risk of circumvention and considered that the Commission was entitled to impute the prior course of conduct to the new legal person, provided that the latter was responsible for the same economic entity.

In connection with the Rheinzink decision Advocate General Rozès also pointed out (at p. 1718) the possibility of circumvention. She emphasized further that it is not the continued act, which will often be the continued unlawful conduct, which is decisive, but on the contrary a balancing of the evidence which the Court of Justice in the sugar cases had laid down as necessary for a finding of a common course of conduct, namely (a) that Suiker Unie had assumed all the rights and liabilities of the cooperatives, (b) that it was not denied that the same undertakings were involved; and (c) that it was not even claimed that Suiker Ünie's conduct on the market differed from that of the former association. Advocate General Rozès refused to accept that only a continuation of unlawful conduct was decisive. What was decisive was the assumption of the rights and obligations of the former undertaking together with the fact that the head office and the management of the undertaking were unchanged. In view of those circumstances, she concluded in that case that even though Rheinzink GmbH and the former undertaking Rheinisches Zinkwalzwerk GmbH & Co. were not identical, the connection, both economic and legal, between the two companies was such as to allow their acts to be treated as continuous so as to justify imputing the infringements complained of to Rheinzink. The Court of Justice held (paragraph 9, at p. 1699) that all things considered Rheinzink GmbH had taken over the former company and had continued its economic activities, stating that a change in the legal form and name of an undertaking did not free the new undertaking from liability for the anticompetitive behaviour of its predecessor when, from an economic point of view, the two were identical.

With regard to the Commission's practice, it might be appropriate first to examine in more detail the peroxygen decision to which the Commission refers in the final paragraph of point 101 of the polypropylene decision, and which is also discussed in these proceedings. In the peroxygen decision (Official Journal 1985 L 35, p. 1, in the fourth paragraph of point 49) it is stated :

In the polypropylene decision the Commission refers to the fact that PCUK, which was the undertaking which committed the infringement, was split up after the infringement had ceased and its peroxygen interests were absorbed by Atochem. The Commission goes on to state that it had held that, since PCUK had ceased to exist as a separate legal entity, Atochem, having taken over its peroxide business and adopted its economic objectives, had to be the addressee of any decision. That PCUK subsequently went into liquidation is quite correct and is supported in addition by the information given in the PVC decision (point 49), but it is not apparent from the peroxygen decision. In particular, it is not apparent that the Commission may have attached weight to the question of how far there existed a remaining part of the undertaking. Rather, from the wording it seems that the Commission attached weight to the takeover of assets and economic objectives, that is to say the peroxygen interests, without the question of the transferor undertaking's continued existence being of importance. That the Commission had probably earlier taken the view ascribed to it by ANIC, namely that liability follows production and earnings, seems also to follow from the PVC decision (point 44), from which it appears inter alia that Norsk Hydro, which had acquired the PVC business from an undertaking that continued to exist, should, according to the Commission's original view, have been liable for earlier infringements, but that the Commission now, that is to say in the PVC decision, recognized that the transferor should have been held liable.

In the PVC and LdPE decisions the Commission stated generally on the question (second paragraph et seq. of point 42):

In both decisions the Commission then examines various specific cases. Without examining those cases in any greater detail, it is probably right to say, in view of the foregoing and what the Commission has stated in the present cases, that the view now taken by the Commission is that if there is a merger whereby the entire former undertaking is integrated into the new one considered as an undertaking, responsibility attaches to the absorbed undertaking. On the other hand, responsibility stays where it is if there is still part of the undertaking remaining, in any case if in its essential form [it] is still in existence.

It would seem that, in its decisions, the Commission has wavered between the individualizing method, which it used for Saga Petrokjemi/Statoil case and a simpler variant whose decisive element is whether the former undertaking in its essential form continues in existence. The Commission was thus an easy target for ANIC's criticism.

As is clear from the case-law of the Court of Justice cited above and also to that extent from the Commission's practice, it must be specifically determined in every case what has become of the undertaking at fault. In the polypropylene decision and its written pleadings in the ANIC case, the Commission gives an excellent account of the factors to be taken into consideration. In the section of the decision concerning Saga Petrokjemi, the Commission also explains in an exemplary way why Statoil should, in its view, bear responsibility.

At all events, the way in which the combination of material and personal elements which must be seen as making up the undertaking fit into the altered structure must therefore be examined very closely. In this connection, it must be recalled once more that liability to a fine is liability for an intentional or negligent infringement and that the personal factors upon which the Court of Justice has insisted in the case-law cited must therefore be accorded great importance.

The view that one might have suspected the Commission of entertaining, namely that any remaining part of the undertaking may be held liable to a fine in any event, is thus probably simply incorrect, even though it may usually reflect the true position. On the other hand, it may be appropriate to emphasize that general considerations concerning the conditions and function of liability to a fine may require weighty reasons for holding, in the absence of special grounds, that liability can be incurred through the acquisition of a production entity with regard to which an infringement of the law has been committed by the transferor. Indeed, the Commission seems to be aware of this. Without its being necessary to come to a conclusion on the decision as far as the Norwegian undertakings are concerned, the question whether the management bodies and/or the group of persons behind the infringements were also transferred will be an important factor for the imputation of liability.

In the case of ANIC, the information is sparse. As against the Commission's assertion that ANIC continues to be a functioning undertaking within the meaning that I have sought to define above, there is only in fact ANIC's statement that all tangible and non-tangible rights were sold, including production plant and know-how.

Confronted with the Commission's assertion that even after the transfer of the polypropylene business ANIC remained ANIC in the eyes of its customers and competitors and its assertion, in point 101 of the decision, that the cases of ANIC and Rhône-Poulenc are different from that of Saga in which the management and personnel were also transferred to Statoil, ANIC has not in particular explained whether the management bodies and/or groups of persons responsible for the polypropylene business left ANIC and moved to Montepolimeri.

In view of the case-law of the Court of Justice cited above and the considerations I have set forth concerning the conditions and the function of liability to a fine, the fact that ANIC itself transferred all tangible and non-tangible rights, including production plant and know-how, to Montepolimeri is not, in my opinion, sufficient for ANIC to cease to bear liability when it must be assumed that in its present form and with its present management and executives ANIC essentially continues to exist, only without its polypropylene production.

D — The interpretation of Article 85

1. Introduction

Article 85(1) prohibits all agreements between undertakings and all concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the common market. Article 85(1 )(a) to (e) enumerates, non-exhaustively, the forms of conduct of which the applicants in the present cases are guilty in the Commission's view. According to Article 85(2), any agreements or decisions prohibited pursuant to Article 85(1) are to be automatically void.

In the cases now before the Court, the interpretation of Article 85 has given rise to argument on the interpretation of the concept of concerted practice and on the question of the extent to which a plurality of acts or, in some circumstances, omissions, which are partly quite heterogeneous in character can properly be regarded as a single agreement or as a concerted practice within the meaning of Article 85(1).

Another question which, conceptually, can be considered independently from the problem referred to above but which is closely allied to it is whether the prohibition in Article 85 covers participation per se in a cartel with the result that the participants, by virtue of their participation, can be held responsible for all the infringements committed in the cartel. In the present cases, the question has been discussed under the heading collective responsibility.

Before I go any further into the question of the interpretation of concerted practice, it might be useful to begin with a few observations on the interpretation of the concept of agreement within the meaning of Article 85.

2. The concept of agreement

The case-law of the Court of Justice enables the concept of agreement to be defined with relative precision. An agreement covered by Article 85 can, of course, have been concluded in such a way that it would be legally binding on the parties but for the fact that it is invalid under Article 85(2). However, it is clear from the case-law of the Court of Justice that an agreement within the meaning of Article 85 may also consist of a gentlemen's agreement, the binding and rulemaking character of which is not due to legal factors but to social psychological factors. An agreement may consist of a continuous contractual relationship and it can probably also be tacit so that it need not be set down in writing.

Anticompetitive agreements are prohibited as soon as they have as their Object the restriction of competition. This has been quite clear since the judgment in the Grundig case. The detailed requirements to be satisfied in order for an agreement to have as its object the restriction of competition will be dealt with below in the section devoted to the argument that the agreements in question did not have the restriction of competition as their objective object.

It is important to emphasize, as the Court has done most recently in the Sandoz case, that the offence involved is a pure conduct offence so that it is not necessary to attempt to explain the actual effects of the agreement. This form of infringement of Article 85(1) has rightly been described as an abstraktes Gefährdungsdelikt.

An agreement within the meaning of Article 85 (agreements ... which have as their ... effect...) may, however, also display the characteristics of pure result offences; in such cases, it is unnecessary to prove an anticompetitive object. On the other hand, it is clear that a particular set of facts may at one and the same time contain the constituent elements of both types of offence.

In the present cases, it is not, however, the concept of agreement which causes difficulties in the realm of interpretation. Rather, as mentioned above, it is the concept of concerted practice which raises considerable problems of interpretation.

3. The concept of concerted practice

(a) The problem defined

The cases in which the Court of Justice has had to consider the question of the interpretation of the concept of concerted practices contained in the prohibition laid down in Article 85 are very few. As will be seen later, owing in particular to the different factual situations in the relevant cases, only limited assistance in the task of interpretation is to be found in that case-law for the purposes of resolving the specific problems arising in the complex of cases now before the Court.

As is apparent from the proceedings in these cases, there are considerable difficulties in arriving at anything like a clear definition of the conceptual content of, on the one hand, the offence consisting of a concerted practice having an unlawful object, and, on the other hand, a concerted practice having an unlawful effect.

Much of the debate in academic circles and in the present cases has left the impression that many commentators have in fact reached the conclusion that the concerted practice/object combination may if anything be described as a conceptual anomaly, which has virtually led to the explaining away of this type of offence. It must also be recognized that a type of offence which appears to combine elements of both a conduct offence (object) and a result offence (practice which is concerted) may be a difficult concept to handle.

The Commission maintains that there is evidence proving that the alleged concerted practice was put into effect by the initiation of parallel measures. It also believes that there was an effect on the market. In the Commission's view, however, in order for an infringement of Article 85 to be found it need only be proved that concertation took place. So, in the Commission's view — as expounded in these proceedings — the concerted practice is constituted by the concertation per se.

The Court might conceivably find that there is evidence that the elements in the case which might rightly be described as a concerted practice had an appreciable effect on the market or at least that there is evidence proving the initiation of parallel measures. It is also possible, however, that the Court might come to the conclusion that it is not possible to establish, with the required degree of certainty, either perceptible effects on the market or at least conduct which can be traced back to concertation of some sort.

It would then be a matter of interest unto itself to know how to judge the situation when one knows that collusion has taken place but cannot establish exactly what happened thereafter.

There is now before the Court a complex of cases in which the factual circumstances are such that the Court will have to decide for the first time whether the term concerted practices appearing in Article 85 can also cover cases in which there is no proof of specific, concrete acts on the market but only of meetings between the market operators at which information on prices, production volumes and so forth, which are normally regarded as matters of business secrecy, was exchanged, and whose object, judging by the subject-matter of the discussions which took place between the parties, must have been to coordinate subsequent conduct on the market. In other words, the question is whether a concerted practice within the meaning of Article 85 presupposes manifestation on the market as a constituent element of the infringement or whether the actual conduct, formed by the concertation itself at the meetings, can be regarded as unlawful under the provision, even if after the concertation no actual conduct, causally related to it, on the part of the undertakings which took part in the concertation can be proved.

The pleadings submitted by the parties in the present cases may also give rise to considerations as to whether an attempt falls within the ambit of Article 85; in other words, whether under Article 85 concerted practices ... which have as their object.... constitute an offence in themselves and not only attempts to commit the offence of adopting concerted practices having an unlawful effect.

(b) The significance of the question in the present cases

It is somewhat surprising that the problem is raised so distinctly in these cases in which, as is apparent from point 87(3) of the decision, the Commission's takes the view that the essence of what took place must be regarded as an agreement within the meaning of Article 85(1). In fact, the Commission sets out only two things to illustrate what, in its view, constitutes a concerted practice rather than an agreement. These are the arrangements which, according to the Commission, were made for the exchange of information on deliveries in 1981 and 1982 and cases where an undertaking's assent to and cooperation on certain aspects, for example Shell's allegedly more passive cooperation on quota schemes, were only tacit.

The Commission did, however, consider that those aspects should be taken into account and at the same time stated, or at any rate suggested, that a whole series of other sub-elements or aspects of the undertakings' conduct could also be said to have characteristics of a concerted practice.

The Commission has omitted to make a distinction, which means that the Court will have to deal with the question whether the two forms of prohibited conduct — agreements and concerted practices — have such a similar structure and such similar characteristics that, for the purposes of the application of the provision, it is not necessary categorically to classify the acts in question as being one or the other.

In the present cases, in which the concerted practice component is, as I will explain below when considering the evidence and as the Commission maintains, limited in extent, it would no doubt have been possible for the Commission without much extra work to frame the decision differently so as to differentiate between the elements which it regarded as an agreement and those which it regarded as amounting to concerted practices in case the Court rejected its principal argument.

But the Commission did not do this, so the question is what happens if the Court does not agree with the Commission on this point. Is the Court itself to undertake the classification or is the decision to be annulled? In view of the existing case-law of the Court of Justice, in which the Court of Justice has carried out its own fairly thorough assessment of the cases, the possibility of the Court carrying out an independent examination of its own can hardly be excluded. However, such a reexamination would meet at least two difficulties. First of all, if the Commission's view is not upheld, it seems to me that the decision will not stand, for lack of reasoning. In that hypothesis, the statement of reasons in points 86 to 88 is simply insufficient to satisfy the requirements of Article 190. The second difficulty is that in point 87 of the decision the Commission indicates that, in its view, all aspects of the applicants' actions may contain elements of both an agreement and a concerted practice. In my view, however, the review undertaken by the Court under Article 173 of the Treaty cannot mean that the Court should in fact go through the case from the beginning, which is what would happen if the Court itself were to attempt to determine the extent to which each individual element of the course of the undertakings' conduct amounted to an agreement or to a concerted practice. If the Commission's view on this point is not accepted, the decision must consequently be annulled in its entirety.

(c) The parties' submissions and arguments

In its letter of 29 March 1985, the Commission set out at length its views on the relationship between agreements and concerted practices within the framework of Article 85(1). In the decision, those views are developed, as mentioned above in this Opinion in section I, A. 1. In points 86 to 88 of the decision, the Commission explains, with reference to the dicta of the Court in the ICI and Suiker Unie judgments, why it considers that it is necessary not to draw a distinction, but to determine the lower limit beyond which cooperation can be regarded as an infringement of Article 85.

In the general part of the defence, the Commission provides a most detailed description. That account was later amplified at the hearing before the Court.

The Commission thus contends that agreements and concerted practices within the meaning of the Treaty cover all types of arrangements by which producers mutually accept a limitation of their freedom of action instead of determining their future competitive conduct in complete independence. Such arrangements always presuppose direct or indirect contacts between competitors, whether these take the form of formal contracts, informal agreements of the kind known as gentlemen's agreements, or simply practical cooperation. The purpose of having the two concepts, agreement and concerted practices, in Article 85 is, in the Commission's view, to avoid any lacuna in the scope of application of the provision. Article 85 can thus be applied to all agreements, express as well as implicit, formal or informal. It can also apply to purely de facto or practical cooperation. Such cooperation is not necessarily identical with a common pattern of behaviour on the market, because the prohibition in Article 85 also covers the mere object of distorting competition.

The Commission then describes the various degrees of cooperation. First, agreements which do not need to be legally binding — which by virtue of Article 85(2) they never actually are if they fall under Article 85(1). Whether one chooses to regard nonbinding arrangements as agreements within the meaning of Article 85 or to reserve the term concerted practices for practical cooperation which has not been given formal expression, it is the whole gamut of anticompetitive arrangements which is caught by Article 85. The term concerted practices refers to practical cooperation of a merely factual nature and cooperation that need not arise from a plan or concertation properly so called. According to the Commission, the judgments of the Court in Suiker Unie and Ziichner show that there can be a concerted practice once contact between competitors takes place prior to their behaviour on the market. This contact may, according to the circumstances, consist in exchanges of information without its being necessary for there to be an agreement on such exchanges. There may conceivably be an agreement to exchange information which can in itself be caught by Article 85 if it has the object or effect of restricting competition. In order for such an exchange of information to be regarded as a concerted practice having at least the object of restricting competition, the information exchanged must relate to the parties' intentions regarding their future conduct on the market and must not be available to competitors by mere observation of the market. The object underlying such an exchange of information is to enable each of the undertakings to determine its own market conduct in reliance on its competitors behaving in parallel. Such an exchange of information cannot be explained except on the assumption that there exists a legitimate expectation between the information-exchanging parties that the others will behave as they previously indicated they would. When, for their part, the applicants contend that a concerted practice must have manifested itself on the market, this indicates to the Commission that they are confusing the question of the proof of the existence of a concerted practice with the concerted practice itself.

In my view, there is little indication that in general the applicants are confusing the concepts or have otherwise misunderstood the problem. They simply disagree with the Commission's point of view.

The most succinct account of the applicants' point of view is to be found in Mr Hermann's oral argument presented at the hearing before the Court. He stated inter alia that in the case of concerted practices the minimum requirement for Article 85(1) to be regarded as infringed is that it should be proved that (a) at least two undertakings entered into concertation by whatever means; (b) the concertation was followed by a corresponding practice on the market; (c) that practice had an anticompetitive effect; and (d) in the case of an anticompetitive effect, this effect was foreseeable.

The applicants also recognize that a concerted practice within the meaning of Article 85(1) can consist in a concerted practice having only the object of restricting competition without such an effect having to be proved; according to the applicants, this may be where the undertakings pursue an anticompetitive purpose through specific conduct on the market but without being in a position to fulfil their project.

In their submissions both sides examine closely the case-law of the Court of Justice but arrive at completely different results. It will therefore be useful first to look at the background to the provision and at what can be deduced from the Court's case-law which, as already stated, is sparse. The Opinions of the Advocates General in those cases also contain observations to which the parties have referred and which will be examined in more detail. Finally, academic literature is of some assistance in the task of interpretation.

(d) Historical background

As is well known, in drafting Article 85(1) the authors of the Treaty were influenced by the concepts concerted actions, concerts of action and concerted practices, those concepts having originated in American case-law on the basis of Section 1 of the Sherman Act and notably the concept of conspiracy contained therein.

It is apparent from American case-law that anticompetitive effects are not necessary as a constituent element of conspiracy, just as no acts other than the conspiracy need be committed.

As Joliét has remarked, the concept of concerted action was of significance for the determination of the legal meaning of conscious parallelism of action, when there was no direct evidence of conspiracy. The concept has thus been important in cases where the problem has been examined from the market aspect and where on the basis of an assessment of the market compared with other evidence it was necessary to determine whether the alleged practice could be presumed to have been concerted unlawfully.

Direct evidence of conspiracy is, however, as stated, sufficient to constitute an infringement of the Sherman Act.

In the present cases, the applicants, in their arguments concerning the genesis of Article 85, assume that the authors of the Treaty, under the influence inter alia of the Allies' somewhat vague postwar decartelization legislation in Germany, did not wish to adopt the American concept of conspiracy, which, in the applicants' view, was also rather imprecise. For reasons of legal certainty, the concepts of agreement, a term with a fairly well-defined meaning, and concerted practices were preferred. The latter concept, I understand, was in fact inspired by American case-law, being a component of conspiracy, which concerns conduct on the market.

The interpretation of the Treaties on the basis of the travaux préparatoires is a notoriously difficult area in Community law, one reason for this being that a large number of the preparatory documents have not been published. In the area of competition law, the difficulties are illustrated, for example, by Ellis's examination of the known, more or less official, preparatory documents relating to Article 85. It is probably also indicative that the applicants nave not pointed to any specific, written elements in the genesis of Article 85 in support of their view.

It is certainly not improbable that the applicants may well be right in their observations on the historical background, but the significance which can be attached to them is hardly decisive. When one considers the wording of the provision, which is plainly intended to embrace all anticompetitive activity incompatible with the common market, it cannot be presumed without very solid evidence that the authors of the Treaty wished to exclude from the scope of the provision a whole category of questionable business initiatives. The Court of Justice has made no such assumption in the cases in which it has had occasion to address this matter, as is clear from the judgments cited below. Nor do I see any decisive criteria for interpretation which would compel the Court to limit the scope of Article 85 in that way. On the other hand, historical considerations do not support the Commission's view either.

(e) The dicta of the Court of Justice concerning the concept of concerted practice

The Court's first judgments on this subject were delivered in 1972. They were later to become the subject of a wide, and in part critical, debate, which will be discussed below under (g).

In those cases the facts differed significantly from the cases now under review. It was price increases implemented on the market which in themselves made the Commission suspect coordination of a practice which had unquestionably been implemented on the market. The dispute was about the extent to which the uniformity of the price rises could be explained by the oligopolistic structure of the market and the judgment refers (in paragraph 96) to only one meeting at which the undertakings could have had the opportunity to arrange the concertation which is the central issue in the present cases.

With regard to the definition of the concept, the Court began by resolving a question which had previously been controversial, namely whether the term concerted practice within the meaning of Article 85(1) had an independent scope of application or whether it was simply a kind of legal rule lessening the evidential burden in cases in which essentially there is only market observation to go on and in which the evidence of an agreement will often be impossible to adduce unless the concept of agreement is to be emptied of all meaning.

In paragraphs 64 and 65 of the judgment the Court stated:

The concept of concerted practice has then, according to the Court of Justice, an independent scope which may be described as covering cooperation which is not an agreement. The key word here is coordination and the question in the present cases is whether it is coordination as such or coordination which becomes apparent from the behaviour of the participants which is decisive.

It should also be mentioned that in paragraphs 118 and 119 of the judgment the Court stated:

However, the question once again is whether it is solely cooperation in establishing a coordinated course of action or in addition the ensuring of its success which was decisive in the Court's view.

Considering that the Court was, of course, speaking in the context of the cases then before it, in which the question was precisely whether particular conduct on the market found to be de facto uniform was due to collusion and there is nothing else in the judgment to indicate that in setting forth those grounds the Court also had in view situations such as that existing in the polypropylene cases, the judgment can hardly be relied upon in support of either the applicants' view or the Commission's view in the present cases.

The Court's judgment in the Sugar cases comes perhaps somewhat closer to the view now put forward by the Commission. It was stated in paragraphs 172 to 176 [(1975] ECR 1942):

In that case, therefore, the argument was about how far it was necessary to find that there was a plan for the purposes of Article 85(1). The Court held that this was not necessary. However, it only addressed itself to the nature of the cooperation which can be caught by the provision and not to the question or the time from which an infringement may be considered to have been committed.

In the Ziichner case a German court had asked the Court of Justice whether in transfers of capital and other payments between banks within the common market the debiting of a general service charge at a rate of 0.15% of the sum transferred was a concerted practice.

In its judgment the Court first summarized its dicta in the ICI and Sugar cases, stating in paragraphs 12 to 14:

Later on in the judgment the Court discussed coordination. In paragraph 21 it stated :

The Court gave the following reply to the question referred to it for a preliminary ruling:

As will be seen from the extracts cited, that judgment provides in relation to Suiker Unie no new factors for resolving the matter. In that case, too, there was manifestly a practice within the meaning which the applicants attribute to that concept in the present cases and the Court's dicta thus actually refer to that practice.

In view of the foregoing, I consider that nothing can be inferred from the case-law of the Court of Justice directly contradicting the Commission's view, but the facts of the cases decided hitherto by the Court have been significantly different from the facts in the cases now before us. The Court's description of a concerted practice must necessarily be seen against that background and thus cannot be assumed to provide any answer to the question whether the offence of a concerted practice referred to in Article 85(1) is constituted by concertation alone or whether subsequent de facto and causally related conduct on the part of the undertakings involved is required.

(f) Observations of the Court's Advocates General on the question

In the Chemiefarma case, Mr Advocate General Gand took the same view as that maintained by the applicants in the present cases. In examining how far a gentlemen's agreement was to be regarded as an agreement or a concerted practice within the meaning of Article 85(1), he stated as follows :

In the Dyestuffs cases, Mr Advocate General Mayras essentially endorsed the view expressed by Mr Advocate General Gand in the Chemiefarma case. However, on one point he goes further than Mr Advocate General Gand. First, he refused to accept that as a legal category concerted practices were simply a particular variant of the concept of agreement; he maintained that the distinction in the Treaty should be given an independent meaning and that concerted practices should be regarded as a separate category in order to avoid circumvention of the provision by undertakings' not leaving any written trace of their agreements. That view was followed by the Court, as I mentioned. Mr Advocate General Mayras went on to state (at p. 671, right-hand column):

In a later section, devoted to the adverse effect on competition, Mr Advocate General Mayras further stated (p. 682, right-hand column) :

What is interesting about that argument for the purposes of the present cases is that Mr Advocate General Mayras tries to introduce a doctrine of attempt into the concept of concerted practice as used in Article 85(1). However, the theory ventured by him has not been supported or commented upon in later judgments of the Court of Justice or by its Advocates General.

The Commission relies upon Mr Advocate General Reischl's Opinion in the Fedetab cases to support its line of argument which the applicants attack. The Commission considers that Mr Advocate General Reischl's remark about an unimportant argument of classification must mean that he shares the Commission's view. What the Advocate General in fact said was as follows [(1980] ECR 3125, at p. 3310, left-hand column):

Given their natural meaning and read in their context Mr Advocate General Reischl's comments tend rather to indicate, in my opinion, that he considers that all the constituent elements of a concerted practice, as conceived by the applicants, were present in the Fedetab case, so that in the case in point it would constitute merely an unimportant argument of classification to call what the Commission in its decision classified as an agreement a concerted practice. Whether this is so is obviously a matter of argument but if more is to be read into the remarks they tend if anything to support the views of the applicants.

Finally, the Commission mentions Advocate General Sir Gordon Slynn's observations in the Pioneer cases. He did not, however, address himself to the question as to what a concerted practice may consist in but only dealt with the question as to how concertation can be proved, namely inter alia by examining the practice.

The most recent observations on the matter come from Mr Advocate General Van Gerven in the Sandoz case (Opinion, p. 2, see footnote 7), in which, without further elaboration, it is stated that the circumstances in question must be proved to have had the effect of restricting competition. However, the problem arising in the present cases is not discussed, this being unnecessary in the circumstances of that case.

To summarize the observations of the Advocates General cited above, it may, in my opinion, be concluded that they do not support the Commission's view. On the contrary, they tend to refute it, since in their argument devoted to concertation both Advocates General Gand and Mayras as well as Mr Advocate General Reischl consider that there must be subsequent actual conduct. With Mayras, however, as quoted above, that view is overlaid with his observations on the possibility of taking action against an attempt. However, Mr Advocate General Mayras seems to confuse what he calls the necessary, Objective conduct, namely the fact that the participating undertakings must in fact have acted in the same way, with the real effect that [a concerted practice] has on the competitive situation and he refers to the actual effect being covered but possibly also the potential effect as an attempt. Mr Advocate General Mayras thus appears to consider that only concerted practices having the restriction of competition as their effect are clearly covered and that concerted practices having the restriction of competition as their object should be treated as attempted concerted practices.

(g) Academic writing

In this section, which is not intended to be exhaustive, reference is made to a number of writers whose observations on the matter highlight the nature of the problem.

Probably the most thorough commentary to be found on the Dyestuffi judgments is the article by Joliét who emphasizes in particular that any confusion between the concept of concerted practice and proof thereof should be avoided. His own definition of a concerted practice (set forth on p. 285, and see p. 271) was the reciprocal communication of intentions by competitors prior to any action on the market by which each of the undertakings concerned leads the others to expect that it will act in a certain way, thus reducing the uncertainty prevailing on the market. In Joliet's view, there is thus a concerted practice as soon as concertation takes place, that is to say whether or not implementing action is later taken by the parties engaged in the concertation and whether or not the concertation had effects detrimental to competition, provided that it was capable of having such effects.

J. A. Van Damme endorses Joliet's view, emphasizing that a concerted practice can be disassociated from the actual effects on market conditions. He bases his argument on the fact that Article 85(1) also covers concerted practices having the distortion of competition as their object. It is, however, interesting to note that Van Damme appears to make a distinction only between, on the one hand, cases in which only concertation itself is proved and, on the other hand, cases in which concertation which has affected market conditions is proved. The last category mentioned by Van Damme is really collusion having as its effect and not only as its object the impairment of competition. He does not therefore distinguish an intermediate category of concertation which has such an object and is followed by subsequent implementing action but which does not succeed and thus has no actual effects on competition.

In contrast, Schapira, Le Tallet and Blaise state that if one wishes to remain within the bounds of a literal interpretation, due consideration must be given to the term practice, which would appear to exclude mere intention. In the view of these writers, it is necessary to prove not only concertation but also the taking of steps to give effect to the concertation.

Goldman states in his commentary on the Dyestuffi case when discussing the definition of concerted practices that the concept can be understood either as an exchange of declarations of intent which the participants did not however intend to be legally binding (or as obligations from which they can unilaterally discharge themselves) or as an arrangement under which concerted participation follows upon exchanges of information and mutual consultations between undertakings but is ultimately freely decided upon by each individual undertaking. Goldman goes on to state that not only may gentlemen's agreements and agreements which can be terminated unilaterally be included in the concept of concerted practices, it is also possible for practices decided upon individually, but following exchanges of information and consultation, to be regarded as concerted. It thus appears from Goldman's description of the concept of concerted practice that he would consider the constituent elements of a concerted practice having the impairment of competition as its object to be present when (i) collusion is proved, for example in the form of consultations and exchanges of information, and (ii) action is subsequently taken, even if that subsequent action is freely decided upon by each of the participating undertakings. It is not therefore necessary for the action to be determined in common; it need simply follow the collusion. In Goldman's view, concerted practices may, therefore, be quite different as well as parallel. Concertation having the restriction of competition as its object is thus not sufficient per se; it must be followed by action.

Piriou (op. cit. p. 53), in her commentary on the Dyestuffs judgments, takes the view that the Court of Justice attached decisive weight to the effect which the concertation has on competition. From this she concludes that concertation must in practice result in the elimination of the risks of competition. She is thus dubious about Mr Advocate General Mayras's suggestion that the potential effect of a concerted practice, that is to say an attempt per se to mount a concerted practice, could be caught by Article 85(1). Piriou concludes (p. 58) that it can be inferred from the judgments that inasmuch as the Court does not lay down a minimum requirement as to the degree of cooperation needed in order for it to fall foul of Article 85, exchanges of information on prices may, for example, be caught, provided, however, that the concertation also manifests itself in an alteration in the actual competitive situation. Piriou thus appears to overlook the intermediate category in which steps have been taken to implement the concertation but they have not succeeded in bringing about the intended effects. Her description seems rather to imply that in her interpretation she ignores the category of concertation having an anticompetitive object because she considers the concertation must manifest itself in an alteration in the actual competitive situation, in other words that the concertation must have produced effects.

Eric Colmarti expresses a view which comes close to that of the applicants in the present cases, namely that a concerted practice may have an unlawful object without having an unlawful effect. According to Colmant, that is the case where the effects of the undertakings' conduct are insufficient to damage competition but are sufficiently clear for it to be concluded that there must have been an intention to enter into anticompetitive activity. Colmant maintains that the existence of a concerted practice presupposes a combination of two factors : first, concertation, the subjective factor; secondly, certain de facto conduct, the objective factor; between those two factors there must be a link.

Van Gerven stresses the importance of proving the anticompetitive activity ensuing from the concertation.

Schröter in Groeben/Boech aligns himself essentially with Joliét and J. A. van Damme. He states inter alia that the prohibition against cartels operates as soon as coordination of the participating undertakings' future conduct occurs, thus even before the intended conduct is translated into action.

Koch in Grabitz likewise states, without providing detailed explanation, that concertation per se can constitute an infringement of Article 85(1).

Bellamy and Child also consider that contact between undertakings, often consisting of meetings, discussions, exchanges of information or soundings out, when their object is to influence market behaviour, falls within the concept of concerted practices.

Kovar interprets the Dyestuffi judgments as laying down a definition with two components: one, objective, the parallel conduct, the other, subjective, its intentional character.

Druesne considers that a concerted practice comprises two aspects: the conduct itself and the intention to act together. According to Druesne, the mere exchange of information can, however, constitute a concerted practice, inasmuch as the undertakings, by making their intentions known to their competitors, exert an influence on their decision-making autonomy.

Deringer is doubtful whether Article 85(1) relates solely to concerted conduct or also covers collusion as an attempt.

As will be seen from this brief examination of some of the academic literature dating from 1973 to this day, there is considerable disagreement about the interpretation of the concept of concerted practice. Some writers regard the concept as meaning that concertation itself is sufficient, thus overlooking the word practice, or consider concertation to be a practice in itself. Other academic writers, particularly the most recent, insist that there must be both concertation and a practice, described by some as subjective and objective elements respectively. It appears clear, however, that it is the concertation which is universally regarded as the crucial element. All the commentators are unanimous that a practice on the market which cannot be traced back to any concertation is manifestly outside the scope of Article 85.

However, it also appears from an examination of the academic literature that those commentators who say that there should also be a practice seem to suppose in some way that evidence must be adduced of an effect on the market and not solely of conduct, even if it has not actually affected competition. Consequently, in my view a study of the relevant academic literature does not give any clear, or more convincing, support for either the Commission's or the applicants' interpretation of the concept of concerted practice.

(h) The starting point for interpretation

If one considers the actual wording of Article 85, a concerted practice within the meaning of that provision comprises a priori exactly the same elements as an agreement, namely there is a prohibition of concerted practices which have as their object the prevention, restriction or distortion of competition and an equivalent prohibition against concerted practices which have such an effect. A literal and grammatical interpretation thus clearly leads to the conclusion that a separate category may be identified comprising concerted practices having an (unlawful) object.

If, like some academic writers and, it would appear, the Advocates General, and in accordance with the wording of Article 85, one takes the view that, besides concertation, a proven practice causally related thereto is required, the problem which arises is to decide how much or how little is needed for there to be a practice in a case involving a concerted practice having an unlawful object. Whether a possible proven practice has had actual consequential effects on competition is unimportant in this regard. On the other hand, in the present cases the Commission is in difficulty in explaining what form is taken by that element of the offence covered by the provision which is called a practice since, in its view, concertation having the restriction of competition as its object constitutes per se a concerted practice.

As I shall endeavour to demonstrate below, the problem can, I believe, be reduced to one question: when is an infringement of the law committed? As will be seen, all this boils down to, in my view, is that the point in the course of events at which one can speak of a completed infringement called a concerted practice having an unlawful object is later than the corresponding point for agreements having an unlawful object.

(i) Do concerted practices require conduct on the market?

As is apparent from sections (d) to (e), in my opinion no support for the Commission's point of view can be found in the case-law of the Court of Justice or in the various Opinions of the Advocates General which I have cited nor is there any clear and unequivocal support for it in the academic literature.

Nor do the wording and historical background of Article 85 lend support to the theory that a concerted practice may be assumed to exist immediately upon, and by virtue of, the exchange of information of competitive significance or if concertation is ensured in any other way.

If we look at the available case-law as cited above in section (e), it will be seen that the cases have been considered from the perspective of the market. Hitherto the task has been to decide whether, on the basis of an observation of de facto conduct on the market and on the basis of the often slender documentary evidence available, it was possible to consider it established that the reason for the observed conduct on the market was collusion between the undertakings in question. It is also clear that when the market behaves in a way which is hard to explain, when one or more meetings take place between undertakings normally in competition, when more or less similar telexes are sent out with, for example, price instructions and so on, these happenings will typically be strong indications that everything is not as it should be, even if there is no conclusive direct evidence.

When one considers a situation from the perspective of the market, it is evident that it will normally be conspicuously uniform behaviour on the market which will arouse the Commission's suspicions. However, a concerted practice can quite conceivably consist in a mutual understanding between the participating undertakings that A will do X (for example, charge a particular price) while B, C, D and E will do Y (for example, charge a particular price less 5 pence). In that case, too, there is a concerted practice. If, for the sake of the argument, we ignore the fact that it is scarcely conceivable for such a sophisticated arrangement to come into existence without something that can best be described as an agreement and we suppose that A is to do a, B is to do b, C is to do c and so on, it is, however, quite obvious that such a concertation may be extraordinarily difficult to prove unless it has also been possible to obtain other, fairly direct, evidence of what has taken place. Nevertheless, there is clearly a concerted practice which simply does not manifest itself in the form of uniform parallel action.

If one is confronted with a market which is behaving in a conspicuous way, displaying other indications or unlawful concertation, the relevant factors must, of course, be concordant. If the market inexplicably shows an actual price which is the right market price plus 15%, then, as far as the evidential situation is concerned, it does not help to have a telex which states the right price minus 15% or, for that matter, the right price plus 150%. It must be possible to prove a connection upon the evidence; it must be possible to infer safely from the behaviour of the market, on the basis of evidence pointing in the same direction, that there was concertation and what form it took.

But how does the situation appear from the other side? How does the matter stand if one is aware of concertation but the other factors are somewhat less clear? Is it sufficient, as the Commission believes, that there is proof of concertation, that is to say, is the coordination or the exchange of information sufficient? Or, in other words, does the concertation constitute in itself a concerted practice?

As explained above, there is little support to be found for assuming that such an interpretation of the concept of concerted practice is correct. That interpretation might perhaps be desirable from the point of view of legal policy but it is difficult to reconcile with the ordinary meaning of the words of the provision; nor is it corroborated by the history of the provision. I therefore consider that such an interpretation should be rejected.

We have to ask ourselves, however, what is it that happens when undertakings have entered into concertation? Why is the concertation something so crucial that, in the view of the Court of Justice, it follows from Article 85 that any direct or indirect contact between such operators, the object or effect whereof is... to influence the conduct on the market of an actual or potential competitor is categorically prohibited, as it stated in the more recent Sugar and Ziichner cases cited above? In my view, the reason is that such undertakings will then necessarily, and normally unavoidably, act on the market in the light of the knowledge and on the basis of the discussions which have taken place in connection with the concertation. They will have received information about the way in which others are thinking; they will be aware that the other undertakings now know something about their own circumstances and they will be fairly confident about what they can expect, or at least what they should be able to expect, from the others in the light of the discussions they have had. They will negotiate with their customers and arrange their production and so forth possessing a different body of knowledge and being in a different state of awareness than if they had only their own experience, general knowledge and perception of the market to rely on.

In my opinion, it can therefore be maintained that in principle concertation will automatically trigger subsequent action on the market which will be determined by the concertation, whether the undertakings do one thing or another with regard to their market policy, that is to say regardless whether they subsequently behave in a more or less uniform way on the market. Something of this sort will, in my view, also occur if concertation in the form of exchanges of information of competitive significance, for example about actual or anticipated prices, takes place without any further coordination between the undertakings, that is to say merely on the understanding between the undertakings that all are presumed to react rationally in relation to their own and the other participants' situation. Thus, in such a case, the undertakings are in a position to assess the market situation with considerably more certainty and to act accordingly. The exchange of information will, all else being equal, entail at all events a considerable risk that market conditions will not be the same as they would otherwise have been. It is obvious that in such a case it will not normally be possible to prove any concrete, specific causal link between the acts (practice) and the concenation (the exchange of information for an unlawful purpose).

According to Article 85(1), all concerted practices are prohibited. That part of the provision is thus in the nature of a catchall provision, which in its broad terms is intended to cover all forms of competitive cooperation between undertakings other than just those belonging to the concept of agreements. The passage from the judgment of the Court of Justice referred to above in section (e) in which it is stated

may be understood in the same way.

In those circumstances if it is certain that concertation having an unlawful object has taken place and if, as I stated above, it can be assumed that the undertakings have acted on the basis of that concertation even if the Commission adduces no evidence of the concrete acts (practice), there is in my opinion nothing to prevent it from being said that there is a concerted practice with an unlawful object covered by Article 85. When undertakings act with greater knowledge and more or less justified expectations about other undertakings than they should have had and normally would have, there is always a clear risk that competition will be less intense than it otherwise would have been. Such a danger is distinctly present where market conditions are such that wholly free competition would lead to drastic falls in prices. If, as stated above in point 2, agreements having an object which is unlawful under Article 85 have the character of an abstraktes befährdungsdelikt. I find it difficult to see what is to prevent the infringement called a concerted practice having an unlawful object, which is completely parallel in that regard, from being interpreted in a more or less comparable way. The infringements have exactly the same character. In the case of a concerted practice, there must and will be, in addition to the concertation, de facto conduct subsequent to and connected with the concertation. Nothing like that is required in the case of agreements. However, in both cases the essential thing about the infringement is that traders in those cases no longer determine independently the policies they will pursue and there thus arises a very considerable risk that conditions of competition not corresponding to the normal conditions on the relevant market will be created.

Contrary to what would follow from the Commission's argument, it is therefore necessary, in my view, for action to be taken with the knowledge and the awareness that results from the concertation. The difference is, firstly, that if no action is taken at all there is no infringement. If, therefore, directly after the concertation, an undertaking has to leave the market for unconnected, extraneous reasons such as the destruction of its production plant, Article 85 cannot apply. It would, however, apply if there was evidence of an agreement having an unlawful object. Secondly, the view contended for here means that it remains at least theoretically possible to prove that the practice being pursued is not concerted in the sense that in some circumstances a party will be able to show that it has cast off the ties or bonds ensuing from the concertation and closed its mind to the knowledge gained from it. This might be conceivable where, for example, an undertaking is taken over and the new management seeks to clean up the business or is simply unaware of the collusion which had taken place. In a continuous process such as existed in the formation of the alleged polypropylene cartel, it is, of course, clear that in practice it may be difficult to argue plausibly that after collusion occurred the previous course of action was changed before the knowledge obtained from the meetings was put to use.

It is evident that, as far as the practical consequences are concerned, there will not normally be any great differences between the Commission's view and the concept of concerted practice for which I am contending. If it is assumed that concertation will have and must have a kind of automatic effect, it will generally be sufficient for there to be proof of the concertation and of subsequent conduct on the market pursued in the knowledge ensuing from the concertation. The Commission need not therefore specifically prove individual, causally-related acts. As far as the undertakings are concerned, it follows from my argument that there is in fact incumbent on the undertakings a certain burden of proof or at least a very broad obligation to provide information if it is to be accepted that the knowledge obtained from the concertation was not used in the determination of the undertaking's policy. It is also worth pointing out in this regard that the applicants admitted during the hearing before the Court that the information obtained in the meetings was useful.

The interpretation of the concept of concerted practice which I am advocating and which, as I have stated, finds no direct or express support in the case-law of the Court of Justice, is, however, in my view congruent or in line with that case-law. My interpretation of the concept is in fact consistent with the opinion of a considerable number of academic commentators and in particular with the ordinary and natural meaning of the words used in the provision. It is to be observed in this regard that there is no question of an extensive interpretation of the provision, which, because of its quasi-penal nature, would be difficult to reconcile with generally accepted legal safeguards, as the Court of Justice also indicated in its judgment in the Parke Davis case. Finally, it is to be observed that the interpretation advocated here ensures that the effet utile of the provision is not neglected, which would be one result of the interpretation which the applicants have sought to persuade the Court is correct.

(j) The structure of the two forms of infringement

In their written submissions the applicants contend that the Commission should have distinguished between an agreement and a concerted practice because these forms of infringement are different in character and in particular because the matter to be proved is different.

In the section devoted to the interpretation of Article 85 as a formal problem, I have already explained the significance of this question. However, if the conclusion which I have reached is relied upon, the parallelism between the structure of the two types of infringement is, however, in my view so great that it is not necessary to make any distinction in a situation such as that now under review.

As far as agreements are concerned, the dispute about proof mainly concerns two matters: on the one hand, whether agreements (having as their object etc.) were entered into and, on the other, whether they produced effects, as the Commission maintains. The existence of effects is in itself a matter to be proved, first for the purposes of establishing the extent to which Article 85 has been infringed because the agreement had those effects and, secondly and in the alternative, possibly for the purposes of demonstrating that the matter is of lesser gravity since the cartel had no, or little, anticompetitive effect in practice. However, the question of effects is also important as a matter of evidence, or perhaps rather as a matter of counterevidence, in relation to the existence of the agreements themselves. One of the applicants' arguments is that what may look like an agreement is not an agreement if it has not had the desired effects. Moreover, evidence regarding the existence of the agreements is also adduced on another basis, or at any rate it is sought to challenge verbally the Commission's allegation that the agreements existed.

In a quite parallel way, it is possible, with exactly the same subject-matter, for the existence of the collusion to be demonstrated, or at any rate for a line of argument to be developed, by discussing the documentary evidence available, motives and so forth, without having any regard to the market. If, as I believe, it is necessary in both cases to attach importance to the object as an independent element of an infringement category where only the time when the offence is committed is different, there is, in my view, no justification for arguing that the infringements are so different in character that a distinction has to be made.

Consequently, even though I cannot share the Commission's view of the concept of concerted practice, I do agree that the two concepts, interpreted in the way I consider correct, have such a parallel structure that the making of a distinction is unnecessary, at any rate in this particular instance. Consequently, when I later examine the evidence in the case I shall not always expressly indicate whether each specific aspect of the undertakings' conduct can be categorized as an agreement within the meaning of Article 85(1) or is more in the nature of a concerted practice.

4. May a doctrine of attempt be propounded in the context of Article 85?

As stated in the previous section, the concept of concerted practice must be understood as meaning that, as far as this component of Article 85 is concerned, the offence is only committed once conduct on the market becomes apparent and I have explained how this view differs from that held by the Commission. In my opinion, an infringement exists if it can be proved that concertation having as its object the prevention, restriction or distortion of competition has taken place and the undertakings have then taken action on the market, which must in principle, as I have argued above, be presumed to have been taken in a causal relation to their concertation if they continue to operate on the market in the relevant product or products after the concertation has taken place. Alternatively, an infringement exists where concertation, perhaps not having a competition-restricting object, has had the effect of restricting competition and the undertakings knew or should have known that such an effect would be produced.

If the Court accepts this view, I see little profit in seeking to determine whether an attempt is punishable under Article 85. Given the way in which the Commission's decision is drafted, it would, however, appear to be necessary to examine whether it can be said that an attempt is punishable, particularly where concerted practices are concerned.

Article 85 comprises, on the one hand, an offence pertaining to abstraktes befährdungsdelikt, namely an agreement having an unlawful object, and two result offences, that is to say agreements and concerted practices and the effects which they produce. In the interpretation set forth above, the concerted practice/object combination occupies an intermediate position.

As far as the offence consisting of the agreement/object combination is concerned, it is clear that an infringement of Article 85 is committed when the agreement is entered into. As far as the offence consisting of the agreement/effect combination is concerned, the infringement is committed once the anticompetitive effects of the agreement (irrespective of its object) have become apparent.

Doubtlessly influenced by the more specific substance of what he called the objective aspect of the concerted practice, Mr Advocate General Mayras, as I mentioned earlier, raised in the Dyestuffs cases the question of the extent to which a concerted practice could be caught by Article 85, that is to say before any action is taken on the market. His arguments are set out above at page 96.

If we consider the agreement/object offence, it seems clear to me from the wording of Article 85 that this concept was not intended to cover the situation where the Commission manages to break into the undertakings' negotiations before an agreement is concluded. The same applies, according to the wording of Article 85, to a concerted practice. Furthermore, elementary considerations relating to legal safeguards would militate against any attempt to introduce, without any express legal basis, a concept of attempt into the ambit of Article 85, which would in reality widen the ambit of that provision. Likewise, neither the more recent case-law of the Court of Justice nor academic literature bear out the assertion that an attempt to commit the offences covered by Article 85 would be unlawful.

I therefore consider that there is no place within the ambit of Article 85 for the introduction of a separate concept of attempt. In any event, according to the interpretation of the concept of agreement and concerted practice laid down by the Court of Justice and in view of my proposed interpretation of concerted practice, the provision has such a wide scope of application that in practice there would be no real need for the possibility of fining attempts.

5. Was the undertakings' conduct objectively inapt as a means f

During the hearing Professor Albach concluded one of his addresses to the Court by venturing his personal view that all the initiatives attempted by the undertakings had to be regarded as having no more than a placebo effect on nervous business managers. He explained amongst other things that a price agreement can be an effective strategy in an oligopoly when sale prices do not cover overheads. But, he continued, theory and experience show that implementation is difficult on a market on which there are seventeen producers and practically impossible in a period of considerable overcapacity. He then described his analyses, which led him to conclude that the so-called target prices could not be regarded as having any significant effect on the highly competitive market. The applicants also sought to show that the price instructions which were issued bore no relation to the target prices discussed at the meetings and that in any event the prices actually achieved did not correspond to the target prices. Furthermore, evidence has been produced, for example, Professor Budd's report, according to which the applicants' conduct not only had no effect on the market but was not even capable of having any effect because the market had to be regarded as resistant to the undertakings' attempted initiatives (whose existence I presuppose in this context).

All this gave rise in particular to arguments concerning the precise meaning of Object in Article 85(1).

Thus, it is maintained (by ICI) that it is clear from the very wording of the provision that it is the object, objectively assessed, of the agreement or the concerted practice which must be considered in judging the case, and not the subjective intentions of the participants in the meetings. The applicant argues that the object of the arrangements must be assessed in the light of the actual economic context in which they are made, and refers to the judgments of the Court of Justice in Société Technique Minière and Volk. According to the applicant, those judgments show that, whether it is the object or the effect which is being examined, the evaluation of an arrangement cannot be divorced from its actual effects upon competition. Thus, in its view, in order to establish that the object of the arrangement was to restrict competition, the Commission must prove that there was a sufficient probability that the object could be achieved. Moving on from the supposition that it had proved that the alleged cartel did not have and could not have effects on the market, ICI then argued that the Commission had not established that the object was to restrict competition. The Commission replied that it could not be deduced from the judgments cited that an agreement or any other arrangement had to have had an actual effect in order for conclusions about the object of the agreement or the arrangement to be drawn. Finally, the Commission argued that it had proved that the cartel had both potential and actual effects greater than was required by the judgments cited.

According to the applicant's line of argument, the conduct displayed cannot therefore create liability when it is established, or so it contends, that it was not possible for it to affect the market, regardless of what was done. If the conduct could not, as matters stood, produce any effect, it could not have the object of restricting competition within the meaning of Article 85 either.

If, for the sake of argument, we assume that the market was not in fact affected, it remains to consider how an indisputable intention to infringe the rules of competition is to be judged.

It is difficult to find in the judgments cited or in any other judgments of the Court of Justice any support for the view maintained by the applicant.

In Volk v Vervaecke it fell to be determined whether an agreement which by its nature can be caught by the prohibition laid down in Article 85(1) could still escape the prohibition for particular reasons. The Court began by stating that an agreement could fall within the provision only when it was possible to foresee with a sufficient degree of probability that it could adversely affect trade between Member States and it had the object or the effect of damaging competition. The Court went on to state that those conditions had to be understood by reference to the actual circumstances of the agreement and concluded that an exclusive dealing agreement may fall outside the prohibition laid down in Article 85(1) when it has only an insignificant effect on the market, having regard to the weak position of the persons concerned on that market. The undertakings in question in that case held only a modest market share.

The present cases do not concern minor undertakings with modest market shares where the risk of there being an actual effect on the market is slight. On the contrary, the largest petrochemical enterprises in Europe, which together account for almost the entire market, are involved. The circumstances of the Volk case were thus quite different from those in the present cases and that judgment therefore has no value as a precedent for determining this point.

It is also hard to see any connection between the applicants' situation and the judgment in Société Technique Minière. The Court of Justice stated that the object of the agreement had to be considered, taking into account the economic context in which it was to be applied and that the effects on competition had to result from all or some of the clauses of the agreement itself. Société Technique Minière also concerned an exclusive dealing agreement.

According to the traditional view, and in particular the view underlying Article 85(l)(a) and (b) of the Treaty, pricing agreements or production agreements are harmful to competition. What happens, however, should it transpire that pricing agreements cannot normally be used to defy market forces, or if this was impossible in this very case? What happens, therefore, if either in general or in specific cases pricing agreements are, or were, inapt as a means?

If that were the case, it would undoubtedly provide cause for raising fundamental questions about competition law. An applicant seeking to get off the hook by using such arguments, which appear to be real substantial innovations in the field of competition law, must, however, produce particularly cogent evidence to substantiate his theories.

In my view, it would at least have to be possible to demonstrate that amongst academic writers on economic matters there was a consensus of opinion that the market could never be affected by the means which parties seek to bring to bear. Nothing of the kind has been proved in this case. The most that can be inferred from Professor Albach's oral testimony is that, given the market conditions, the participating parties were scarcely able to exert any appreciable influence on the market. The other documentary evidence produced does not corroborate such an assertion in any substantial way either, nor has there been produced any evidence which, without having been produced for the occasion, is of general assistance in resolving the question.

For those reasons, I consider that the arguments put forward by the applicant (ICI) on the Objective interpretation of the concept of object in relation to concerted practices must be rejected in the present cases for the simple reason that the fundamental condition requiring the means to have been objectively inapt cannot be considered to have been fulfilled.

6. Is there one agreement and one concerted practice?

The applicants also object to the fact that the whole course of events from 1977 to 1983 is classified as an agreement and a concerted practice or as a framework agreement, a term which has been vigorously contested by some of the applicants.

It is difficult to see how the classification per se can have such crucial importance in the context of these cases. I therefore consider it a matter of no decisive importance whether what happened is eventually to be designated a continuous infringement or whether it is more accurate to describe it as an agreement and a concerted practice.

If one considers the original, ordinary meaning of the word agreement, it would certainly be straining this meaning to use it to describe a series of meetings and so forth at which over a period of years plans were discussed, agreed, adopted, amended, abandoned, postponed and so on.

It would perhaps be just as natural to label what is alleged to have taken place as a continuous infringement. If, as the Commission maintains, the representatives of a certain number of undertakings, who were aware of their common problems, agreed to discuss what could be done in order to prevent a ruinous price war and if in those discussions it was agreed to meet at regular intervals and that idea later crystallized so as to evolve into a quite intricate scheme designed to deal with the obvious difficulties involved in making a price cartel consisting of seventeen undertakings function on a market with significant overcapacity, I really see no reason why it should not be characterized as an agreement and a concerted practice.

The applicants contend, however, that the term framework agreement is an artificial concept meant to cover up holes in the Commission's case. I do not believe that that is so. It must, however, be acknowledged that when the existence of an agreement has been inferred from only a large volume of evidential material, one must be particularly careful not to deduce from that concept something which was not there before.

As the case has been presented by some of the undertakings, there may be reason to emphasize that one can regard framework agreement as a collective concept, which the Commission also appears to do, particularly in point 81 of the decision. Nevertheless, it must be acknowledged that the Commission's observations on this matter contained in the decision are somewhat lacking in clarity. The fact remains that no prior agreement to take measures which perhaps were not carried into effect until years later has been proved. I therefore accept the slippery slope theory which was discussed during the hearing.

The fines imposed by the Commission were for a series of initiatives on prices and on other matters as well as support measures and it is these initiatives which must be judged, irrespective of whether they are to be classified for the sake of description as an agreement or a concerted practice. As means of providing an overall description of what happened, the term agreement and concerted practice is therefore satisfactory, but those terms must not be used to cover up weaknesses in the case made out.

If, therefore, from the point of view of a continuous infringement one can identify a gap from the middle of 1978 until autumn 1979, this must be equally true of any framework agreement, which thus no longer consists of one agreement but of two agreements, with all the problems of limitation which this may entail.

In my view, therefore, the applicants' line of argument on this point cannot in itself lead to the annulment of the decision, but it remains to be verified whether the Commission has been able to prove its allegations and has not been induced by its own description of all the events which took place into deducing facts for which there is otherwise no evidence.

7. Collective responsibility under Article 85(1)?

At times in this case there has been a fierce debate on the question whether the Commission in its decision imposed collective responsibility on the undertakings in the sense that one undertaking or a number of undertakings were to be held responsible for something which others had done but in which they themselves had not participated. The question is closely linked to the description of the infringement contained in Article 1 of the decision. In the Commission's description of the infringement as an agreement and a concerted practice the applicants perceive a risk that they may be held liable for all aspects of the infringement as described in Article 1(a) to (e) of the decision, even if it were to be accepted that they did not participate in one or more aspects of the infringement.

As is clear from my observations in the previous section, there is, in my view, no justification for arguing that the existence of a prior basic agreement on practically all aspects of what is later supposed to have taken place may be presumed. At the hearing, this was indeed acknowledged by the Commission, which claimed that in the decision it asserted nothing of the kind. Therefore, for that reason alone, it is not possible to hold each individual applicant responsible for all the acts done by all the parties participating in the cartel, even if it is proved that they took part over a certain period.

It must, however, be acknowledged that in a number of places the Commission has formulated its decision in such a way that it might convey the impression that it quite simply considered all the undertakings responsible for everything that happened in the cartel during the period in which they were involved in it. Thus, in the second paragraph of point 83 of the decision it is stated:

In its defence in the BASF case, for example, the Commission has this to say on the matter:

Whilst that last statement by the Commission hits the mark without however saying very much, the second paragraph of point 83 of the decision is apt to cause confusion.

The proper description of the infringement given in the decision appears, however, as I have stated, in Article 1, according to which the infringement resides in the act of participating... in an agreement and concerted practice consisting of a number of elements. The description of the infringement as participating in an agreement, that is to say in a cartel such as that which is defined in the decision as an agreement, does not appear to be open to objection in itself. To contend that an undertaking has been party to an agreement or has participated in a concerted practice is in itself a correct description of the infringement.

When, however, the Commission states in the second paragraph of point 83 that each participant must bear responsibility not only for its own involvement but also for the execution of the agreement in its entirety, this suggests responsibility going beyond the activities in which the undertaking itself participated and it is here that the problem lies. It must, however, be added that in the same point in the decision the Commission explains that each participant's responsibility arising from its involvement in the cartel relates only to the period in which that participant was in the cartel.

It is of course correct that, as the Commission states in its defence in BASF case, the infringement of Article 85 is a collective infringement in the sense that two or more undertakings must cooperate in order to commit the infringement and that one undertaking's participation in the other's infringement is a necessary and logical consequence of its collective character. However, if the basic requirement is that a party must have entered into an agreement or been involved in a concerted practice, I do not consider that a party should be held responsible for infringements committed by others to an extent greater than that party's own participation. It is therefore simply incorrect to say that each individual undertaking may be held responsible for more than ensues from the undertaking's own participation, as the Commission would appear to suggest.

The argument about collective responsibility in this case has been long and protracted; however, it must first be established what cannot be classified as collective responsibility, even if that is perhaps the view of some of the applicants.

If one considers how the alleged cartel probably worked in practice, it is plain to see that it can be extraordinarily difficult to determine in detail the degree of involvement of each party. Who had the idea of taking this or that initiative? Who sought to persuade those others who were perhaps less enthusiastic? Who came to the meetings best prepared? And so on. It is self-evident that where there are no admissions on the part of the participating undertakings, it is often not possible to unravel all those threads in an administrative procedure in which most of the evidence is based on written documents. Of course, in cases of this nature there are also limits on how far it is necessary to go into the finest detail. For technical reasons of a legal nature, it may therefore be justifiable to make do, if necessary, with a slightly broader description of each party's participation. After all, in a case of banknote forgery, for example, it would hardly be a crucial bar to the passing of sentence under the legal system of any Member State if it were not possible to prove who filled the printing press with ink and who operated the press when each party must be presumed to have been present and to have taken part.

Ltøon a reasonable and limited application of that argument, there is no justification for claiming that some are to bear collective liability; on the contrary, it is only that the legal system resists the overstretching of its requirements to include the provision of a detailed account and detailed evidence on points on which this will rarely be possible and where the presumed perpetrators are themselves responsible for getting into such a situation, so that some relaxation of the requirements of proof must be regarded as unobjectionable. Otherwise, in many cases the Commission would in all likelihood have to abandon prosecution from the outset in cases where there is unquestionably an unlawful cartel but where it is not possible to adduce detailed proof of each party's involvement in the cartel's activities. Such a result would in practice rob Article 85 of much of its effectiveness.

Another aspect of the question mentioned above concerns the general requirements to which the Commission's evidence is subject. For example, if on the basis of an overall assessment of the available evidence it may be assumed that an undertaking was involved in a price initiative notwithstanding he fact that no written document relating to price instructions or such like has been found, there is no question at all of imposing collective liability on that undertaking; it is solely and simply a matter of assessing the evidence.

The question of what the undertakings themselves call collective liability arises only where it cannot be presumed that an undertaking has participated in one or more aspects of the infringements.

It is perhaps these problems relating to proof which show through the not very felicitous first sentence of the second paragraph of point 83 of the decision.

If, like myself, one regards the Commission's statements as merely reflecting those aspects, there is nothing to reproach the Commission for.

If the Court takes the view that there is not sufficient evidence for holding that an undertaking participated in the infringement in a certain respect, that element of Article 1 of the decision must be struck out as far as that particular undertaking is concerned. As far as I can see, the Commission did not set about adopting an all or nothing decision (see my previous observations concerning the nature of the framework agreement). Therefore, in my view, there is nothing to preclude the striking out, where appropriate, of an element, or elements, of Article 1 of the decision without this affecting the fundamental finding that the undertaking or undertakings concerned have otherwise infringed Article 85, provided that it is proved that the undertaking or undertakings in question participated in other aspects of the infringements.

Upon a reading of the Commission's decision, in particular point 78 and points 107 to 109, there is no obvious indication that the Commission did in fact impute to certain undertakings responsibility for what others had done. However, this question must be specifically examined in relation to each undertaking, which will be done at a later stage in my Opinion when the evidence concerning each undertaking is examined.

Although during the hearing the Commission consistently denied that it sought to impose collective liability on the undertakings, some passages in the written pleadings, for example the defence in the Hercules case, at pages 65 and 66, are hardly to be explained except by the fact that the Commission was alleging at least an element of collective responsibility. Thus, in the passage just mentioned, it is stated that the decision found the applicants guilty of one (continuous) agreement and concerted practice whereby the producers generally (Commission's emphasis) carried out the activities described in paragraphs (a) to (e) of Article 1.

In my view, no one can object to the applicants' firm repudiation of such a view. However, the wording in question is not to be found in Article 1 of the decision, which, together with the statement of reasons for the decision, is determinative. It is, of course, for the Court of First Instance and the Court of Justice, and not for the Commission, to provide the definitive interpretation of the decision. As expressed in its defence in the Hercules case, the Commission's view comes perilously close to making mere contact with lawbreakers or belonging to a group having an unlawful object liable to a fine and, as explained earlier, this has no basis in Article 85. If there were to be a need for such a rule, it would be for the legislature to introduce it.

E — General remarks on the assessment of the evidence and on the categories of evidence and the evidence itself

1. General

Leaving aside wholly elementary questions, evidence is difficult to deal with on a theoretical basis. There is, however, often a great similarity of views when two or more persons, whether professional judges, lay judges or others who have to deal with evidence, must come to a decision on a concrete matter of evidence.

It might therefore be tempting to proceed directly to consider the evidence. The applicants have, however, made such a considerable number of points relating to the evidential situation in these cases that it would appear to be appropriate to make a few general remarks concerning the questions of evidence.

2. Unfettered evaluation of the evidence

It is important first to point out that the activity of the Court of Justice and thus also that of the Court of First Instance is governed by the principle of the unfettered evaluation of evidence, unconstrained by the various rules laid down in the national legal systems. Apan from the exceptions laid down in the Communities' own legal order, it is only the reliability of the evidence before the Court which is decisive when it comes to its evaluation. In the present cases, only the evidence which cannot be used by the Commission against the undertakings because it was not communicated to them during the administrative procedure is to be treated as an exception to that principle, as I have explained above.

For example, the Suiker Unie case shows clearly that the Court of Justice allows only an overall assessment of a document's probative value and simple rules of evidential logic to be decisive in the evaluation of evidence (see in particular paragraphs 156 to 166 of the judgment, pp. 1939 to 1941).

The same principles must apply in the present cases. However, conclusions drawn from the evidence must never, of course, develop into ill-founded speculation. There must be a sufficient basis for the decision and any reasonable doubt must be for the benefit of the applicants according to the principle in dubio pro reo.

A very important factor in the present cases is the overall view of the evidence. It is clear that even where it is possible to give a reasonable alternative explanation of a specific document, which may be isolated from a number of documents, the explanation in question might not withstand closer examination in the context of an overall evaluation of a whole body of evidence. It must accordingly be permissible to apply, as the Commission does, conclusions drawn from periods where the evidence is fairly solid to other periods where the gap between the various pieces of evidence is perhaps larger. After all, there needs to be a particularly good explanation to convince a court of law that in a particular phase of a series of meetings things occurred which were completely different from what had transpired at earlier or subsequent meetings when the meetings were attended by the same people, took place under similar external conditions and indisputably had the same primary purpose, namely to discuss the problems within the industrial sector concerned.

3. Oral evidence

In the administrative procedure the Commission does not have the power to compel persons to give evidence under oath. That is one of the reasons why the Commission's decisions in competition cases rest to a large extent on documentary evidence. The same is true of the present cases. However, pursuant to Article 3(3) of Regulation No 99/63, undertakings may ask for persons to be heard. In the polypropylene cases, however, as in many other cases as far as I know, there was no question of the persons directly connected with the events in question explaining what happened at the meetings and what they meant in the notes they wrote. On the contrary, on all essential points the undertakings' internal and external legal and economic advisers and representatives have put forward the undertakings' views.

The Commission has relied on that fact in support of its view that if those persons were not heard there is less reason to give credence to the various explanations which some applicants — with, moreover, outstanding forensic skill — have advanced in order to prove that the documents produced in evidence do not demonstrate what they appear to indicate at first sight. I consider that the Commission is right to take that view. It is remarkable that it should have been necessary to put forward alternative hypotheses as to the true context of the case when persons connected with the undertakings, and in whose employment a large number of them presumably still are, could explain what all the documents really meant.

No decisive weight should, of course, be attached to the fact that certain explanations have not been provided, but, in my view, the point made by the Commission is at any rate an evidential factor which is at odds with the applicants' attempt to weaken the Commission's case.

It could be claimed, as Shell did, that so little confidence was held in the oral part of the administrative procedure before the Commission that it was thought wiser not to take part in it. But what was to prevent the applicants from requesting that the persons concerned be heard before the Court of First Instance pursuant to the Rules of Procedure? No such request was made, however. The fact that the applicants, who clearly have difficulties in producing an explanation, do not request the hearing of witnesses capable of providing an explanation does not tell in their favour.

4. Documentary evidence

Most of the applicants have contested the reliability of the records of the meetings and the notes upon which the Commission has founded its decision.

The most important body of the Commission's evidence is in fact the written accounts, including the notes, found on ICI's premises. The Commission claims (point 70, fifth paragraph, of the decision) that the records of the meetings give an exact, trustworthy and consistent account of what took place at the meetings. The Commission further emphasizes that for their part the applicants have not put forward evidence which might cast doubt on the reliability of the documents on which the Commission relies (point 70, first paragraph, of the decision).

On this point, it is contended by the applicants, for instance by ICI, that as a means of understanding what was happening in the applicant undertakings generally, the notes must be regarded as seriously deficient. It is maintained further that the notes had the purely limited purpose of informing the colleagues, within each undertaking, of those who attended the meetings of what had been said at the meetings and what could be deduced from this, and that after the meetings the notes were supplemented by analyses and comparisons for the author's own use.

It is further claimed that even when the notes recorded faithfully what was said, they are far from being reliable as a record of the defacto situation in the industry or of the intentions of the individual producers. In this connection, I refer to the quite frequent remarks about the lack of trust between the parties and the fact that the undertakings' subsequent conduct demonstrates a lack of will to keep to the agreements, which, according to the notes, were entered into at the meetings.

For its part, the Commission refers in particular to the fact that the documents were drawn up directly after the relevant events and that the person or persons who took the notes had no reason at all not to give a correct account, just as the contents of the documents, according to the Commission, do not bear any trace of exaggeration. As mentioned above, the Commission claims that none of the undertakings asked for the persons involved to be heard pursuant to Article 3(3) of Regulation No 99/63.

In assessing the evidential value of a reporting document regard should be had first and foremost to the credibility of the account it contains. Regard should be had in particular to the person from whom the document originates, the circumstances in which it came into being, the person to whom it was addressed, and whether, on its face, the document appears sound and reliable.

Quite generally, it is not at all improbable for the representatives of undertakings to meet in order to enter into agreements or to concert their practices in the way alleged by the Commission. As we know, the Commission has successfully argued before the Court that this is what occurred in many cases.

The documents in the present cases originate from employees who have to provide their colleagues and superiors with an account of what took place at meetings which they attended. If we exclude the Hercules case, there is no indication that those persons did not fulfil their duties quite loyally and to the best of their abilities. Under the general rules of evidence, the fact that the documents were drawn up immediately after the meetings and clearly without any thought for the fact that they might fall into the hands of third parties must be regarded as having great significance. An overall evaluation of the contents of the reports of the meetings shows, in my view, that they were drawn up with care by persons who were fully acquainted with the matters they were describing, setting out in concise, measured and matter-of-fact language what the person or persons who drafted the reports understood from the meetings. They are clear and logical and do not bear any impression that the persons in question may have seriously misunderstood or misinterpreted what took place. It must also, of course, be regarded as improbable that large industrial undertakings would send their employees to meetings which the undertakings certainly considered important if those employees were not capable of reporting what happened at the meetings in a sensible and reliable way.

As far as the ICI case is concerned, it is maintained, as mentioned above, that the notes of meetings were very often supplemented by analyses and additions for the author's own use. That is not immediately apparent from the notes and despite the Commission's request to this effect ICI has not specified what the supplementary material and analyses were. Furthermore, and this must be repeated, ICI has not requested that the person or persons in question should be heard in order to substantiate its claims.

In view of the foregoing, I consider that there need be no hesitation in assuming that the notes are a reliable source for understanding what took place at the meetings and given their ordinary natural meaning they thus provide a basis for surmising the significance of the matters discussed.

5. The economic analyses

Economic analyses often make up an important part of the evidence in competition cases and can be of great value to the Court in understanding the relevant economic context. It is thus important to obtain information about how an oligopolistic market might react in different circumstances. But — and this is the important point — the findings of economic experts cannot take the place of legal assessment and adjudication. Thus, when Professor Albach makes his observations about what target prices might be in an economic context, it must be emphasized that his views are not, and cannot form, a legal assessment. Even if it were to be found that there were no significant effects on the market, that does not prove that no agreement was reached or that no exchange of information took place with a view to regulating prices. It is for the Court to consider what is prohibited under Article 85(1) and the evidence for the commitment of prohibited acts, and not for economic theorists.

On the basis of an overall assessment of the evidence, the Court must therefore consider whether what took place constitutes prohibited conduct, even if the expression target prices from the point of view of economic experts might conceivably mean something other than unanimity on the prices which all should strive to obtain. It is the content of the documentary evidence which must show whether the persons attending the meetings had the intention of influencing prices or whether they simply wished to tell each other what they thought were reasonable prices on the basis of market evaluation and it is the Court which must determine were necessary whether it is unlawful for parties to inform one another over a very long period of time about what they think the market can bear.

6. Target prices and volumes as the subject-matter of agreements or of concertation

According to the Commission, it was sought to implement the alleged price initiatives by applying so-called target prices. In assessing the evidence before the Court we must first establish what the term target prices probably meant to the undertakings during their discussions at the meetings. Once that meaning is established, it must be decided whether target prices are caught by the prohibition in Article 85(1).

On this question the Commission states, for example, that the jargon used in the documentary evidence relating to the meetings speaks for itself. The Commission is entitled, it says, to take a common sense approach and to interpret the expressions used in the documents in accordance with what they actually say. Target prices, sutes the Commission further, must ordinarily be understood as reflecting a specific price level which the producers had agreed to attain as far as the market allowed.

In my view, too, there is no other way to interpret the notes of the meetings.

However, it is perhaps the very term target prices which lies at the root of much of the confusion characterizing the proceedings in these cases. For, as the Commission says, the term target prices, as used in the notes of the meetings and so on, must precisely be understood as reflecting agreements to the effect that the undertakings should as far as possible attain a certain price level. As proved to be the case, and as the Commission is also aware, it was often difficult, and, according to the applicants, impossible, for them to attain the prices to which they aspired. Indeed, the applicants believe that the market did not allow them to implement any target prices and that the target prices were actually determined by the market.

Hence the interminable debate about whether it was the applicants who were steering the market or the market which was steering them.

It accounts, too, for some of the impressive presentations of evidence by the applicants in the form of audits and reports by economic experts and so forth, which all tend to show that the prices striven after were not obtained.

The next link in the chain of argument of most of the applicants is the attempt to show that they had not come to any agreement or at any rate did not regard themselves as bound by any agreement, which is purportedly proved by the very fact that the prices they were together striving to achieve were not obtained.

The applicants have thus in fact set themselves the very difficult task of conjuring away the very certain and well-documented content of the negotiations and the impression of them created by the notes of the meetings. In my view, they have not succeeded in that task and I think that there need be no hesitation in assuming that target prices and so forth were agreed to, probably in the form of a gentlemen's agreement, as alleged by the Commission and to the extent that is otherwise borne out by the documentary evidence.

The fact that there could be any such discussion at all and that the applicants devoted such considerable resources to it is possibly due to the elasticity of the meaning of the term target prices viewed in relation to the rules of competition.

Some would perhaps argue that an agreement solely designed to strive to obtain a specific price as far as the market will bear is not per se a manifestation of anticompetitive conduct.

Yet, to judge by the notes, that is not what was intended. On the contrary, the purpose was to obtain a higher price than the participants reckoned on or could have obtained. The intention was clearly to attempt to push the price level upwards. The object within the meaning of Article 85(1) emerges quite clearly.

In view of the foregoing, it cannot matter, in my opinion, that the very term target prices is so conveniently vague that in the mutual relationships between the parties participating in the cartel it would be difficult to determine when a participant had ignored such an agreement. The fact that the agreements were akin to gentlemen's agreements because of their rather uncertain and vague substance and thus could not have served as a basis for determining the parties' reciprocal legal and moral obligations, is, according to the case-law of the Court of Justice, not decisive.

Therefore, it can in itself only be of minor importance whether the undertakings themselves considered that they were under any obligation towards the others, since the substance of the obligation was from the outset rather indistinct.

The conclusion to be drawn from the above must be that even a type of agreement which, according to its substance, takes account of actual market conditions, as it undoubtedly must, is covered by Article 85(1). From the evidential point of view, it can, moreover, be concluded that any non-application or incomplete application in practice of the prices striven after is not suitable counterevidence in the face of agreements of that type.

Many of the applicants do not in fact dispute the factual circumstances, but only their legal consequences. Thus, ICI, for example, states that the targets discussed at the producers' meetings were consensus recommendations to which the producers aspired with different degrees of enthusiasm.

In my view, that explanation lies square with a description of an agreement or concertation prohibited under Article 85, which shows that to a certain extent only a problem of definition or categorization is involved.

The entire case does in fact show that the undertakings did not feel themselves bound by the agreements entered into or the concertation which took place in the same way in which they would have felt themselves bound by an agreement to deliver a consignment of polypropylene to a particular customer on a particular date. When customers went elsewhere in order to seek a lower price after a price rise had taken place, those in charge of sales could obviously not tolerate a cessation or serious drop in sales. The notes of some of the meetings show quite clearly, however, that to a large extent it was felt necessary to endeavour to work out a common course of conduct, even though there were obviously often some whose solidarity was in doubt.

Generally, it is not possible to establish which undertakings snowed a greater or lesser degree of solidarity, even if some individual undertakings were in the spotlight more than others; moreover, they need not always have been the same. It is possible that one or two undertakings demonstrably had more inclination to cheat than the others, but the picture is far from clear and the fact that the undertaking or undertakings in question possibly looked after their own interests at the others' expense cannot, in my opinion, be taken as an indication that they did not take part.

Exactly the same considerations apply as regards volume targets and quotas.

F — Is the existence of a cartel proved?

1. Introduction

As mentioned above, for procedural reasons the question of the existence of the alleged cartel must be decided on the basis of the documents which the Commission communicated to all the undertakings in the manner required by the case-law of the Court of Justice. The evidence which may consequently be used to establish the existence of the cartel essentially consists of the 101 pieces of documentary evidence which were sent with the general part of the statement of objections as well as the documents enclosed with the Commission's letter of 29 March 1985.

2. The alleged floor-price agreement

According to the information available, at the end of 1977 a series of meetings began, attended by a number of polypropylene producers who supplied the European market. The meetings continued to take place until the end of September 1983. They were held with increasing frequency, ^'ius, the applicants do not seriously dispute that in 1978 six meetings appear to have been held, in 1979 an unknown number, but probably fewer, in 1980 it would seem that there were six meetings, in 1981 ten meetings, in 1982 twenty-three meetings and in 1983, until the end of September 1983, fifteen meetings. According to the Commission, the basic aim of the meetings was to attempt to push polypropylene prices upwards by the setting of so-called target prices and by various support measures which were to serve to achieve that aim.

Prior to that series of meetings, there had, in the Commission's view, been a period in which a number of undertakings had sought to raise the price of polypropylene through an arrangement under which some undertakings allegedly agreed to apply a so-called floor price, and similarly other undertakings are alleged to have made arrangements together in a manner which constituted an unlawful concerted practice.

The Commission's basic evidence as far as the floor-price agreement is concerned is a note supposedly made some time during the first half of 1977 — possibly in June — by the Marketing Director of Hercules. The notes concerned were made by the Marketing Director in the course of a telephone conversation with an unidentified person. In the document, which is rather fragmentary — as notes of telephone conversations often are — and which is not immediately comprehensible in all its details, it is first reported that the major producers have made an agreement. Besides that statement are the names of the undertakings Monte, Hoechst, Shell and ICI. Next is written, in telegraphic form, System floor prices — Floor prices from July 1 — definitely August 1 when present contracts expire — Floor prices for 4 month period only. Then follows a remark about an alternative, which possibly refers to something then existing. In between those notes there are some remarks about importers which do not appear to belong in any context of which we otherwise have knowledge. Thereafter there is a sentence which probably means that a meeting is to be held in October to review progress. Finally, there is a remark to the effect that tonnage restriction would operate next year, possibly — this is at any rate what the Commission believes — subject to the floor-price system working. At the end of the note there are columns with price information in French, British, Belgian and German currency for various grades. It does not look as if what is written on the next page is in the same hand and its more or less illegible text does not seem to have been relied upon by the parties.

We know therefore that the Marketing Director of Hercules received information from an unknown source to the effect that an agreement had been reached between the Big Four the substance of which is outlined in the telephone notes — although not as coherently as the Commission's decision gives the impression.

It thus appears that there existed what the Commission later calls a core agreement (see the letter of 29 March 1985, p. 3) between the four major producers and that Hercules had knowledge of it. We learn nothing further from that document and in particular we learn nothing about other producers. It is, however, not unreasonable to infer from the wording of the document that an agreement was entered into between the four major producers in particular and not between other undertakings. Indeed, that possible interpretation is put forward by the Commission both in its letter of 29 March 1985 and later as an alternative.

The four major producers make the following comments on the document and the alleged agreement:

Montedipe denies that there was an agreement under Article 85(1) within the meaning which this applicant accords to the term, which is that an agreement exists only when the parties unequivocally express their intention to bind themselves and any subsequent conduct by them clearly reflects that intention. The applicant refers on this point to the Chemiefarma judgment, possibly paragraph 112. Whatever one might think of Monte's interpretation of that judgment, the interesting thing, however, is how that undertaking itself describes the events in 1977. Monte says inter alia that that single document demonstrates neither an agreement nor a plan. The reality was different, it says. It continues (application, p. 100):

In the course of those contacts, Monte states further, the prices which would have enabled those difficulties to be alleviated were indicated.

On this point, the Commission contends that in the first place there was no reason for the undertakings to calculate between themselves a price which would have covered costs. Each could have done so independently. The Commission further states that ICI and Shell have acknowledged that the purpose of the action in question was to check the slide in prices.

In my view, Monte's observations must in fact be taken as an acknowledgement of the factual circumstances on that point. As the Commission points out, it does not stand to reason that very large undertakings should afford each other assistance merely for the purpose of carrying out simple price calculations. Things can always be called by a name different from what they are normally called and the joint indication of prices is at any rate normally an expression for the coordination of the undertakings' pricing policies. In my view, there must be extremely strong evidence for concluding that such a joint indication had no significance as far as the undertakings' pricing policies were concerned, and such evidence is not present.

ICI also broadly admits the factual circumstances. It thus does not deny that the producers were in telephone contact in the crisis year of 1977 and that the discussions might have been about the collapse of Western European prices. It admits that it is also possible that some producers may have suggested the desirability of a minimum price level in order to halt the slide in prices. ICI then refers to the abovementioned note, which it does not, however, find cogent because the price in fact fell to DM 1.0/kg.

On this point, the Commission states that the note appears credible per se and that both ICI and Shell have admitted that there was telephone contact between the producers for the purpose of finding a way to halt the slide in prices. With regard to the actual prices, the Commission mentions that in November 1977 Monte announced a price increase very close to that indicated in the note and within a few days the three other undertakings mentioned in the note gave notice of similar price increases. Finally, ICI and a number of other producers publicly lent their support to the initiative. In the result, the actual evolution of prices shows, according to the Commission, that the price initiative was successful.

Shell admits that discussions took place between itself and Monte in or around November 1977 and that Monte might have mentioned the possibility of implementing price increases and might have sought Shell's views on its reactions to any increase. In Shell's view, that is not sufficient to establish an infringement of Article 85 on that point. Nor, in Shell's view, does the note on the floor-price agreement give sufficient grounds for establishing an agreement as alleged by the Commission. In support of its view Shell refers to the document of 6 September 1977, which is mentioned in the fifth paragraph of point 16 of the decision, according to which a meeting took place on 30 August 1977 in Belgium between Shell and Solvay in order to discuss the price of polypropylene. In Shell's view, the document goes to show that it was not guilty of any infringement. Shell refers principally to the fact that, according to the notes of the meeting, the undertaking's policy was to maintain coûte que coûte its sales to a number of customers in Benelux and that Shell saw no solution other than to try to develop the polypropylene market by innovation in conversion processes and development of applications. Furthermore, Shell emphasizes that those attending the meetings did not refer, directly or indirectly, to any agreement on prices. The collapse in prices was a major concern to both companies, and the absence of any such reference would be extraordinary if the Commission's speculation about a floor-price agreement were correct.

The document referred to, which was first submitted in connection with the hearing before the Court, is not to be found among the 101 pieces of documentary evidence and was not, moreover, communicated to Shell. Therefore — as was recognized by the Commission at the hearing — it can be used solely in Shell's favour.

For its part the Commission insists that the note concerning a floor price appears reliable, especially seen in the light of the other information available. Furthermore, it is the Commission's view that the Solvay document from September 1977 shows nothing to indicate that there was no agreement on prices and nothing at all to suggest that prices were not discussed.

Hoechst does not directly address the question of the extent to which there was contact between the undertakings in 1977 nor whether price discussions might have taken place. On the note itself, Hoechst says that it is not unequivocally apparent from that note that the term agreement that was used was intended to signify an agreement within the meaning of Article 85. One could equally well imagine that the term meant only that there was unanimity on the need to raise prices. Finally, according to Hoechst, it is not excluded that the person who spoke with Hercules' employee let it be understood that there was consensus between the Big Four so as to obtain Hercules' support for his own efforts. Hoechst adds that the Commission's assertion that the price increase was postponed is not supported by the documentary evidence.

On this point, the Commission points out that the Hercules note is an authentic document which was drawn up at the time of the events alleged and its accuracy is not subject to any doubt. Furthermore, the note is thoroughly consistent with the uniform price increases which a number of producers announced at the E. A. T. P. (European Association for Textile Polyolefins) meeting in Paris on 22 November 1977 and in the trade press.

As an indication of the note's factual accuracy the Commission also refers to the fact that Hercules, which was therefore aware of the alleged agreement, had already expressed a view that it was for the traditional industry leaders to bring some order into the chaos that reigned. It is, however, the events in November 1977 mentioned above which, in the Commission's view, corroborate the substance of the note on the agreement between the Big Four. On Friday 18 November 1977 Monte announced a price increase in the periodical European Chemical News, reported as a minimum price of DM 1.30 for raffia and DM 1.40 and 1.50 for two other grades. On Tuesday, 22 November, a meeting of the E. A. T. P. organization was held in Paris attended by both producers and consumers of polypropylene.

At the meeting Hoechst announced that it had given notice of a price increase the day before. ICI declared its support for Monte's initiatives; its representative could give no specific details of its proposals but said that its customers would shortly be informed. Shell too expressed itself in a way which can be viewed as support for the initiative. In addition Rhône-Poulenc, Hercules, Chemie Linz, Saga and Solvay declared their support for the initiative. Chemie Linz announced that it would have to follow those new prices. Hercules' representative said that even though he was not yet aware of the details of Monte's initiative and did not know at all whether the initiative would be supported by the other recognized industry leaders, it would be Hercules' position to give complete support to any effort to increase polypropylene prices to a reasonable level relative to manufacturing costs. Rhône-Poulenc mentioned that it had learned of Monte's planned price rises the previous Friday from the press. According to Rhône-Poulenc, it was not possible to return, in one go, to the economically acceptable level, which was around FF 3.50, but Rhône-Poulenc had decided to follow Monte's lead so that its basic price as from 28 November 1977 would be FF 3.00. Solvay announced that it would fall in line with the prices quoted by the main producers.

What thus occurred was subsequently reported in the same week's edition of European Chemical News under the headline Montedison PP price rise backed by European producers.

3. The continuation of the floor-price agreement

The price of polypropylene subsequently rose in the following period to DM 1.20/kg for basic grade raffia and in April 1978 European Chemical News reported that Montedison was planning a price rise of not less than 10% before the summer holidays, probably in June.

At an E.A.T.P. meeting on 26 May 1978, Monte announced that the price increase which had taken place in November 1977 had not brought the industry back to satisfactory price levels and that the November initiative could therefore be considered only a first step which had to be followed by a second one, probably during the second half of 1978. At the same meeting DSM said that it would support the movement towards bringing prices up to a reasonable level. Hercules said inter alia that the need to bring prices up to a minimum level of DM 1.30/kg had not diminished and would not do so. ICI's representative said inter alia that ICI would give its full support to reaching the price level announced in Paris. Rhône-Poulenc expressed the view that the earlier price rise constituted only one step and that a further increase of about 10% would be necessary. Shell did not express any clear view on prices. Solvay emphasized that current prices did not allow for a sufficient profit to producers and that prices should continue to be increased.

4. The meetings between the producers

According to the information provided by ICI, before the meeting in May 1978 referred to above a group of producers had begun to meet in order to discuss the problems in the industry. The meetings began, according to ICI around December 1977. ICI states that in 1978 about six bosses meetings took place and that the idea of target prices was developed at the earlier meetings held in 1978 (thus not necessarily, as the Commission says, at the beginning of 1978). Referring to the fact that six meetings were held in 1978 and that the idea of target prices was developed in 1978, the Commission therefore takes the view that the solidarity expressed at the E.A.T.P. meeting in May 1978 must have been based on a previous agreement between the producers.

5. Assessment of the situation in 1977

In the period before 1978 there was a sort of opening phase which to some extent can be considered separately, as the Commission does. In terms of evidence, that phase is characterized by the decisive weight attached by the Commission to the abovementioned note concerning the floor-price agreement of the four major producers. As will be seen, thereafter the evidential basis on which the Commission founds its decision shifts inasmuch as, so far as the period from the end of 1977 or the beginning of 1978 and thereafter is concerned, decisive weight is attached to the fact that meetings took place at which it is not disputed that the producers discussed the problems faced by the sector.

As far as I can see, the Hercules marketing manager who received the telephone call undoubtedly obtained some important information and in the light of the telephone note there is no reason to doubt that there was an agreement or at least an understanding between the four major producers. The note is very specific as regards the prices to be attained and even the level of the price specifications corresponds fairly well to that which, in Spring 1978, became the actual average price and which, both in Autumn 1977 and Spring 1978, came to be regarded as an acceptable minimum price, namely DM 1.30/kg. It is difficult to imagine that the information contained in the note could have been pure invention. Likewise, some corroborating evidence is required before it can be assumed that someone planted the information with Hercules in order to induce it to join something which had not yet come into being. Furthermore, at least two of the four major producers have largely admitted the factual circumstances. The document on which Shell relies as evidence of its noninvolvement in anything unlawful does not, in my view, lend sufficiently solid support to its claim. It is true that no price agreement is mentioned in the document but if the agreement did have the four major producers as participants, as the Hercules note indicates, it cannot be surprising that price agreements were not expressly discussed with Solvay. Furthermore, one must, of course, always be careful not to attach too much weight to what is not contained in a document.

Consequently, I believe that we must proceed from the assumption that in the middle of 1977 there was an agreement or concertation as indicated in the note. We do not know what happened from then on until November. We do know, however, that Monte then increased its prices and very quickly received the other major producers' support, that price increase largely corresponding, as mentioned, to the indication in the note. We know, too, that the price did actually rise to approximately DM 1.20/kg. In view of those facts it is not unreasonable to make the further assumption that the original price agreement was postponed for later implementation, as maintained by the Commission.

The Commission believes that this constitutes the core agreement discussed previously and it asserts that the fact that Rhône-Poulenc, Hercules, Chemie Linz and Solvay lent their support to the price initiative at the Paris meeting is a strong indication that those producers were implicated in the agreement/concerted practice. That is further corroborated, according to the Commission, by the fact that ICI and Shell admit having been in contact with other producers although it was not possible to establish which. According to the Commission, another possibility is that only the four major producers entered into the core agreement and that the others did not take part in its elaboration and had not even expressed their endorsement of it to the Big Four before the E.A.T.P. meeting. They did, however, according to the Commission, at all events participate in a concerted practice by knowingly cooperating with the Big Four on the implementation of that agreement.

As far as the Big Four are concerned, there exist, as I mentioned, the necessary grounds for regarding them as guilty of an infringement as the Commission maintains. There does not seem to be any reason for doubting the accuracy of the Hercules note and given its ordinary meaning it fully bears out the Commission's view. Furthermore, two of the undertakings expressly mentioned in the note have in fact admitted the facts. The subsequent events tally very well with the content of the note and it is not difficult to imagine that the implementation of the price increases had to wait until the time was right, which was also the case on several later occasions.

As regards the other undertakings which the Commission considers to be implicated, the position is different. The Hercules note expressly mentions the Big Four and it is therefore logical to assume that the others were not involved. Hercules obtained information about the agreement and probably others, too, but we do not know which. All things considered, we know nothing very much about what took place from that date until November, when Monte announced its price increase. So, the question is whether the undertakings which at the meeting in Paris lent their support to the price initiative can be presumed between Friday 18 November and Tuesday 22 November to have reached a decision of their own accord to follow the initiative, possibly in the knowledge that the Big Four were in agreement. The alternative possibility, of course, is that the whole thing was stagemanaged for the benefit of the undertakings' customers and the public.

I do not believe that it is inconceivable or improbable that in a situation like that prevailing at that time quick decisions were taken to follow the leading producer, all the more so since on 22 November the undertakings had their own forum available to announce their intentions. As I have mentioned, we know that some undertakings had had discussions with the Big Four, but we do not know which, and identification would be necessary in order to establish that particular undertakings were actually involved. It is clear that there might well have been discussions going on in the corridors in connection with the meeting in Paris and that decisions might have been taken after telephone conversations, but we just do not have any evidence at our disposal. It must therefore, I think, be assumed that the other undertakings, that is to say Rhône-Poulenc, Chemie Linz, Hercules and Solvay, took the decision each for themselves to follow Monte.

The same is true, in my view, even if it must be assumed that the said undertakings were aware of the Big Four's (unlawful) agreement, which Hercules must be considered to have been. For an undertaking to align itself to a price level in the fixing of which it was not involved cannot, in my opinion, be unlawful per se, any more than it can be unlawful to follow a price-leader and indeed what is involved here is a form of price-leadership put into effect by the Big Four acting together. If the position were otherwise, it would mean that there would be a prohibition against undertakings' reacting intelligently to the existing market, which cannot be the intention of the competition rules (see the judgment of the Court of Justice in the ICI case, paragraph 118).

For those reasons, I cannot agree with the Commission's alternative or subsidiary view as to what occurred. It is correct that at the E.A.T.P. meeting the undertakings expressed their support for price increases and the aim of achieving a price level of DM 1.30/kg. But if it may be assumed that the decisions were taken individually, is it then unlawful to give notice of such a decision in a forum which may be described as public? A priori that should not be the case. Of course, the possibility cannot be excluded that on the basis of an underlying understanding public channels are being used to exchange signals which, taken as a whole, must be considered a concertation of a practice, but in the present cases there do not seem to be sufficient grounds for believing that was the case.

As far as 1977 is concerned, it should accordingly be held that the Big Four — Monte, ICI, Hoechst and Shell — participated in a floor-price agreement as the Commission alleges, whereas there is no evidential basis for holding that the others did likewise.

6. Assessment of the situation in 1978

As mentioned, the Commission considers that the solidarity expressed at the E.A.T.P. meeting in May 1978 can be traced back to the secret discussions which, in the Commission's view, may have taken place at the bosses' meetings held in 1978 and attended by a significant number of the producers. As further corroboration of its supposition, the Commission refers to ICI's explanation that the idea of target prices was developed in the course of 1978.

Finally, the Commission contends generally that the subject-matter of the meetings for which records are available can broadly be assumed to provide a measure of the subject-matter of the meetings about which no detailed information is available. That view is generally disputed by the applicants. More specifically, it is alleged (by Rhône-Poulenc) that the Commission has failed to take account of the fact that the cartel evolved in the course of time, starting from perhaps a more modest beginning and later turning into a more refined system.

More direct evidence, such as to that available for later periods, is missing for 1978 and the sole price initiative that the Commission considers it has found is the second part of the floor-price agreement of 1977. It is, of course, suspicious if competing undertakings hold more or less regular meetings to discuss the market, market demand, the balance of supply and demand and the price level for individual sectors of the market, as ICI explains. It becomes even more suspicious when target prices, target quotas, market shares, compensating arrangements and account management systems are discussed, as ICI also mentions. It is clear that, if something unlawful occurred in 1977 and, as I shall discuss later, there is evidence of unlawful activity in 1979 and thereafter, this must cast a dubious light on the first meetings of which we have knowledge. On the other hand, we know from an overall assessment of the cartel that it evolved into an increasingly finer network of arrangements for realizing the undertakings' price aspirations.

The question is, accordingly, from which point in time it is defensible as a matter of law to find that the discussions which took place condensed into common projects to such a degree that the situation may be considered to be caught by Article 85(1) as an agreement or a concerted practice. We know from ICI that the idea of target prices originated during the course of 1978, but it was not possible to specify exactly when. We also know from ICI that until the end of 1978 or the beginning of 1979 only so-called bosses meetings' were held, whereas the experts' meetings between lower-ranking employees with more knowledge of sales first began in 1979.

In my view, the course of events of which we are aware, does not leave much doubt that discussions were held in 1978 which had some sort of unlawful object and later led to something unlawful taking place, but in the final analysis there must be evidence that the discussions led to concrete results in one form or another in the shape of agreements or concerted practices. We do not, however, have such evidence in respect of this period.

In view of the foregoing, I consider that it must be concluded first of all that there is not sufficiently solid evidence for finding that the target-price schemes were put into effect by the beginning of 1978. It is thus, in my view, doubtful whether it can be regarded as proven that the statements made at the E. A. T. P. meeting in May 1978 represented a prearranged move. At all events, as far as the producers other than the Big Four are concerned, there is, in my view, no solid evidence at all. Just as in the case of 1977, their support for the price increase can very easily be explained as a reaction to the price increase announced by Monte, possibly reinforced by some passive knowledge of an agreement entered into by the Big Four.

As far as the latter are concerned, there may be less doubt in concluding that they continued the unlawful activity in which we must assume they engaged in 1977. With regard to these four major producers it is, however, also true that there is in any case a marked lack of evidence concerning what actually occurred for a not insignificant part of 1978.

To summarize, I therefore conclude that the Commission has not adduced any evidence of agreements or concerted practices contravening Article 85(1) as far as any of the undertakings are concerned, at least as regards a significant part of 1978.

I shall return to this matter latter, but as far as the four major producers are concerned, the liability to a fine for matters which the Commission has, in my view, demonstrated with regard to 1977 and the beginning of 1978 must be regarded as time-barred.

7. Assessment of the situation in 1979

The documentary evidence concerning 1979 consists of tables found on ICI's premises, discussed in point 54 of the decision, and a note concerning a meeting held on 26 September 1979. In addition there are the price instructions which concern the first of the price initiatives mentioned in the decision.

In the first table (annex 55 to the general part of the statement of objections), which is headed producers' sales to West Europe (Turkey not included) are set out, in the three first columns, the producers' sales in the years 1976 to 1978. These are the figures for all the applicants in the present cases as well as for the other West European producers. In a fourth column there are the figures for the year 1979 with the additional word actual. In the fifth column there are corresponding figures under the heading revised target 1979. The next table (annex 56), which is divided up into a main column under the heading Sales and columns headed + % gg. 1979, Quote 1980 and Proposal I and Proposal II and Proposal III, probably originates, according to ICI on whose premises the table was found, from one of the German or Austrian producers. In the latter table are calculated, according to ICI, first the producers' actual sales to West Europe in the years 1976 to 1979. Again according to ICI, the column + % gg 1979 contains a calculation of the individual undertakings' figures for 1979 with the addition of either 5 or 10%. The column Quote 1980 is the author's suggestion for target volumes for 1980. The columns Proposal I, II, III were intended to be filled in by ICI's representative. The column Proposal III contains the final target quotas suggested as the quota for the year 1980.

The figures in the first table's column 1979, actual corresponds exactly to the column for 1979 in the second table. The figures in the first table's column revised target 1979 differs from the column 1979, actual by a few percentage points (up to 5-7). The same figure for 1979, actual recurs in other tables.

As far as production volumes are concerned, it is first stated in the note of the meeting of 26 September 1979 that it was accepted that a fixed quota system was important and then follows the statement: go for 80% scheme as per recent Zurich note. With regard to this latter cryptic remark, ICI explains (Article 11, response, p. 23): At a meeting in Zurich the producers who were represented took total industry sales made in West Europe during the first eight months of 1979 and averaged these in order to arrive at a monthly rate. It was considered that monthly demand in the last three months of that year would be at no more than 80% of the average monthly demand in the first eight months of the year and accordingly at the meeting it was suggested that producers should limit their volume ambitions to take into account this fact.

A further document concerning the French producers, which the Commission mentions in its decision at point 54 (3), must (see above, Section A, 5, c) be held inadmissible on procedural grounds because it was not sent to all the undertakings.

A quota arrangement for 1980 is relatively well documented, and I shall return to this point later. There can scarcely be any doubt that the term revised target for 1979 must mean that there was some arrangement relating to some part of 1979. We do not know when the first of the tables described above was drawn up, but a comparison of the actual figures for 1979 with some of the revised target figures for 1979 strongly suggests that for the author it was intended as a subsequent review to see whether the figures tallied or not. Logically, there is, of course, nothing to prevent target sales figures from being set in the course of 1979, possibly late in the year, and, possibly, revised target figures, too. The record of the meeting of 26 September shows in any case that at that point in time the problem was being tackled.

As described in the following section, there is, on the evidence available, no reason for doubting that there was a price initiative in the last half of 1979. Any quota arrangement alleged to have been introduced by the beginning of 1979 is, however, in my view not sufficiently well documented. It may be assumed that there was some sort of arrangement in the final months of 1979 but — according to the evidence to hand — not the quota agreement alleged by the Commission.

In the light of the foregoing, I consider that — if the Court is otherwise in agreement with my evaluation of the evidence concerning what occurred later — it should hold it an established fact that the first infringement, apart from the floor-price agreement of 1977, took place at a time after the middle of 1979 and not, as the Commission contends, in 1978 and at the beginning of 1979.

8. The price initiatives

(a) The price initiative from July to December 1979

The Commission's main evidence concerning this price initiative is a note of a meeting held on 26 and 27 September 1979. The note, which was found on ICI's premises, mentions first the level generally of 1.70-1.75 DM, 4.0./4.1 FF and 26-27 BFR. It is then stated that 2.05, which indisputably must be DM, remains the target. Next, it is stated that that price, that is to say 2.05, is clearly not achievable in October nor in November and that the plan now is 2.05 on 1 December. Finally, as far as prices are concerned, it states 1 Nov. 1.90 or 1.95 — yet to be decided — but decision will be swift, whatever that might mean in the context.

As to the Commission's further evidence, it must first be observed that the documents originating from Shell mentioned in the second paragraph of point 29 and the first paragraph of point 31 of the decision must not be taken into consideration since they were not sent to all the undertakings in the course of the administrative procedure (see above, p. II-897 ff.)

In the 30 July 1979 edition of European Chemical News it was reported in an article headed Montedison announces PP prices rise in Europe, that Montedison intended to raise prices throughout Europe by about 8% from 27 August. According to European Chemical News, that price rise brought prices to LIT 920/kg, DM 2.05/kg, FF 4.70/kg and BFR 33/kg for products sold to large users. Montedison's plastics division director general is then reponed as having hinted, like other major producers, that a second round of price rises might be on the cards before the end of the year. ICI's polyolefins director is reported as saying that ICI totally supports Montedison's move, and then it is added that many industry insiders reckon that ICI will announce similar increases shortly. Finally, it is reported that Shell was backing the Montedison move and that Shell's operating companies in Europe would be aiming to secure Montedison's target DM 2.05/kg. Shell reckoned that its polypropylene prices would rise towards the end of August or early September.

In a letter dated 29 August 1979 ATO-Chimie notified its prices applicable from 1 September 1979 for consignments of 20 t as FF 4.70 for fibre, FF 4.70 for moulage homopolymère, FF 5.00 for moulage copolymère and FF 4.90 for film homopolymère. The prices for other grades are also on the list. Further surcharges for various grades and for quantities of less than 20 t are given.

In almost identical circular letters of 24 and 26 July 1979 BASF gave its offices in Germany, the United Kingdom, Belgium, France and Italy notice of price rises to apply from 20 August 1979 in Germany and from 1 September 1979 in the other countries. In the letters the reason for the price increases was given as the rise in the price of raw materials in the third quarter and it was stated that holidays dictated the time for the price rises, although they were already very badly needed. The prices set to apply from the said dates were listed as follows: Grobtextil 2.05 DM/kg, Spritzguß 2.05 DM/kg, Folie 2.10 DM/kg in a grade called Novolen 11 and 10 pfennigs higher in a grade called Novolen 13. The prices given were described as minimum prices, whereas maximum prices were fixed as DM 2.45/kg and DM 2.55/kg for the two grades. In some of those letters containing price instructions it was also stated that the price rises were necessary even though the price rise announced at the end of May 1979 has not in any case fully penetrated. In the letters to the offices in other countries the prices for Raffia and Spritzguß are given as BFR 33/kg, FF 4.70/kg, LIT 920/kg and 500 or UKL 510/t.

In a circular letter of 20 July 1979, referring to the rising price of propylene, Hoechst warned of new prices for polypropylene on the basis of the following basic prices for a grade called Hostalen PP — Homopolymer natur/Granulat im Sack: Germany DM 2.06/kg, France FF 4.70/kg, Great Britain UKL 500/t, Italy LIT 925/kg and Belgium BFR 32.50/kg. According to the letter, the price rise was to apply for all deliveries taking place in September. The new price list was sent out with the circular letter of 1 August 1979 and was to apply to all deliveries taking place from 3 September. In the price list, which was supposed to be for internal use only, prices are given for deliveries of 20 t with minimum and maximum amounts. Six different grades are given at prices from DM 202 to 207/100 kg, whereas all other grades are given at the higher prices. There is in addition a column with slightly higher minimum prices for customers who take less than 20 t per delivery. According to the circular letter, the prices were indicated in that way in order to allow latitude for negotiation. Finally, it is stated that in some cases called special cases/large customers the head office should be contacted.

In a telex of 30 July 1979 from ICI Europe to ICI Germany, minimum target prices for September are given: raffia 2.05; ho moulding 2.05; co moulding 2.20; ho cast film; 2.10, the last figure being crossed out by hand and amended to 2.05 DM/kg. In notes to the list of prices it was mentioned that premiums and discounts remained as they were and that DM 1.90 was to be paid for oriented film, which was to become DM 2.05 from 1 September.

In a circular letter dated 30 July 1979 ICI announced a general price rise as from 1 September of UKL 25 per tonne. Thus homopolymer — film yarn was to go up from a floor price of UKL 475 per tonne to UKL 500 per tonne. Moulding and film were to rise respectively to UKL 510 and UKL 550 per tonne and copolymer — general purpose was to rise to UKL 550 per tonne. It was further stated that ICI expected that the price level on the continent would rise to DM 2.05/kg in September. In a later circular letter dated 28 September 1979 it was explained that the price level on the continent was still somewhat below the level of DM 2.05 which Montedison had announced at the end of August. Information is then given of the price levels in Germany, France and Belgium, stated as DM 1.70 to 1.75/kg, FF 4.00 — 4.10 and BFR 26.00 to 27.00 respectively. The letter went on to say that it was clear that 2.05 would not be achieved on the continent, but that ICI still thought that it was achievable in November/December and that it would actively support the attempts to achieve that price level. In consequence the instruction was given that UKL 500 per tonne was not to be asked from 1 October but that large customers could continue to buy at UKL 475 per tonne for another month.

In a telex of 20 June 1979 Chemie Linz informed its sales organization in Germany that on 19 June a decision had finally been taken to raise the price to DM 1.90 for raffia. It then pointed out that its competitors were trying to obtain a price increase for 1 July 1979, which would be reflected in information given to their local organizations. In addition it was mentioned that it was appreciated that a necessarily reduced flexibility in prices on the part of competitors would mean more contact from consumers. It finally explained that it would be necessary and practicable to aim at a further price increase of 15/pf. for 1 September. DM 2.05 is added in brackets and Raffia is inserted by hand.

In a letter of 30 July 1979 Shell announced that it intended to raise its prices on 1 September by UKL 25 per tonne for all grades. It instructed that letters were to be sent to all customers in the period from 30 July to 1 August. In the letter the prices on the continent were also discussed and it was mentioned that Montedison had issued a press statement to the effect that it would be raising its polypropylene prices in Europe, with effect from 27 August, to the prices which are also mentioned in the article in European Chemical News. It went on to state that in Europe some business was being transacted at UKL 425 per tonne, and therefore the extent of the price increase that needed to be achieved would obviously be very substantial. Shell considered that there might be some phasing up of price in September rather than a clean increase as of 1 September. Finally, Shell was said to be hopeful that this situation would not occur in the United Kingdom. Beneath the signature there is added by hand:

Finally, a cutting from European Chemical News dated 6 August 1979 and put forward in evidence by Atochem should be mentioned. Beneath the heading ICI joins latest PP initiative, it is reported that ICI, following the previous week's pricing initiative by Montedison, had announced its intention to lift polypropylene prices on the European market by 5 to 10% from 1 September. Homopolymer would thus rise to UKL 500 per tonne, compared to the 1 June level of UKL 480 per tonne.

If, as mentioned above on page 156 the minutes of the internal Shell meetings of 5 July 1979 and 12 September 1979 cannot be used as evidence against the undertakings, the basis for some of the Commission's conclusions in points 29 to 31 of the decision is considerably weakened. Thus the claim that the price initiative was to be applied from 1 July 1979 is to a certain extent supported by Chemie Linz's telex of 20 July 1979 alone. Nothing direct emerges on this point from the minutes of the meeting of 26 and 27 September.

On the other hand, an initiative from 1 September 1979 is, on the basis of the evidence briefly set out above, in my view well documented. In the first place, it appears from the minutes of the meeting of 26 and 27 September 1979 that a target price was set to be applied from an unspecified date before 26 September. Monte announced a price rise which was reported in European Chemical News on July 30 to take effect from 27 August. Hoechst, BASF and Chemie Linz had, however, shortly before that public announcement, given notice of exactly the same price rise to take effect at more or less the same time. Shell and ICI sent out their announcements on 30 July and it is of course conceivable that, as Shell states, those undertakings acted on the basis of a price announcement which Monte might have sent out a few days before the publication in European Chemical News. In any case, those undertakings certainly reacted extremely rapidly. Given the fact that (a) it is documented that meetings were held, (b) it is clear that at least three undertakings, for reasons difficult to explain, arrived at exactly the same view as Monte with regard to the price level to be applied from a particular date, and (c) it appears from the said note of the meeting that an effort was to be made to attain precisely that price level before 26 September, I consider that there need be no hesitation in assuming that the level and timing of the price rise must have been the object of an agreement or concertation.

(b) January to May 1981

For the meetings which, as the Commission contends and ICI and other applicants accept, took place in 1980, the Commission was unable to find any notes. The price initiative for 1981, which appears to have taken place in stages, is divided in the annex to the Commission's letter of 29 March 1985 into three phases, covering January 1981, February/March 1981 and May 1981.

The internal Solvay document mentioned in point 32 of the decision and the information contained therein cannot be taken into consideration since it was neither sent to the undertakings in the course of the administrative procedure nor otherwise adduced in evidence.

In the decision the Commission stated that a price initiative was mounted at the end of 1980 to set a target price of DM 1.50/kg. The only basis for that allegation is a number of price instructions from various undertakings, the aim of which was to push prices up to the level that the producers were reported by European Chemical News of 10 November 1980 as believing could be attained before the year end.

In a memorandum of 8 December 1980 DSM gave internal notice of guidelines for sale prices to be applied from 1 January 1981. For a grade called HEX figures of 1.30/1.50, 3.50/3.50, 320/325 are mentioned for West Germany, France and the United Kingdom respectively. There is a note to the tables saying Rock Bottom/Verkoop Richtlijn, which in its description of the document the Commission interprets as a minimum and a target price respectively.

On 22 October 1980 ICI gave notice to its subsidiary in Germany of target prices for the end of 1980. For November the prices were given as 1.30, 1.50 and 1.85 and for December 1.50, 1.70 and 2.00 for raffia, homopolymer and copolymeri general purpose respectively. In an internal circular dated 1 December 1980 ICI announced minimum prices applicable from 5 January 1981 as UKL 320 per tonne for raffia, UKL 380 per tonne for homopolymer; moulding and UKL 430 per tonne for copolymer — general purpose. In the letter it was stated that Shell was reported to have written to its customers to announce a price rise of UKL 30 per tonne as from 1 January 1981. In addition, instructions were given concerning the procedure to be followed in cases where it might be necessary to arrive at a compromise in relation to target prices. In telex notifications, the dates of which are illegible, to ICI in France and Italy the prices stated as desired minimum prices from the beginning of next year are 3.50, 4.10, 4.00 and 4.30 for January and 4.00, 4.40, 4.30 and 4.60 (FF) for February and 720, 760, 750 and 850 for January and 830, 860, 850 and 950 (LIT) for February.

In a Hoechst internal document originally dated 29 October 1980 Mindestpreise and Zielpreise/minimum are given, probably applicable for December 1980 and January 1981. In the document basic grades are given with DM 1.50/kg, DM 1.70/kg and DM 1.95/kg as target prices. The minimum prices are shown as 25-30 pf lower. The information in the document is crossed out and in handwriting there is the date of 5 January 1981; February 1981 is given as the date for implementation. Target prices for the basic grades are given as DM 1.75, 1.85 and 2.00/kg.

By a telex of 31 December 1980 Chemie Linz informed its sales office in Germany that the prices from January would be DM 1.50, 1.70 and 1.95/kg for basic grades and DM 1.75, 1.85 and 2.00/kg. from February.

It is stated in that letter inter alia that there were sure to be individual cases where temporary solutions would have to be found.

On 9 December 1980 Monte informed its Italian sales offices that from 31 December 1980 the price would be LIT 720, 750 and 870. It was stated that a watch would be kept on how the situation developed so as to ensure the best possible result with regard to the price/volume relationship.

On 17 December 1980 Shell issued an announcement that a number of its companies had set minimum prices applicable from 1 January 1981 of DM 1.50/kg, DM 1.70/kg and DM 2.00/kg for basic grades. It went on to state that notice was further given of a price rise of DM 0.25/kg from 1 February.

Finally, the instructions given to Saga in Great Britain contain similar figures.

As mentioned above on II-975, in its circular letter of 1 December 1980 ICI stated: It has become clear today via various messages from the marketplace that Shell.... ICI uses this as justification to its subsidiaries for itself seeking similar price rises: we will be supporting Shell's initiative .... On the other hand, no other evidence has been put before the Court in these cases to prove that a price leader was to announce a price level which the others merely followed. The somewhat vague and anonymous statements in European Chemical News thus plainly cannot explain why the price level of so many undertakings was set as uniformly as it was. Furthermore, the uniformity of the price announcements is particularly striking.

In the light of the foregoing, I consider that the description in point 32 of the decision is correct, but subject to the proviso that it is not apparent how the Commission supports its supposition that the original plan was to apply those levels from 1 December 1980.

The contents of the decision is corroborated in the very detailed report of the two meetings held in January 1981 and attended by a large number of undertakings (but not Petrofina, as the Commission acknowledged at the hearing). Here reference is made to the price levels in a number of countries. With regard to Germany it is stated that prices had moved up but not to agreed levels especially for raffia.

The report concludes that it was agreed that the DM 1.75 target should remain and that DM 2.00 should be introduced without exception from 1 March, as mentioned in point 33 of the decision.

The price instructions issued and referred to in the last paragraph of point 33 of the decision essentially correspond to that plan, except for ATO's instructions which are slightly at variance.

Therefore, in view of the above, I see no reason for doubting the essence of the Commission's findings of fact as far as the content of points 32 and 33 of the decision are concerned.

According to the evidence available, the final part of the initiative of January to May 1981 would appear to have been somewhat of a fiasco. Examination of the price instructions referred to by the Commission in point 34 of the decision shows that the undertakings named therein apparently endeavoured in vain to move prices up as stated. As I have explained above, however, I do not believe it is decisive in assessing whether an infringement occurred that the initiative more or less collapsed. In those circumstances and taking into account the fact that immediately beforehand a meeting had been held between the undertakings and that no evidence has been put forward providing an alternative, credible explanation, there can be no reasonable doubt that the prices must have been concerted or agreed.

(c) August to December 1981

The detailed examination set forth above of the documentary evidence concerning the first price initiatives has shown that essentially the Commission's assessment of the evidence had a firm foundation. It is thus mainly the leaving out of account of documents which — as explained in more detail in Section A, 5 (c) above — cannot be used on procedural grounds that alters the picture slightly.

Bearing in mind the length of this Opinion, I shall therefore confine myself to examining passages in the decision concerning later price initiatives which, in my view, call for particular remarks.

According to Table 3 of the Commission's decision, in the period relating to the price initiative referred to in the subheading, meetings were held on 28 July, 4 August, 28 August, 17 November, 20 November and 16 and 18 December 1981.

At the meeting on 15 June 1981, mentioned in point 35 of the decision, Mr C. R. Green of ICI, Mr J. E. Lane of Shell and Mr E. Zacchi of Monte discussed various possible solutions to the problem posed by the fact that the upward trend in prices was slowing down in the first four to five months of the year. One of the possibilities mentioned was a 20 pf/kg price increase with effect from 1 July.

As the Commission states in the decision at the beginning of this point, Shell and ICI had already, that is to say in the letters referred to above, given notice that there might be price increases in September/October.

According to the decision, within a few days of the meeting ICI and Shell had instructed their employees to prepare the market place for a price rise in September based on a raffia price of DM 2.30/kg. The direct connection thus suggested between the meeting of 15 June and ICI's and Shell's announcements, which were issued on 17 and 19 June respectively, does not withstand close examination. Shell does indeed talk of a minimum price of DM 2.00/kg in July but the only information corresponding to that is the information contained in a note of a meeting held on 27 May 1981 between Mr C. R. Green of ICI, and Mr J. E. Lane and Mr G. Dewick of Shell, in which a comparison of prices shows an average price of DM 1.8 for raffia and DM 2.1-2.2 for copolymer (annex 64/1 to the general statement of objections). However, that document is not relied on by the Commission in this connection. From more specific information in the two statements it appears merely that a price rise to DM 2.30/kg at the beginning of September was wanted and that has simply nothing to do with the possible price rises for 1 July 1981 discussed in the note of the meeting. No direct connection can therefore be found between the two documents.

The letter of 17 July 1981 from Solvay to its sales offices in Benelux which, as far as we know, has not been put forward in evidence cannot be used as evidence against the undertakings.

According to the information available to us, an experts' meeting was held on 28 July 1981 and we have a document which looks as if it might have some connection with that meeting consisting of a note written by one of Hercules' managers to Mr Bastiaens. In this note it is stated that the Official prices for August and September are DM 2.00 and DM 2.20 for raffia and DM 2.25 and 2.40 for moulding and DM 2.40 and DM 2.55 for copolymer.

In the second paragraph of point 35 of the decision, the Commission states that there was an original plan to go for a price of DM 2.30/kg in September 1981 and that that plan was probably revised at the meeting on 28 July. As will have become clear from the foregoing, the sole ground for supposing that there was an original, concerted plan consists of the ICI and Shell letters dated 17 and 19 June 1981 respectively and we have no information as to whether meetings were held in the period before 28 July, apart from the meeting on 15 June.

What we know therefore is that after the meeting on 15 June ICI and Shell both had plans to attain a price level of DM 2.30 by 1 September. The first more solid evidence which emerges is, however, the holding of the meeting of 28 July 1981 together with the note from Hercules of 29 July concerning the official prices for August and September.

For those reasons I am not satisfied that there is sufficient evidence of an original plan as mentioned at the beginning of the second paragraph of point 35 of the decision.

The price instructions referred to in the third paragraph of point 35 concerning prices for 1 October 1981 were issued as follows: on 7 August BASF issued a notice of prices to apply from 1 September and on 7 September a notice of prices to apply from 1 October. DSM issued price instructions in a letter of 6 August 1981. Hoechst sent out two letters, (probably) on 29 July and 13 August. ICI sent out letters on 4 August and 7 September 1981 concerning prices to apply from 1 September and 5 October respectively. Monte sent out a letter on 4 September 1981 for the months of September and October. On 28 August Shell issued a notice of prices to apply from 1 September and 1 October. In that letter, the alleged intended price increase mentioned in the first paragraph of point 36 of the decision is simply referred to as nov. 2.5 DM/kg.

On 23 December 1981 ICI issued a notice of prices to apply from January 1982. On 5 November 1981 Shell issued a notice of target prices for November and December. On 27 November prices were issued for January 1982.

Examination of those price instructions shows that the Commission is justified in describing them as nearly identical. Bearing in mind that it is known that meetings were held shortly before most of the price instructions were issued and also that there are otherwise well-documented price initiatives, I have no hesitation in assuming that a similar initiative took place with regard to September and probably for the rest of 1981.

(d) The periods from June to July 1982 and from September to November 1982

With regard to this period, the Commission alleges that there were two price initiatives. These are very well documented by notes of meetings and do not give rise to any particular comment.

(e) The end of the price cartel

The last price initiative mentioned in the decision is stated therein to have begun in July 1983. With regard to the period leading up to the last meeting, held on 29 September 1983 (possibly 30 September), before the Commission's investigations took place, the content of the decision does not call for comment. As with the period from June 1982 to November 1982, the evidence is so massive that there is no reason to go into it in detail in this Opinion. I should point out that two documents originating from Shell and mentioned in point 49 of the decision cannot be relied on as evidence for the existence of the cartel. The documents concerned are a document called PP W. Europe-Pricing and an internal report dated 14 June 1983. According to the Commission, all the applicants had the opportunity to acquaint themselves with these documents when they were given access to the files. However, as stated earlier, this is not sufficient in my opinion.

The fact that those two documents cannot be used does not, however, weaken the Commission's conclusion.

Finally, the question has been raised as to how the period subsequent to the last meeting, possibly after the Commission's investigation, is to be evaluated.

A number of applicants have argued that the price cartel must be regarded as having ended no later than after the last meeting was held. The Commission, however, takes the view that the undertakings continued to concert their practices even after 29 September 1983 on the basis of the previous agreements.

The question whether the cartel had any effect even after the Commission's investigation does not appear to be particularly important in this case since it is strictly a question of whether the cartel might have produced effects for a further few months. If this further period is compared with the duration of the cartel, it must be obvious that the question how far agreements entered into before 13 and 14 October 1983 were followed for any time after those dates is not of great significance for the determination of the amount of the fines.

The applicants' view of the legal situation as regards the last period of the cartel is however, incorrect. The unlawful activity cannot be regarded as having ceased when the Commission intervened and thereby possibly put an end to the organized meeting activity. It is the effects of the cartel which are decisive.

As far as this question is concerned, the Court of Justice stated in the Binon case (at paragraph 17, p. 2040): Moreover, Article 85 would also be applicable if parallel conduct on the part of publishers were continued after the termination of the former agreement and in the absence of its replacement by a new agreement.... The system of competition rules established by Article 85 et seq. of the EEC Treaty is concerned with the economic effects of agreements or of any comparable form of concerted practice or coordination rather than with their legal form. Similarly, in the EMI judgment (at paragraphs 30 to 32, p. 848), it is said that agreements are also covered by Article 85 if they continue to produce their effects after they have formally ceased to be in force. The situation in the two cases cited was somewhat different from that in the present cases. They concerned proper agreements which were formally terminated and not arrangements made within the framework of a cartel in which the agreements, by their the very nature, are characterized by the fact that the parties to the agreement knew perfectly well that their purpose was unlawful.

The same principle must, however, apply. If, therefore, it can be established that a similar pattern was followed in the period after the meetings ceased to take place, the obvious assumption is that the agreements or concertation previously entered into were being put into effect in the period after the Commission's investigation. As I see it, there is not much doubt that the price instructions for the period in question, that is to say October and November 1983, show that agreements or concertation must have existed. I therefore agree with the Commission that the cartel cannot be regarded as having terminated until a few months after the last meeting. This conclusion applies even though the document referred to in the third paragraph of point 51, which was found on ATO's premises, cannot be used against the undertakings since it was not sent to all of them.

9. The quota arrangements

As is clear from my remarks in section F, 7, above, concerning the situation in 1979, I do not think that there is a sufficient basis for concluding that there was a quota arrangement for all of 1979. The first quota arrangement to be dealt with in this section will therefore be the alleged quota arrangement for 1980.

As the Commission mentions under heading VII of the decision and as is shown by the documentary evidence, the undertakings operated with volume targets, target volumes, quotas and aspirations. These terms raise exactly the same difficulties of interpretation as the term target prices. As mentioned above in section E, 6, it must, however, be assumed that agreements or concertation fall within the prohibition in Article 85(1) even if their substance essentially takes account of actual market conditions. The same holds true for target quotas.

(a) The arrangement for 1980

In the note of the 26 September meeting, it is stated, as discussed earlier, that a fixed quota system is essential. That demonstrates that at that time the problem was being addressed and, as mentioned above in section F, 7, there are certain indications that a form of quota arrangement was in existence at the end of 1979.

At the other end of the timescale covered by the alleged quota arrangement we find the note of the experts' meeting held in January 1981. In this note provisional sales figures for December 1980 are set out and sales figures in the fourth quarter of 1980 are compared with those for the corresponding period in 1979. It can be seen that sales figures for each month of the fourth quarter of 1980 exceeded those for the corresponding months in 1979 and that total sales in 1980 were only 3% lower than those in 1979. Sales figures for each individual undertaking are then set out in a column. Alongside there is a second column in which is given the target figure for each undertaking, based, as stated in the note of the meeting, on a 1.21 million tonne market in Western Europe in 1980. In a comment on the figures it is said that of the four major producers only ICI was off target. DSM, it continues, disputed any undertaking to cut back from its original target; the Amoco figures were largely guesswork and the Hercules figure did not seem to fit in with views about its level of activity in 1980. It is explained that figures were not forthcoming from Solvay and BP but adding the rest together would suggest a combined total of 90 kilotonnes versus a target of 71 kilotonnes.

ICI's target figure is given as 139.2 kilotonnes and actual sales as 128.1 kilotonnes. With regard to ICI's situation it is said that part of ICI's problem had clearly stemmed from the severity of the recession in the United Kingdom but more important had been the virtual withdrawal from Dundee as a result of trying to lead prices up early in the year.

From the year 1980 there is a table (annex 60 to the general statement of objections) to which reference is made in the first and second paragraphs of point 55. On the typed table headed Polypropylene — Sales Target 1980 (kilotonnes) the date of 28 (but possibly 26) February 1980 is written in by hand. In the table there are four columns headed 1980 Target (Based on 1979; Petrofina Adjust.); Opening suggestions; Proposed adjustments and Agreed targets. In the latter column are given the figures for each undertaking, which added together produce max 1390 (kilotonnes). At the root of the document is typed 1390 kt would represent + 12.1% on 1979. Beside the figures for Saga, Tagsa/Paular, BASF and Petrofina there is an asterisk referring to a note in the document that they are to be rechecked. Against Petrofina, in the column Proposed adjustments is written max. 20, which reappears in the column Agreed targets. That tallies with the fact that, according to the information available to us, Petrofina did not enter the market until 1980.

The figures set out in the column Opening suggestions correspond almost without exception to the figures set out in the column Proposal I in the document previously referred to (annex 56) found on ICI's premises, but which may originate from one of the German producers. Furthermore, the same column in that document contains a series of figures in brackets which, compared with the figures in that document's Proposal II column, correspond to the first-mentioned document's Agreed targets column, as far as the undertakings for which adjustments were proposed are concerned. In the Proposal III column in annex 56 figures are given which correspond essentially to the Agreed targets in the first-mentioned document which, when totalled, amount to 1378 kilotonnes. There are discrepancies as regards Petrofina (15 instead of 20), BASF (60 instead of 64) and Saga (35 instead of 38), that is to say 12 kilotonnes in total, which relate to three of the four undertakings which, according to annex 60, were to be ťechecked.

The figures found in annex 60 in the column Agreed targets recur in annex 57, a document originating from ICI, dated 8 October 1980, in which a comparison is made between 1980 quotas and the various undertakings' theoretical capacity. They are again found in annex 58 in the column Sales 1980, Aspirations. In that document, which originates from ICI and is dated 9 October 1980, actual sales for the individual undertakings are also given, together totalling 1170 kilotonnes. Finally, in annex 59, which also originates from ICI, there are columns showing targets for 1980, totalling 1382 kilotonnes, compared with actual sales listed in two columns, one typewritten in round figures and one handwritten with exact figures, totalling 1207.9 kilotonnes. Those figures correspond, with minor discrepancies, to the figures mentioned in the note of the meeting of January 1981.

In considering all that information set out above it is to be noted first of all how striking it is that from three different sources, namely ATO, ICI and an unidentified German undertaking, there are detailed schemes specifying quotas in the form of an opening suggestion, a revised suggestion and a final or agreed proposal, which correspond to each other down to almost the smallest detail.

When the notes of the meetings of 26 and 27 September 1979 and January 1981 are also taken into account, confirming that discussions took place in the latter half of 1979 and corroborating the contents of the tables, I see no reason to doubt that the course of events recounted in point 55 of the decision as far as 1980 is concerned must be regarded as an essentially accurate and well-founded account. I find especially revealing the comment in the note of the meeting of January 1981 to the effect that DSM disputed any undertaking to cut back from its original target. DSM is quoted as having sold 46.1 kilotonnes, which should be compared with the fact that in the document of 28 February 1980 that undertaking is mentioned as having a quota of 45 kilotonnes.

(b) The period 1981-82

According to the Commission, there was no definitive quota agreement for this period (points 56 to 59 of the decision). In Article 1 (e) of the decision the alleged infringement by the undertakings is described as a sharing of the market in default of a definitive agreement covering the whole year by requiring individual producers to limit their sales in each month by reference to some previous period ....

A situation such as that described undoubtedly constitutes an infringement of Article 85(1) if the Commission is able to adduce evidence for its allegation.

1981

The first document cited by the Commission in this connection is the note of an experts' meeting in January 1981 referred to above. As mentioned in section 8(b), it is stated in the note that it was agreed that the DM 1.75 target should remain in February (1981) and that DM 2.00 should be introduced without exception in March. It is then stated: In the meantime monthly volume would be restricted to 1/12 of 85% of the 1980 target... with, it says, a freeze on customers.

From an undated internal note found on ICI's premises (mentioned in the third paragraph of point 56 of the decision) (annex 63) it appears that at a meeting various alternatives had been discussed as regards quota arrangements. It is not clear whether this was an internal ICI meeting or a meeting between producers. According to the note, ICI's attitude was that volume should be limited to the level that the market was expected to reach, namely 1.35 million tonnes. It is then stated that although there had been no further discussion with Shell, the four major producers could set the lead by accepting a reduction in their 1980 target market share of about 0.35% provided the more ambitious smaller producers such as Solvay, Saga, DSM, Chemie Linz, Anic/SIR also tempered their demands. Finally, it is stated that, provided the four major producers were in agreement, the anomalies could probably be best handled by individual discussions at senior level, if possible before the meetings in Zurich. There then follows a table setting out a possible compromise with the various undertakings' shares expressed as a percentage of their target volumes for 1980.

There is also annex 61 to the general part of the statement of objections, discussed in the section devoted to 1980, in the table of which there is also a column headed 1981 Aspirations. In addition there is a plan, in annex 62, setting out three different proposals on which cooperation is sought compared with the undertakings' aspirations.

From the note mentioned in the decision (annex 64) of a meeting on 27 May between Shell and ICI it is clear that no quota agreement had been entered into at that time. From the note of the meeting dated 17 June 1981 also referred to it appears that no quota arrangement had been established at that point either.

In a table found on ICI's premises, but apparently originating from Monte (annex 65) a comparison is made for the period January to December 1981 between what is called actual and theoretical sales. A final table in this connection (annex 67), dated 21 December 1981, shows all the producers' monthly sales for 1981. Up to and including October the figures are typewritten, but for November ana December they are written in by hand. A further column contains a handwritten comparison for the whole year.

The final piece of documentary evidence on which the Commission relies is a note prepared by an employee of ICI, Mr M. E. Robinson, in the middle of 1981 when he took over the post of marketing director for thermoplastics (annex 66). The relevant part of the note is quoted by the Commission as follows: Do not have 81 agreement but people use 80 as a model.

As stated, the Commission believes that reciprocal undertakings were given by the firms to limit production. In the second paragraph of point 57 the Commission refers to a stopgap measure which consisted in sales being monitored against a notional split of the available market based on the 1980 quota.

If the first months of the year are left out of account, it appears fairly clear from the documentary evidence that the producers could not agree on a quota scheme. To support the Commission's assumption that there must have been a stopgap measure of such a nature as to be covered by Article 85(1) there is, however, only the statement that people used 1980 quotas. The fact that at some point which might have been at the end of 1981 or the beginning of 1982 a Monte or ICI employee made a comparison between a theoretical and actual basis would seem to be a rather flimsy foundation for the far-reaching conclusions drawn by the Commission. It is of course suspicious that an ICI employee was apparently in possession of exact sales figures for each individual undertaking and there are thus good grounds for assuming that information on sales was exchanged at meetings. In my view, however, there is not any sufficient correlation between the evidence referred to in order to justify a finding that the undertakings required each other to limit their sales.

For the reasons set out above I consider that there is sufficient evidence to prove that in the first months of the year a quota arrangement was agreed in order to support the price initiative in February and March in accordance with the contents of the note of the January meeting. However, apart from that evidence, there are not, either in the tables available or in the lapidary comment in the note cited, sufficent grounds for the conclusions the Commission draws from evidence. I would therefore hesitate to endorse the Commission's view on this point.

1982

The Commission believes that the undertakings similarly required each other to limit sales for 1982 without otherwise managing to reach a consensus on a quota scheme. Annexes 69 to 71 of the general part of the statement of objections show the proposals put forward, but the note of the meeting of 10 March 1982 mentioned in point 58 of the decision cannot be used against the undertakings because, according to what we know, it was only communicated to ICI. According to the defence in some of the cases, the document was communicated solely to ICI, which does not claim that the document was not communicated in the proper way. However, the document does not seem to have been put forward as evidence in the case at all. The Commission admitted that this was an error and referred instead to a note of the same meeting originating from Hercules, that note having been annexed to the general part of the statement of objections as annex 23. As regards the proposals for volume control, that note contains nothing significant, however, other than a list of the various undertakings' theoretical production capacity. In my view, there thus is no sufficient basis for assuming that in the first half of 1982 the undertakings undertook, on the matter of volume control, anything other than the exchange of information concerning actual sales and more general information relating to the individual undertakings' situation.

Point 59 of the decision contains an account of ICI's initiative upon the taking over by the undertaking of the chairmanship of the group. As regards quota arrangements, it is clear that ICI wished to have, and considered essential, an effectively functioning quota system which was to take effect from the beginning of 1983. Meanwhile, it was emphasized (from the note of the meeting of August 1982 it appears that it was ICI which emphasized this to the others) that the undertakings had to endeavour to limit their sales to the market share which they had had in the first months of the year and (also according to the note of the August meetings) in relation to a market of 120 kt.

On the matter of agreements or arrangements, the information mentioned in the second and third paragraphs of point 59 originating from ATO contains only one remark, to the effect that at the end of 1982 quasi-consensus had been reached concerning aspirations and market shares.

In my view, that evidence does not demonstrate with sufficient certainty that a consensus had been reached to limit sales in 1982. The only indication we have on this point is a proposal from ICI and there is no precise evidence to indicate how this was received by the others. It thus appears that efforts were concentrated on seeking to achieve a workable system for 1983.

In those circumstances I cannot agree with the Commission that there is evidence proving that the undertakings attained any form of consensus to limit sales in 1982.

(c) Quota arrangements in 1983

For the first two quarters of 1983 the Commission considers it has established that there was a consensus between the producers on a quota system.

It is indeed clear from the documents put forward in evidence that upon taking over the chairmanship of the group in the middle of 1982 ICI attached great weight to the introduction of a quota arrangement. The evidence shows, moreover, that ICI's managers made great efforts to set up such an arrangement.

Nevertheless, it is difficult to find any documentation showing that consensus had been reached on a quota agreement as alleged by the Commission, particularly if documents which cannot be used against the undertakings on procedural grounds (that is to say, ICI's briefing note for a meeting with Shell in May 1983 — second paragraph of point 63 of the decision — and Shell's plan for the first quarter of 1983 — third paragraph of point 63 of the decision) must be ignored. The only item in the documentary evidence which shows anything like consensus is the internal Shell document mentioned in point 64 of the decision (annex 90) in which there is a reference to the agreed Shell target. However, there is nothing else concerning the second quarter of 1983. In my view, that material is too slender for regarding it as proven that there was an agreement for the second quarter.

With regard to the first quarter, the Commission states in the first paragraph of point 63 of the decision that a number of undertakings found acceptable an allocation of quotas which had been discussed at a meeting on 2 December 1982. Nothing is said, however, about how the others reacted; furthermore, it is not clear how those who agreed would react if the others demanded larger shares. In my view, nothing in the material otherwise available provides a basis for concluding that a consensus was reached for the first quarter.

Thus, in my view, there are no sufficient grounds for assuming that there was an agreement on a quota arrangement for any part of 1983.

10. Account management and account leadership

In Article 1(c) of the decision, the account management system is expressly mentioned as one of the measures taken by the undertakings with a view to making possible the price rises they sought.

The purpose of the account management or account leadership systems, which are described in detail in point 27 of the decision, was to prevent customer tourism, which is discussed in several places in the documents produced in evidence. As ICI states, attempts were made to implement probably two systems. There is not much information available concerning the first, but it appears that preparations at least had been made to implement an arrangement and detailed plans were laid for a number of countries. This took place at a meeting on 2 September 1982. At a meeting on 2 December 1982 a suggestion was made to implement the idea at a more general level and a long list of customers was selected to make up the system. From a meeting of 3 May 1983 there is a note of very thorough discussions of each individual customer's situation.

In my view, on the basis of that evidence there cannot be any reasonable doubt that the Commission is right in believing that one or more management systems were used in order to try to halt or neutralize customer tourism. It is not hard to imagine that a system such as that described collapsed after a time. It is certainly no easy matter to make such a system work in a market with 20 sellers and a very large number of purchasers. The notes of meetings in 1983 show, however, that it did function for some months.

11. Other ancillary measures

Most of the other measures referred to in point 27 of the decision and intended to facilitate the implementation of price increases may be regarded as an integral part of the main infringements, namely price and volume agreements. Therefore, it is hardly necessary to deal with these matters separately, being as they are details in a much larger picture. There is, however, on the evidence available, no reasonable doubt that the Commission's assumptions about what took place in this respect are right.

There is one exception, however. It concerns the Commission's allegation that the undertakings agreed to divert their supplies to deep sea markets in order to create a shortage in Western Europe conducive to a price increase. In the light of the evidence available I find it doubtful that the undertakings did anything other than inform each other of their sales to deep sea markets, called rest of the world in the documentary evidence. It is quite normal for undertakings to attempt to dispose of surplus production in markets other than the nearest. However, very firm evidence is required before it can be found that the undertakings clearly intended to create a shortage on the West European market. In my view, there is no such evidence to hand and the proper course for the Court is to ignore the Commission's allegation that there was an agreement to divert supplies from Western Europe.

G — Limitation

Pursuant to Article 1(1 )(b) of Regulation No 2988/74, the power of the Commission to impose fines or sanctions for infringements of the rules of the European Economic Community relating to transport or competition is subject to a limitation period of five years for infringements such as those involved in the present cases.

Article 1(2) provides that time is to begin to run upon the day on which the infringement is committed. However, in the case of continuing or repeated infringements, time begins to run on the day on which the infringement ceases.

According to the first sentence of Article 2(1), the limitation period is interrupted by any action taken by the Commission, or by any Member State, acting at the request of the Commission, for the purpose of the investigation or proceedings in respect of an infringement. The limitation period is interrupted with effect from the date on which the action is notified to at least one undertaking which has participated in the infringement.

In the present cases a number of undertakings claimed even during the administrative procedure that the floor-price agreement of 1977 operated separately from the subsequent arrangements and that therefore liability to a fine must be precluded by limitation as far as that part of the case was concerned. In point 103 of the decision the Commission rejected that argument, referring to the clear factual and circumstantial lien between the arrangements.

Some of the applicants invoked the limitation rules in the proceedings before the Court. Given the nature of the limitation rules, it is hardly necessary for an applicant to refer expressly to Regulation No 2988/74. The fact that the undertaking claimed that a fine should not be imposed for a particular aspect of the case must be sufficient.

The Commission regards what happened quite simply as a constant or continuous infringement and in the decision also alleges that there is such an interrelationship between the floor-price agreement of 1977 and what subsequently took place that the infringement which began in 1977 can be regarded as having ended only some time after the Commission's investigations at the undertakings' premises on 13 and 14 October 1983.

The applicants maintain their view that the arrangements which may have existed in 1977 were essentially different and completely separate from what later took place.

It is undisputed that any limitation period which might have begun to run was interrupted on 13 October 1983 when the Commission began its investigations and the question is therefore whether any infringement or part of an infringement can be considered to have been committed or to have ceased before 13 October 1978.

If the Court agrees with my assessment of the evidence as set forth above, I have no doubt that it must rule that the case concerning the floor-price agreement of 1977 is time-barred. In my judgment, the floor-price agreement was at the most a sort of precursor of what took place later. It may well have had the same purpose as the subsequent agreements, but from what we know it was not the result of negotiations within the same framework. According to my findings based on the evidence, there is moreover a clear time interval between the floor-price agreement and the first result of the applicants' series of meetings that we can establish with sufficient certainty, namely the price initiative in the latter half of 1979. Of course, the beginning of the series of meetings which formed the framework for the later infringements bears some temporal connection with the floor-price agreement, but no concrete unlawful results of the series of meetings prior to 1979 have been demonstrated.

In my view, it should therefore be held that the infringements concerning the floor-price agreement came to an end in Spring 1978 at the latest and that liability to a fine for those infringements has therefore lapsed. This means that the fines should be reduced to some extent as will be explained in Part HI devoted to the question of the fines.

II — The individual cases

In this part I shall consider, on the one hand, the evidence regarding each individual applicant and, on the other, individual submissions made only by one or a few applicants. I shall not comment on every detail of all the applicants' arguments, which would hardly be possible if my Opinion is — despite everything — to be kept within reasonable bounds. This applies, in particular, to the submissions and arguments of an economic nature, to which, as will have become apparent from the foregoing, only limited weight can be attached in determining whether Article 85(1) has been infringed. The same applies with respect to the evidence, since detailed consideration of every item of evidence may be regarded as somewhat superfluous in cases where the incriminating evidence is clear and unambiguous. I would recall that in Section I, F. 5 above I have expressed my views on the question of the participation of the smaller undertakings in the floor-price agreement as from 1977, and so in principle I shall not comment on that aspect in this section.

A — Rhône-Poulenc (Case T-1/89)

It is established that Rhône-Poulenc disposed of its polypropylene business at the end of 1980. It is therefore one of the undertakings in respect of which there is only relatively sparse direct evidence. The period to be taken into consideration runs from Autumn 1979, when the cartel — as stated above in Section I, F. 7 — can on the basis of the evidence be assumed to have begun its activities, to the end of 1980. The fact that the direct evidence is not comprehensive has made the parties' arguments in the case quite specific in nature and ipso facto means that one of the fundamental problems of proof in the case, namely the evidential significance of the very fact of taking part in the meetings, must be squarely addressed.

In my view, the evidence in the Rhône-Poulenc case is weaker than in any of the other cases. The only documentary evidence is the report on the meeting held on 26 and 27 September 1979 together with the tables in which quota arrangements are mentioned, including the note found at ATO regarding the producers operating in France. Further evidence is ICI's information that Rhône-Poulenc took part in the cartel's meetings from at least 1979. Finally, the case is founded on a series of inferences drawn from the other evidential elements. The case therefore prompts reflection on the requirements regarding the strength of the evidence which the Commission must possess as a basis for a decision. In this connection, considerable importance must be attached to the fact that competition cases of this kind are in reality of a penal nature, which naturally suggests that a high standard of proof is required. On the other hand, the importance and value of circumstantial evidence should not be underrated.

Rhône-Poulenc claims that the Commission is not in possession of substantial evidence against it. It points out in particular that the bulk of the existing documentary evidence concerns a period after Rhône-Poulenc left the market which, in the applicant's view, must mean that that evidence cannot be relied on against that undertaking: the Commission's handling of the evidence was a sort of retroactive conjecture and is not therefore sound; in arriving at its decision the Commission, amongst other things, took no account of changes on the market since the situation up until 1982 was characterized by imbalance and low profitability so that the assessment of the undertakings' conduct should not be the same before and after the end of 1980; furthermore, the Commission failed to take account of the cartel's evolution.

With regard to documentary evidence concerning the period when Rhône-Poulenc was present on the market, the applicant states first that the report on the meeting held on 26 and 27 September 1979, which is merely an anonymous handwritten note, does not name any of the participants. As regards the tables concerning quotas, it claims that no significant indication can be drawn from those tables of any agreement in which Rhône-Poulenc took part. In that connection Rhône-Poulenc points out first of all that, although AMOCO, Taqsa, Paular and BP are named in those tables, the Commission expressly or implicitly recognized that those undertaking did not take part in any cartel. Secondly, Rhône-Poulenc claims that it is not possible to do as the Commission did and classify as a quota something which is in fact only a sales target, which, moreover, explains why those figures were periodically altered to take account of actual sales in the course of the year. With regard to the Commission's assumption that Rnône-Poulenc took part in the bosses meetings' and experts meetings' from 1979, Rhône-Poulenc points out that the evidence of such participation consists only of the information from ICI. According to Rhône-Poulenc, that information is cast in vague and general terms and is not in itself sufficient to prove Rhône-Poulenc's participation in the meetings or in the alleged cartel. Finally, it points out that no other evidence whatsoever has been found to show its engaging in conduct on the market such as price instructions or the like.

The Commission for its part first refers to its main premiss that the matters discussed at the meetings of which no reports are available must have been the same as those discussed at the meetings of whose subject-matter we do have knowledge. The Commission thus bases its view precisely on the retroactive presumption which is challenged by Rhône-Poulenc. In support of its view the Commission refers in particular to ICI's information that the idea of target prices developed in 1978 and that the cartel may in any event be assumed to have been fully operational from 1979. With regard to the tables found, the Commission observes that they do not appear to contain precise figures for AMOCO and Hercules and that it appears from the report and from a later meeting that the figures for AMOCO were described as guesswork. The Commission further states, as regards the tables concerning 1980, that the quota which, according to the documents, was allocated to Rhône-Poulenc closely corresponds to its actual sales figures (2.9% as against the 2.98% of a market which was originally estimated at 1382 kt but proved to be 1207.9 kt). The Commission sees this as conduct which shows that Rhône-Poulenc took seriously the agreements entered into. The Commission considers that Rhône-Poulenc's participation in the meetings is C roved by the information given by ICI and y the fact that Rhône-Poulenc has never denied taking part in the meetings.

As already mentioned, Rhône-Poulenc's case in particular prompts consideration of the question how far mere participation in meetings, in conjunction with a few additional indications, can be regarded as sufficient proof of participation in the cartel.

As far as its attendance is concerned, according to ICI's information, Rhône-Poulenc attended regularly, at least from 1979. Neither Rhône-Poulenc nor anyone else has been able to give any explanation why ICI should come forward with such information if it did not believe that it could vouch for it. There seems to be no logical reason why ICI should seek to implicate its competitors in this way and it may be assumed instead that ICI has been careful in answering the Commission's questions. As regards the reliability of ICI's information, it is further worth noting that most of the other applicants who, according to ICI, took part in the meetings have in fact acknowledged their participation as indicated by ICI. On the other hand, there is only Rhône-Poulenc's assertion that ICI's information does not prove Rhône-Poulenc's participation. However, Rhône-Poulenc has not denied its participation. Even if Rhône-Poulenc, as appears from the reply to the Statement of Objections, no longer has any employees who were concerned with polypropylene production at the time, the least that could be expected was that Rhône-Poulenc itself would have examined the question thoroughly enough to venture to deny having taken part in the meetings. In that case, the Commission would have had the possibility of carrying out further investigations in order to either confirm or disprove that denial. In my opinion, it is not enough merely to adopt a passive attitude in the face of information giving a credible indication of participation. For that reason I consider it justifiable to assume that Rhône-Poulenc did take part in meetings from 1979 until the end of 1980.

In so far as Rhône-Poulenc is concerned, the only evidence of its involvement in the cartel, apart from the references in the quota tables, is its participation in the meetings. As I stated in the introduction to my Opinion and as is accepted in American case-law, participation in a single meeting at which others agree to something unlawful is generally not sufficient to prove involvement in the unlawful acts. Usually it will be possible for a party to provide a reasonable explanation to the effect that it attended without any intention to take part in an infringement of the competition rules. But to explain in this way participation in more than one, or possibly quite a few meetings, without dispelling the suspicion of complicity in the unlawful acts of the other participants in the meetings will, in my view, be a difficult matter. If, on top of this, no facts come to light to provide a reasonable alternative explanation of the purpose of the attendance at the meetings and if the name of Rhône-Poulenc — unlike those of AMOCO and BP — appears beside precise figures, it is in my view justifiable to presume that Rhône-Poulenc did take part in the cartel from its beginnings in autumn 1979 until Rhône-Poulenc left the market at the end of 1980.

With particular regard to the price initiatives referred to by the Commission, of which only the initiative from July to December 1979 took place while Rhône-Poulenc was present on the market, Rhône-Poulenc claims that its involvement is not documented in any way and that there is no evidence of it. Rhône-Poulenc's conduct is in fact completely unknown.

The Commission points out that having taken part in secret meetings at which target prices were agreed Rhône-Poulenc cannot escape responsibility for involvement in the cartel merely because no written price instructions have been found in its case. The Commission further points out that the information to hand concerning Rhône-Poulenc's compliance with the quota system suggests that its participation in the meetings was not entirely gratuitous and that its conduct was in accordance with what had been agreed. Furthermore, in the Commission's view, the quota arrangements and the price arrangements cannot be dissociated from each other.

The first observation which must be made in response to Rhône-Poulenc's arguments is that the fact that conduct is not known does not of course mean that there cannot have been any conduct in accordance with the agreements made. As the Commission has pointed out, there is furthermore considerable evidence suggesting that Rhône-Poulenc did in fact comply with the quotas fixed for it for the year 1980. In those circumstances and having regard to the fact that Rhône-Poulenc did take part in the meetings, there are in my view no substantial reasons for doubting that Rhône-Poulenc made use of the knowledge which it obtained from the meetings.

With regard to the degree of Rhône-Poulenc's involvement, there is no information to suggest that Rhône-Poulenc was particularly active or particularly restrained in its conduct.

One argument connected to a certain extent with the matter of evidence is Rhône-Poulenc's assertion that the Commission disregarded the principle of equal treatment in so far as it treated Rhône-Poulenc as a participant in the cartel while not prosecuting BP and AMOCO even though, in Rhône-Poulenc's view, the Commission had better evidence against those two undertakings than against Rhône-Poulenc.

The Commission states that the decisive factor in its assessment of the evidence was that the involvement of those two undertakings in the meetings of the cartel could not be proved and that the Commission therefore decided, despite holding certain suspicions, to refrain from taking further action against those two undertakings.

I do not believe that Rhône-Poulenc is right in the view it takes. Irrespective of whether the Commission could have been obliged, by a complaint from an entitled party to attempt to prosecute BP and AMOCO — which in itself is very much open to doubt — the failure to prosecute those two undertakings cannot in normal circumstances give rise to any automatic consequences for others. In this instance, owing to the evidence available in respect of the various parties, the case of AMOCO and BP on the one hand and that of Rhône-Poulenc on the other are not the same. Furthermore, it has not even been claimed that the Commission's decision to prosecute some undertakings and not to prosecute others rests on anything other than objective reasons. Where nonobjective reasons can be ruled out, the Commission must in my opinion be allowed a wide margin of discretion in deciding which cases it can pursue and which it cannot. In the polypropylene cases the Commission drew the dividing line where it identified the existence of evidence of participation in the meetings and on that basis it did not believe that the cases against BP and AMOCO would stand up. Such an assessment of the evidence must clearly lie within the Commission's margin of discretion even if others might consider that the cases could have stood up. Moreover, in my opinion, the Commission's assessment appears reasonable. On those grounds I therefore consider that Rhône-Poulenc's submission in this respect must be rejected.

Finally, Rhône-Poulenc has claimed that the statement of reasons for the decision is inadequate in its regard in so far as the Commission took no account of the fact that the cartel developed in the course of the period during which it existed and thus wrongfully called Rhône-Poulenc to account for actions undertaken only after it had withdrawn from the market.

It is difficult to see the justification for that point of view. The decision must of course be understood as charging the undertakings with having taken part in what happened while they were involved in the cartel and the information in the cases provides no grounds for believing that the Commission intended to do otherwise.

B — Petrofina (Case T-2/89)

It is plain from what has been said above that I agree with the Commission that the very fact of taking part in meetings at which discussions took place which indubitably were concerned with arrangements restricting competition is weighty evidence against an undertaking. In so far as more regular participation cannot be established, other solid evidence is necessary in order to find that the undertaking took part in the cartel. With Petrofina as with Rhône-Poulenc the question of participation in meetings gives rise to certain problems.

As is mentioned inter alia in the eighth paragraph of point 78 of the decision, from the beginning of 1980 until March 1982 Petrofina collaborated with Montedipe in operating the production plant at Feluy in Belgium. Their collaboration was implemented by the jointly owned company Montefina. In March 1982 Petrofina itself took over the sale of its share of production through its own sales department for chemical products.

In its pleadings, Petrofina acknowledged that it took part in meetings from May 1982 and until the end of the series of meetings in September 1983. With regard to the period from March to May 1982 there remains some uncertainty as to what actually is Petrofina's position. In its reply of 12 January 1984 to the Commission's request for information, Petrofina states that its employees took part in meetings from March 1982. In an annex to its reply, Petrofina then lists the meetings between May 1982 and September 1983 in which it took part, beginning with the meeting on 18 May 1982, and also states which employees represented Petrofina there. On that basis it might be assumed that Petrofina took part in meetings from March 1982 but was not in a position to specify who took part in them. On the other hand, Petrofina expressly denies that it was represented at the meeting which, according to the second paragraph of point 58 of the decision, was held on 10 March 1982. There are no documents in the case to refute that denial.

As regards the period prior to March 1982 the situation is rather unclear. In its reply of 12 January 1984, Petrofina states that it did not have the power, on the basis of its collaboration with Montedipe, unilaterally to commit Montefina and that it therefore could not answer for that undertaking. With regard to one or two specific meetings, namely the meetings in January 1981 of which reports are available (see the third paragraph of point 33 of the decision), Petrofina stated in answer to a question from the Court that, in the light of a fresh investigation, Petrofina did not seem to have taken part in those meetings.

In its answer to a similar question from the Court, the Commission stated that the indication in the third paragraph of point 33 of the decision that Petrofina took part in those meetings is based on an error. Regarding the possibility of Petrofina's participation during this period, the Commission further states in the eighth paragraph of point 78 of the decision that it is not certain whether before March 1982 Petrofina was separately represented at the meetings.

In view of the fact that the Commission does not directly seek to claim that Petrofina was represented at the meetings in January 1981 or at the other meetings up until March 1982 and that Petrofina's statements are best construed as a denial, the Court should, in my opinion, proceed on the basis that Petrofina employees did not take pan in those meetings. The least that could have been expected of the Commission is that, when Petrofina stated that it could not answer for Montefina, the Commission should have examined what lay behind that answer.

As regards the meetings in the period from March to May 1982 I consider that, in the light of what Petrofina itself acknowledged in its answer to the Statement of Objections, it may be held that Petrofina did take part in the meetings from March 1982 (except for the meeting on 10 March 1982).

However, the Commission believes that Petrofina was also involved in the various cartel arrangements in the period from the beginning of 1980 until March 1982 and the eighth paragraph of point 78 of decision implies that Petrofina may have been indirectly represented at the meetings through its collaboration with Montedipe concerning the plant at Feluy.

The Commission's basis for regarding Petrofina as having taken part in the cartel or as sharing responsibility for what happened in the period from 1980 until March 1982 stems from two quite different points of view. As is apparent from the third paragraph of point 102 of the decision, the Commission claims principally that Petrofina took part in the cartel in its own right from 1980. In the alternative, however, it considers that Petrofina must share responsibility for Montefina's participation in the cartel up to March 1982.

In the individual Statement of Objections addressed to Petrofina it is stated in this regard that the two parent companies, Petrofina and Montedipe, must be liable to the extent that infringements of the Community competition rules were committed (before March 1982) by Montefina. A similar passage is also contained in the Statement of Objections addressed to Montedipe.

However, there is not sufficiently complete information to show whether Montefina may have been independently represented at the meetings and in that way or in other ways may have incurred the responsibility of the parent companies. Nor is there any information as to whether Montefina may have directly represented the parent companies, notwithstanding the fact, as mentioned above, that Petrofina's first answer should have prompted the Commission to ask again. In those circumstances, the Commission's alternative point of view may be disregarded altogether. A third possibility was raised by the Court at the hearing in the form of a question to the Commission, which was asked whether it considered that Montedipe had also represented Petrofina through its collaboration in Montefina. The Court did not really obtain an answer to that question.

It must therefore be concluded that there remains considerable uncertainty as to how — if at all — Petrofina was represented at the meetings, or what connection Petrofina otherwise had with the meetings.

The other evidence on which the Commission founds its assumption that Petrofina was involved in the cartel as from 1980 is summarized in the Commission's answer to the Court's written question number 33. The Commission there states that it is principally relying on Annex 60 to the general Statement of Objections, which refers inter alia to 1980 Targets (Based on 1979 + Petrofina Adjust.). That shows, according to the Commission, that Petrofina took part as from 1980 and that it must therefore also have taken part in the price initiatives taken after the beginning of 1980. The Commission also bases its view on the fact that during the period under consideration Petrofina was allocated its own separate quotas, which were thus not included in Montedipe's quotas.

Since, as described above, Petrofina's involvement in meetings remains altogether unclear, the evidence relied on by the Commission is, in my view, so sparse that I believe that the Court should conclude that Petrofina cannot be held to have taken part in the cartel before March 1982. Since Petrofina collaborated with Montedipe in the period up to that date, it is indeed not inconceivable that Petrofina was implicated in one way or another, but there is not the slightest evidence to prove it.

As regards the remainder of the period, namely from March 1982 to November 1983, Petrofina states that it did indeed take part in the meetings but that its participation was purely passive and its purpose was only to collect information relevant to Petrofina's competitive prospects.

Only two written price instructions are available from Petrofina, which explains that price instructions were given orally to its salesmen.

The fact that price instructions were not given in writing, however, does not by any means signify that instructions were not given. Details about the content of oral price instructions simply form no part of the evidence in the case. They may have accorded exactly with the written instructions sent out by the other undertakings and they may have been quite different. Given the means of obtaining information at the disposal of the Commission under the relevant provisions, it hardly has much chance of procuring solid information about the oral instructions. But Petrofina for its part has done nothing to substantiate its assertion that it acted completely independently of any agreements or concertation entered into by the other undertakings. The fact that an undertaking, as Petrofina maintains, sold at prices below the target prices does not, as pointed out above, serve as evidence to refute the assertion of its participation in an agreement or concerted practice. That would require the undertaking to come forward with information which showed that it had not even tried to use an agreed or concerted target price as its own target or otherwise allowed its conduct on the market to be influenced by the agreements or the concertation.

The absence of written price instructions can thus not in itself be regarded as an indication that an undertaking has not taken part in the cartel.

In the present case there are, as pointed out above, two price instructions. The first is dated 11 March 1982, that is to say the day after a meeting in which, as I have said, Petrofina must be presumed not to have taken part. Petrofina considers that it was legitimately entitled to expect that the Commission would not use the document against it since that document was neither included amongst the documents that were sent together with the Commission's letter of 29 March 1985 nor mentioned in the decision or the annexes thereto. I would simply observe in this respect that the annex was sent with the individual part of the Statement of Objections in which the annex was expressly relied on against the applicant and that there is nothing else in the case to show that the Commission may have refrained from relying on that annex. The explanation why the annex was not sent with the letter of 29 March 1985 is, of course, that the Commission did not expressly allege a price initiative on 1 April 1982 as evidence for the cartel. But it is now established that the price instruction from Petrofina corresponds to what appears from the report on the meeting of 10 March 1982. If an undertaking has generally been involved in meetings, the fact that, perhaps for fortuitous practical reasons, it was absent from one single meeting is not, as the Commission points out in point 83 of the decision, decisive. It was in the interests of all parties to ensure that the absentee was informed of what had happened at the meeting. Whether a document which concerns a circumstance not expressly mentioned in the decision should be used at all against the undertaking is indeed doubtful, but the fact that the document does exist certainly does nothing to contradict the Commission's point of view. As to the contents of the document, there is nothing much to add to my general comments regarding price initiatives. It sets out a list price together with a certain margin for the sales department's negotiations. That goods are not sold at list prices is quite common, but it certainly does not mean that that was not the price aimed at.

The second written price instruction was given on 20 July 1983. It specifies that with immediate effect prices are to be charged which seem to correspond to what was agreed at a meeting on 1 June 1983. In the telex message reference is made to technical problems which had caused production to be halted. Petrofina states that the telex had no connection with any agreements concluded at a meeting held on 1 June. Its object, according to Petrofina, was merely to curb sales, which was then done by setting prices at a level above the prevailing market price. That statement is not contested by the Commission, which, however, emphasizes that, as regards the amount, the price instruction corresponds in any case with that which was agreed to take effect from 1 July 1983 at the meeting on 1 June 1983 in which Petrofina took part.

On the basis of the information available, it must be assumed that Petrofina's written price instruction was sent for the reason given by that undertaking. However, it is established that the price instruction directed the price agreed at the meeting on 1 June to be applied. Even if the document can, therefore, hardly be taken as showing that Petrofina only gave information regarding the price mentioned because it was agreed at the meeting, the price instruction in question also does not contradict the Commission's view. It is therefore scarcely possible to attach any significance to the document at all.

But even if it is not possible to lay too much weight on the written price instructions themselves, I consider that Petrofina's participation is beyond doubt in the light of the other evidence. Petrofina is one of the undertakings which explain their presence at the meetings by saying that they were obliged to take part so as to gain the necessary information in order, as new undertakings, to cope with the competition. Quite apart from the fact that the question of liability for merely passively receiving information might arise in such a situation, the explanation given is quite simply not plausible. It is so improbable that a completely passive role should have left no trace whatsoever in the written evidence that the explanation can be rejected.

Adopting the metaphor used in the discussion at the hearing about various undertakings' claims to have had a purely passive role, I must say that it is just not likely that a group of business managers from various EEC countries should repeatedly sit around a table and simply listen, like a bevy of birdwatchers, for sounds which do not come.

The evidence also rules out such an assumption in Petrofina's case. The documents contain details regarding Petrofina's participation in the adoption of price initiatives, volume targets and account management.

In my opinion, it should therefore be held that Petrofina did take part in the cartel, but only from March 1982 until it came to an end around November 1983.

C — Atochem (Case T-3/89)

According to Article 1 of the decision, Atochem took part in the cartel at least from 1978 and until it came to an end. That part of the decision fits awkwardly with the reasons given by the Commission in the second paragraph of point 105 of the decision where Atochem is treated in the same way as ANIC, BASF, DSM and Hüls. To that extent, the reasoning is defective. But since, upon my assessment of the evidence, the cartel's infringement of Article 85(1) cannot be regarded as proven in respect of the period before Autumn 1979, that defect is of no significance.

Atochem acknowledges that it took part in a certain number of meetings with other polypropylene producers between 1978 and 1983 but it denies having taken part in any infringement of Article 85(1). Atochem claims that as a newcomer on the market it needed to gather any information whatsoever, including information given at meetings of producers, without committing itself to any agreement at all regarding either prices or quantities. Atochem is mentioned by ICI as a regular participant throughout the whole period.

Atochem did send out written price instructions which correspond to what was mentioned at the producers' meetings and the evidence for its participation in quota arrangements and account management systems is the same as that regarding the other undertakings.

Atochem claimed in particular that the written price instructions it gave were prepared on the basis of information it obtained by reading European Chemical News. The Commission for its part points out that the decisive factor is not who was the first to announce a price increase but how it was adopted.

On this point I consider that the Commission's views must be upheld. In the case of the price initiatives adopted in 1979 and thereafter, there was no possibility, as in 1977, for a small group of producers independently to adopt decisions which were followed by the others after reading about them in the trade press. On the contrary, it is clear from the evidence that the situation was such that at the meetings decisions were taken or understandings secured regarding target prices which were then communicated to the public inter alia through the trade press.

In view of the foregoing, I consider that there is sufficient evidence to support the conclusion that Atochem took part in the cartel from Autumn 1979 until the cartel came to an end in Autumn 1983.

D — BASF (Case T-4/89)

Together with DSM and Hüls, BASF is mentioned in Article 1 of the decision as a participant in the cartel from a time between 1977 and 1979 and until Autumn 1983. If the Court accepts my premiss that the activities of the cartel can be regarded as constituting an infringement of Article 85(1) only as from Autumn 1979, much of the discussion in the case as to when BASF's participation began becomes redundant.

In its response to the Commission's request for information, BASF states that its employees took part sporadically in meetings during the period before 1 January 1980. With regard to the period after that date, BASF mentions four meetings in 1980 in which it took part and a large number of meetings in and after June 1982. In its application BASF describes its participation before June 1982 as occasional.

I do not think one should attach too much weight to BASF's description in the pleadings of its participation in meetings before June 1982. Moreover, the applicant's participation in the cartel from Autumn 1979 is demonstrated in other ways, in particular by the price instructions which the Commission used to prove the cooperation which followed the agreements entered into or concertation of practice that took place at the meetings. Thirdly, the applicant is mentioned by ICI as a participant in the meetings without any limitation in time. The least which may be concluded is that participation in the meetings was sufficient to have given BASF the opportunity to keep abreast with what was happening at the meetings and to adapt its conduct accordingly.

With regard to volume targets, it is clear from the tables produced that BASF took part on the same footing as the others. BASF's denial of having taken part in any account management system does not appear tenable. In addition, it would appear that it was Mr Arenz of BASF who was the first to draw attention to the risk involved if all the undertakings charged exactly the same price at the same time when approached by a customer. According to the report on the meeting held in September 1982 at which that point was made, it was agreed that undertakings other than the individual customer's principal supplier would offer to sell at a price which was a few pfennigs higher than the price agreed at that time, which was DM 2.00. The reports on the subsequent meetings show that BASF took part in the same way as the other undertakings in the attempts to limit the consequences of customers' shopping around to find where they could obtain the most advantageous prices.

On that basis I consider that it may be concluded that BASF did take part in the cartel during the period from Autumn 1979 until Autumn 1983 and did so on the same footing as the other undertakings, as alleged by the Commission.

E — ANIC (Enichem) (Case T-6/89)

ANIC has acknowledged that it took part in the producers' meetings at least from Autumn 1979. On the other hand, it is still not altogether certain when ANIC's participation in the meetings came to an end. In its answer to the Commission's request for information, ANIC writes that it began to take part in the producers' meetings at a time which was most probably not long after the start of the series of meetings. The first paragraph of point 19 of the decision states that ANIC did not take part in the meetings after about the middle or end of 1982 when ANIC's interests in the polypropylene sector were taken over by Montedipe. However, the Commission considers that ANIC's participation in the cartel lasted until the end of 1982 or the beginning of 1983 (see Article 1 of the decision).

ANIC states in its reply to the Commission's request for information that the last time it took part in a meeting was in October 1982 in Zurich. ANIC later stated that that information was erroneous and that it probably stopped taking part in meetings at the beginning of 1982.

A report on a meeting in May 1982 states that ANIC/SIR were no longer coming. A report on a meeting in September 1982 records that ANIC must be regarded as a problem. It was stated that it was necessary to exercise pressure and that Mr Zacchi (Monte) was requested to get Mr Morioni to talk to Mr Corradini (ANIC). In a report concerning a meeting held on 2 November 1982 it is noted with regard to the situation in Italy that ANIC, according to Monte, together with SAGA, AMOCO and BP, gave cause for concern.

As the Commission acknowledged at the hearing, there is hardly any evidence in the case that ANIC took part in meetings after the middle of 1982 and in any event the comment in the report of the meeting held in May 1982 to the effect that ANIC was no longer taking part does not seem to be contradicted by other evidence. It is possible that ANIC may have given information to be used in attempts to bring about a quota agreement for 1983, but in my opinion there is not sufficient evidence in the material available to assume that ANIC took part in the activities of the cartel after the middle of 1982. Accordingly, it should be assumed that ANIC took pan from Autumn 1979 until the middle of 1982 and not, as found by the Commission, until the end of 1982 or the beginning of 1983.

In the period up to the middle of 1982, ANIC's participation in the cartel's price initiatives is largely borne out by its participation in the meetings. As mentioned in the section devoted to Rhône-Poulenc, this must be regarded as sufficient evidence in the present situation where participation in many meetings over a long period is hardly conceivable without participation in the measures decided on at the meetings. For the rest, there is no basis for assuming, as ANIC submits, that actions which are attributed to ANIC could just as well have been undertaken by representatives of SIR. In point of fact ANIC is not denying its own participation and in most of the tables which have been found the two undertakings are indeed mentioned separately.

The fact that ANIC and SIR are mentioned together in the meeting report for May 1982 presumably signifies that neither of those undertakings was any longer taking part in the meetings.

With regard to the quota arrangements for the period during which ANIC did take part, the evidence is the same as against the other undertakings. The account management systems were introduced only after the time at which ANIC's participation must be taken to have come to an end.

F — Hercules (Case T-7/89)

Hercules has specifically raised the question of the extent to which an undertaking may be liable for a fine in a case where an employee may have acted on his own initiative and contrary to his superiors' instructions.

It claims that Mr Bastiaens's participation was unofficial, that he did not have his superiors' approval and that his participation was contrary to company policy. The Commission for its part contends that the participation of Mr Bastiaens was known or should have been known to his superiors, who, in the Commission's view, must at least have given their tacit agreement.

There is some mystery surrounding Mr Bastiaens's participation in the meetings, or at least an attempt to cloak it in mystery. The first question which arises is, of course: What was he doing at the meetings at all? The explanation given by Hercules for his participation is that its management did not make it sufficiently clear to him that the applicant did not wish to take part in the arrangements of which it became acquainted during the first phase, namely in 1977. It is suggested that Mr Bastiaens, like a moth, fluttered around the flame until he was engulfed by it and began to take part in the meetings. According to Hercules, he was responsible for predicting price movements in the market and preparing price guidance for the affiliated sales companies and his ability to perform his job would of course be enhanced if he knew the other undertakings' intentions. Hercules acknowledges that it may accordingly be found that its senior management did not exert sufficient vigilance as regards what was occurring but it believes that a lack of vigilance cannot be equated with a knowing participation on the undertaking's part.

The Commission, as stated above, considers that the superiors of Mr Bastiaens must have known what was happening and it points to the original note on the floor-price agreement and to an internal note dated 29 July 1981 both of which the Commission says, without being challenged by Hercules, were addressed to Mr Bastiaens and written by his superior. The latter note says : If you can believe it, here are the official prices for Aug. and Sept. and then sets out a series of prices. It ends: These brought to me on two tablets by a bearded gentleman coming down a mountain.

The applicant further explains at length that Mr Bastiaens's position did not entail any power to commit Hercules, and in particular that he did not have authority over production and thus could not undertake to adopt agreements on quotas. It is stated that he was subordinate, on the one hand, to a sales director and, on the other, to the undertaking's managing director, to whom the sales director reported. The Commission, for its part, lays great weight on the fact that Mr Bastiaens had the title of Marketing Manager. The applicant stresses that Mr Bastiaens repeatedly indicated that his participation was unofficial and it states that he probably did not so much give information as gather it. The Commission, on the other hand, points out in particular that Mr Bastiaens was apparently very active and had good ideas and that in any event he was present.

The picture given by the information thus available is somewhat unclear. Is it, as Hercules maintains, a case of an ambitious employee who for career reasons disregarded orders which were perhaps not framed particularly clearly? Or is it, as is indeed implicit in the Commission's view, a case of an undertaking which was deliberately trying to reduce the risks of being discovered and fined?

It appears from the case-file that Mr Bastiaens was a marketing manager and was responsible for assessing the market and consequently may actually have had an influence on the fixing of prices. The applicant thus indubitably gained some advantage from the information obtained. It is also demonstrated by the Commission that Mr Bastiaens's superior in some cases did know what was happening at the meetings and the note dated 29 July 1981 does not in any event give the impression that either Mr Bastiaens or his superiors were unaware of the general issue. On the contrary, there appears to be a considerable degree of understanding. Finally, it has been explained that Mr Bastiaens occupied a post on the third rung from the top in a large undertaking.

How it could be possible to keep it a secret that an employee in a fairly senior position was taking part in meetings which the senior management only became aware of from time to time in some other way is puzzling. It is possible that to some extent Mr Bastiaens deceived his superiors. It is also possible that, through Mr Bastiaens, Hercules deceived its competitors and used Mr Bastiaens's somewhat mysterious participation in an attempt to reduce its risks. On an overall assessment, however, I am inclined to believe that the superiors of Mr Bastiaens must have known about and at least tacitly approved what was happening and that the applicant's objection can be rejected on that ground alone.

In the case of Melchers & Co. the Court of Justice laid stress on the powers inherent in the position of someone acting on behalf of an undertaking. The Court held that the burden of proving that the powers inherent in a post had been exceeded lay on the undertaking.

Where an employee is vested with powers by virtue of his position, it is clear that the limits on those powers must be determined a priori on the basis of objective factors. Normally, an undertaking cannot incur liability for acts which third parties can see are clearly outside an employee's powers. The legal position appears to be somewhat different, however, if an employee acts within the general powers inherent in his position but contrary to his orders or at the fringes of the powers inherent in his position.

There is no indication in the documents before the Court that Mr Bastiaens clearly and explicitly drew attention to the fact that he was in no way empowered to act in any way on the applicant's behalf and that his participation would in no way influence its conduct on the market. Anything of that kind would, moreover, be hard to reconcile with the position he actually held in the applicant undertaking; furthermore, I consider it doubtful whether the other participants in the meetings would in that case have allowed him to attend at all. His very position in Hercules gave the other participants no cause to think that his word carried no weight. On the contrary, to judge by the evidence available, it seems to have been assumed that he was in a position to act on behalf of Hercules.

The post actually held by Mr Bastiaens, and acknowledged by Hercules, combined with the fact that he had a position such as to enable him over a long period of time to give others the impression that he held certain powers must, in my view, lead to the conclusion that Hercules has failed to demonstrate that Mr Bastiaens exceeded the powers inherent in his post. It is, I believe, not enough for the applicant to maintain now, after the event, that Mr Bastiaens did not have a decisive influence on production and so forth if in point of fact it has allowed a senior employee who was known to have substantial influence on price policy to have such a free hand that he was able to take part. I therefore consider that this argument must be dismissed.

According to the Commission, Hercules took part in the cartel from its beginnings until it came to an end in November 1983. The applicant, on the other hand, contends that at most it played an essentially minor role and took part for a considerably shorter period than is alleged by the Commission.

Hercules acknowledges that it took part in meetings from May 1979 but contends that its participation until May 1982 was sporadic and after that date certainly not regular. After August 1983 no representative of Hercules took part in the meetings or had contact with its competitors. The applicant points out in particular that ICI, in its response to the Commission's request for information, states that Hercules's participation at the meetings was not regular and that at the meetings Hercules did not give information regarding its own figures. Hercules further points out that Monte, in its answer to the Commission, stated that the major European producers generally took part in the meetings, apart from AMOCO, Shell, BP and Hercules, although Monte was not ruling out altogether the possibility that Hercules attended some meetings. Hercules considers that the Commission, for no reason, failed to take account of the applicant's low level of participation both in evaluating its blameworthiness and fixing the fine.

According to the evidence available, in the period before May 1982 Mr Bastiaens's participation was without doubt not as regular as that of many of the others. Hercules itself in the context of an internal inquiry discovered that prior to May 1982 Mr Bastiaens certainly or with great probability took part in meetings on 10 or 11 May 1979, 1 or 2 October 1980, 27 or 28 August 1981, 16 December 1981, 4 February 1982 and 9 or 10 March 1982. In addition, there is a meeting in Amsterdam, possibly in 1979, and perhaps another meeting in London. Furthermore, the Commission has pointed out that in the 1983 meeting reports Hercules is regarded as a regular participant.

In so far as concerns reports of meetings and similar evidence, there is the report for a meeting held on 10 March 1982 which is referred to in point 15(b) of the decision (Annex 23 to the general part of the Statement of Objections) and also a report on a meeting held on 13 May 1982. Written price instructions exist for the period from March 1982. For the period before March 1982 there is the abovementioned note stating that the official prices were given on two tablets. There is also a copy of a survey of price developments allegedly used by Mr Bastiaens as a basis for a talk to the applicant's salesmen. It stated inter alia that a general resolute attitude had brought prices up to DM 2.05, the closest they had yet come to published target prices. Finally, it should be pointed out that the note concerning the original 1977 floor-price agreement stemmed from Hercules.

It is clear from the foregoing that Hercules must have taken part in the cartel at least from March 1982. I would not attach decisive importance to the fact that, according to the information available, Mr Bastiaens did not himself come forward with figures concerning his employer. The part he played in meetings and discussions and the fact that Hercules did send out price instructions which corresponded to what had been agreed leave no reasonable doubt that the other undertakings could count on the applicant's price policy following theirs, which must be sufficient to find that Article 85(1) has been infringed. The evidence regarding the period between Autumn 1979 and March 1982 is somewhat weaker. But it is established that Hercules acknowledged that it took part in a not inconsiderable number of meetings in that period and the other evidence shows that Hercules in any event was extremely well informed of what was happening in the cartel. I am therefore compelled to take the view that there are the requisite grounds for concluding that Hercules's participation extended from Autumn 1979 until the cartel came to an end in 1983. It should be noted that Hercules's participation in the 1980 quota arrangements, whose existence must be held to be proven, is no less well documented than that of the other undertakings.

In its pleadings Hercules makes much of the fact that the documentary evidence on which the Commission relies consists of second or third-hand hearsay about circumstances which Hercules was prevented from examining more closely, in particular because of lack of knowledge about the origin of the documents. As stated above in Section I, E, 2, however, it is the overall assessment of the evidential weight of a document which must be decisive and the bulk of the documents relied on by the Commission constitutes, by any reasonable yardstick, very strong evidence. It is, moreover, difficult to see what may have prevented Hercules from seeking to ascertain who was the originator of which documents to the extent that the originator is known, for example by approaching the Commission.

G — DSM (Case T-8/89)

DSM acknowledges that it took part on a fairly regular basis in the meetings of the cartel from 1 January 1981. As regards the period prior to that, it denies taking part in the meetings on a regular basis or in any structured form.

The Commission, on the other hand, which in Article 1 of the decision held that DSM had taken part from some time between 1977 and 1979, contends that, according to the available volume target schemes for 1979 and 1980, DSM must at least have taken part from 1979.

If the Court endorses my assessment of the evidence concerning the period from 1977 until Autumn 1979, the matter in dispute must accordingly be DSM's participation in meetings from Autumn 1979 until the end of 1980. In view of the fact that ICI stated in its reply to the Commission's request for information that DSM was one of the regular participants, undue weight should not be attached to DSM's own description of its participation in the meetings before 1 January 1981 as unsystematic or unstructured, particularly where it is a question of a period in which meetings were held far less frequently than they were later. Since in addition DSM is mentioned on the same footing as the other participants in the schemes regarding quotas and volumes for 1979 and 1980, I have no hesitation in holding that DSM did take part in the cartel from Autumn 1979.

Furthermore, DSM did send out a considerable number of price instructions and it is mentioned in the documentation as an account manager. With regard to quota arrangements, the evidence against it is as strong as that against the other applicants.

DSM further refers to the maxim in dubio, pro reo and claims that the Commission may not demand that the undertakings should produce a convincing alternative explanation for what the Commission regards as incriminating evidence. I find it hard to see how such general considerations cast any new light on the questions of proof arising in this case. The practical reality is that there must be proof of an infringement and that whoever is to judge the evidence must be satisfied, upon an overall assessment of the weight of the evidence, that the case made out by the Commission is sound. As will have become apparent, I consider that the standard of proof should be set at a higher level than that applied by the Commission; however, there is nothing in the text of the decision to suggest that generally the Commission disregarded general principles regarding the appraisal of evidence, including the principle in dubio, pro reo. I therefore consider that the submission should be rejected.

H — Hüls (Case T-9/89)

According to Article 1 of the decision, Hüls took part in the cartel from a time between 1977 and 1979 and at least until November 1983. On the view I expressed in the general part of my Opinion, there is, however, no question of liability for any period before Autumn 1979.

With regard to price initiatives, Hüls has largely based its case on the contention that there was not or could not have been any agreement or concerted practice falling under Article 85(1). The first part of its argument relates to the question of interpretation at the heart of the present cases, namely the interpretation of the concepts of agreement and concerted practice with regard to what occurred in these cases. The second part of its argument rests on considerations over the question whether its conduct had any effect on the market and the conclusions which some purport to draw therefrom. Hüls stresses in particular Professor Albach's view that actual price changes are difficult to reconcile with any assumption that prices were fixed by means of cartel agreements.

With regard to that part of the applicant's case, I can refer generally to my comments set out above regarding the interpretation of Article 85 and the evidence as a whole, although it may be appropriate to emphasize once again that economic science cannot determine and must not be allowed to determine, on the basis of theoretical considerations, when an agreement or concerted practice within the meaning of Article 85 exists. That is a matter of legal determination to be made on the basis of an overall evaluation of all the evidence produced, including, of course, the experts' opinions presented to Court in this case.

There remains the question of the applicant's participation in meetings and what is otherwise disclosed by the documentary evidence about the role played by Hüls in the case. In this connection Hüls states first that, according to the annex submitted by the Commission, Hüls took part in meetings on one occasion in 1981 and more frequently as from 1982. It did not take pan in local meetings concerning territories within the EEC. Hüls also did not take part in a meeting held in the United Kingdom on 18 October 1982.

In its answer to the Commission's Statement of Objections, Hüls does not seek to deny that it took part in the plenary meeting in January 1981, nor is the possibility ruled out that its employees took part in a few other meetings in 1981. In any event, it says, it must be assumed that Hüls took part at the earliest in meetings as from January 1981 and took part on a regular basis at the earliest from around May 1982. Regarding local meetings, it is said that a representative of the applicant only took part in one or two discussions concerning the Scandinavian market.

The Commission, for its part, refers first and foremost to ICI's information regarding the participation of the other undertakings in the meetings in which ICI states that Hüls was one of the regular participants in the meetings. The Commission also emphasizes that Hüls is mentioned in the lists concerning revised volume targets for 1979.

In the light of the foregoing, Hüls may be considered to have essentially acknowledged that it took part from the beginning of 1981 and that its earlier participation is denied. On the other hand, there is information provided by ICI, whose reliability has, moreover, largely been supported by the other applicants' varying degrees of acceptance, and the fact that there is no other information, as was the case for Petrofina, giving any other indication that Hüls took part from a somewhat later date than Autumn 1979, the time which must be assumed to have been the real beginning of the cartel.

Moreover, the written price instructions point unequivocally to Hüls having taken part on the same footing as the other smaller producers and Hüls is mentioned in exactly the same way as the others.

For the rest, the essential thrust of the applicant's defensive strategy is to maintain that at the meetings it followed a policy based on a combination of mental reservations and misinformation, which is said to be apparent in particular from the differences between the target prices set and the prices actually achieved.

As will have become apparent from my general observations, any difference between target prices and the sales prices actually obtained do not, in my view, signify that the undertaking did not take part in the implementation of the common plans. It is almost inevitable that undertakings which are also in competition with one another will to a significant extent seek to promote their own interests at the expense of others where there is an attempt at collusion but it would hardly have been possible to maintain over a long period of time an attitude of continuing disloyalty to the others, as Hüls claims to have done, without this showing up in the documentation available. The strongest indication of this happening is that some of the notes on meetings suggest that some of the smaller producers (not Hüls incidentally) were ambitious and that some were hooligans. In the absence of concrete indications to the contrary, it must be justifiable to describe Hüls's explanation as implausible. The documents in the case show that Hüls also took part in the quota arrangements and account leadership system.

In those circumstances I consider that there is sufficient evidence to support a finding that Hüls did generally take part in the cartel from Autumn 1979 until around November 1983.

I — Hoechst (Case T-10/89)

In its decision the Commission regards Hoechst as one of the four major producers who must bear particular responsibility for the activities of the cartel — see in particular points 67, 68 and 78 of the decision.

Hoechst does not deny that it took part in the meetings that were held but points out that those meetings were not particularly frequent in the period before 1981.

Hoechst's defence is essentially that sales were not transacted at the agreed target prices and that its turnover did not match the agreed quotas either. In its pleadings Hoechst also points to the weaknesses which are actually to be seen in the Commission's evidence. Finally, Hoechst points out that it was most unlikely that the information exchanged at the meetings could dispel the uncertainty regarding the future conduct of the other undertakings on the market and thus eliminate competition. The questions thus raised have all been dealt with in the general part of my Opinion.

With regard to Hoechst's participation in the special group of big producers it is apparent from the annex concerning the 1977 floor-price agreement that Hoechst is included in this group. As is clear from my evaluation of the evidence regarding the situation in 1977, I agree with the Commission that it may be taken to be proven that at that time Hoechst took part in the core agreement alleged by the Commission. According to ICI's reply to the Commission's request for information, there existed, as mentioned in the first paragraph of point 68 of the decision, a special understanding between the four biggest producers that if prices were to be increased those four undertakings would have to take the lead even at the expense of their own sales. A Shell note (Annex 94 to the general Statement of Objections) records that price initiatives taken by the big producers (Hoechst, MP, ICI, Shell) had hardly any effect. At the end of 1982 the Big Four began to hold separate meetings. According to Table 5 annexed to the decision, all those undertakings took part in a total of seven meetings of that kind between 13 October 1982 and 22 August 1983. Hoechst denies that it took pan in the meeting on 13 October 1982 at Heathrow and indeed ICI has not clarified whether Hoechst took part. It is further apparent from the second paragraph of point 67 of the decision that Hoechst also did not take part in a meeting with the other big producers on 17 June 1981 at which the various possibilities were discussed.

In my view, the abovementioned written evidence, together with the fact that Hoechst — subject to very few exceptions — does not deny its participation either in the general meetings or in the special pre-meetings of the four big producers and that Hoechst does not contest the information provided by ICI about the subject-matter and purpose of the meetings, provide quite sufficient grounds for finding that Hoechst was a participant in the activities of the cartel, as maintained by the Commission, with the exception of that part of the activities which, as I have observed in the general section of my Opinion, cannot be held to be sufficiently proved.

J — Shell (Case T-11/89)

It is plain that Shell's role in the activities of the cartel is the least well documented since it did not take part in the bosses meetings' and experts meetings' of the other undertakings.

However, as mentioned in the preceding section, Shell did take part in meetings between the four biggest producers and also in a series of so-called local meetings. It is known that the meetings between the four biggest producers, which are described in the second paragraph of point 68 of the decision as pre-meetings and which were held in advance of the so-called bosses meetings', took place on the day before the bosses' meetings for, as ICI says, practical reasons, because the participants' hierarchical level was the same as in the bosses' meetings.

In the procedure before the Court, as in the administrative procedure, Shell denied that those meetings could in any way be regarded as pre-meetings for the bosses' meetings or that the meetings served to coordinate the participants' positions on matters to be discussed the following day. At the hearing before the Court we were given the impression that they were social gatherings without any specific purpose. The explanation thus given by Shell appears at first sight to be somewhat implausible and, as the Commission points out in its decision, Shell's assertion is contradicted by the information available, including the information about the subject-matter of the meetings between the four big producers in October 1982 and May 1983.

As far as Shell's participation in the original floor-price initiative is concerned, my assessment of this question is apparent from the relevant observations set out in the general part of my Opinion — see Section I, F,5.

As far as the period between Autumn 1979 and the end of 1982 is concerned, we know that Shell has acknowledged that it often received information from ICI concerning target prices and so forth. We also know from various meeting reports that there was significant communication in the opposite direction, with information reaching the participants in the plenary meetings regarding Shell's position on various matters. It is evident from the reports on meetings on 27 May 1981, 17 June 1981 and 9 and 10 September 1982 that there were frequent contacts between ICI and Shell. The meeting on 1 July 1981, also attended by Montepolimeri, referred to in the second paragraph of point 67 of the decision, should also be mentioned.

From Shell International Chemical Company and Shell in the United Kingdom there exist price recommendations and price instructions respectively which follow up on the agreements that were concluded or concertation that took place at the plenary meetings. There is further sufficient evidence for assuming that Shell took part in the quota arrangements for 1980. Finally, according to the evidence, the national Shell organizations took part to a certain extent as coordinators in the account leadership systems.

That body of evidence as a whole constitutes, in my view, an adequate basis for finding that Shell took part both in the cartel itself and in the separate group constituted by the four big producers as alleged by the Commission. The fact that Shell did not physically attend the plenary meetings, either at the expert or boss level, was not, in my view, decisive to the extent that it is documented, as in this case, that Shell supplied to and received from the meetings the relevant information and appears to have acted according to the outcome of the meetings.

Finally, I would point out that I attach no particular weight to Shell's objections concerning the Shell group's structure or internal organization which is relatively decentralized — see the first and second paragraphs of point 102 of the decision. The main thrust of Shell's argument here is that the undertaking to which the decision was addressed, namely Shell International Chemical Company, had no form of authority to issue instructions to the companies which sold polypropylene and that Shell International Chemical Company, which is described by Shell as a mere service company, had to obtain approval from the national Shell undertakings in order to conclude any agreement whatsoever. However, according to the relevant documentation, it is established that it was Shell International Chemical Company which took part in the pre-meetings and coordinated the notification of prices. There can therefore hardly be any doubt that it was that company which stood at the centre of events as far as Shell was concerned and Shell's arguments based on its internal structure are therefore, in my view, best regarded as an attempt to use that structure to evade liability to a fine, which is of course unacceptable.

K — Solvay (Case T-12/89)

In Solvay's reply to the Commission's request for information, Solvay acknowledges its participation in the meetings of producers from the beginning of 1978, which corresponds to the information given by ICI. Solvay thus took part in the cartel during the period in respect of which there is proof of an infringement of Article 85(1), that is to say from Autumn 1979 until the end of 1983. I would especially point out here that participation in the original floor-price agreement cannot be considered proved, in particular because the note of 6 September 1977, which is referred to in the fifth paragraph of point 16 of the decision and was the subject of much discussion in the course of this case, was not even properly communicated to Solvay in the manner laid down by the Court of Justice.

Solvay is one of those undertakings which describes its own role as completely passive. That argument is refuted by the meeting reports to hand and the factual circumstances on which the Commission bases its assessment are in actual fact hardly challenged. Solvay states in its application that it took part in the meetings only with the object of gathering technical and commercial information that could be used to overcome the handicap stemming from the fact that Solvay was a newcomer to the market. But it further states that the applicant otherwise does not seek to deny that that exchange of information, to the applicant's knowledge, could have been capable of limiting the effects of the crisis which had arisen because of the excessive production capacity on the market. Solvay further points out that Solvay itself and many of the other undertakings acted duplicitously. It is thus pointed out that at the oral hearing before the Commission Solvay stated that at the meetings in question the art of bluff reached extreme heights.

On reading the meeting reports one gains the distinct impression that there was a climate of constructive mutual mistrust between the persons attending the meetings but also that they worked seriously on the problems confronting them. I do not consider it possible to derive from Solvay's duplicity theory anything that might be of vital importance in determining whether there was an infringement of Article 85(1).

The same applies to the argument that the interests of the participants were so diverse and mutually incompatible that this fact alone precluded the conclusion of any agreements or the mounting of any concertation. This line of argument in fact turns largely on the participants' motives: while the undertakings well established on the market in 1977 had to try to hold on to their market shares and still try to obtain higher prices, the interest of the new producers lay in winning a share of the market, if necessary by undercutting the existing undertakings.

As is evident from the proceedings, that argument is factually incorrect. There are many references in the documentary evidence to the fact that other parties, that is to say in particular the big producers, are restricting their sales and voluntarily giving up market shares. As a theoretical argument, it is also untenable. As stated above in footnote 1, in some circumstances it may, from the point of view of an established undertaking, make just as much sense to attempt to integrate a new undertaking on the market peacefully as to start a price war, for example.

Solvay's other arguments concern points which have been treated above in the general part of my Opinion. In Solvay's case, too, there are no grounds for doubting that price and quota agreements were concluded and concertation entered into, even if products were sold at lower prices and even if the quantities sold did not always correspond to the quota allocated.

I therefore conclude that it must be held that Solvay did take part in the cartel from Autumn 1979 until the cartel came to an end in Autumn 1983.

L — ICI (Case T-13/89)

The position as regards evidence is most straightforward in the case of the applicants right at the centre of events, ICI and Montedipe. These cases in fact largely concern arguments going to the question of the limits of lawful conduct, and not so much to an assessment of the evidence against these two undertakings, which without exaggeration can be described as rather overwhelming. Both undertakings acknowledge the factual circumstances constituting in the Commission's view the offence covered by Article 85(1) and both seek by various means to demonstrate that the conduct displayed was lawful.

Otherwise, ICI's strategy is to attack every point where that is in any way possible. In the general part of my Opinion I have essentially addressed ICI's arguments.

ICI further maintains inter alia that it cannot be assumed that the conduct displayed by the applicants actually affected trade between the Member States. In its view, the activities of the cartel neither increased nor diminished trade and had no influence on the structure of competition either. Trade between the Member States was already considerable and it increased during the period in which the cartel was in operation.

Those arguments must be compared with the point made by the Commission (in points 93 and 94 of the decision) that the fixing of target prices etc. must have had an effect on patterns of trade and must have distorted them.

It is obvious that if one assumes that the cartel had no effect whatsoever on competition, it must at the same time be concluded that it did not affect trade between Member States either. However, it is equally clear that agreements and concerted practices of the kind used by the cartel as a means to try to obtain higher prices may affect trade between Member States, which is indeed the criterion laid down by Article 85(1) and the established case-law of the Court of Justice. One or more undertakings might, for example, have been induced to withdraw from the market because competition without the cartel was too intense. It is clear that in such a case the market would have been different from what it in fact became. ICI's objection must therefore be rejected.

ICI's leading role in the cartel also calls for a number of other observations. ICI claims that chairmanship of the group did not entail any greater degree of involvement: the chairman was merely responsible for practical coordination. The Commission, on the other hand, believes that ICI played an altogether central role in the cartel. Apart from taking part in plenary meetings, Big Four meetings and in a long series of so-called local meetings in many countries, it appears from the evidence available that ICI acted more or less as a kind of whipper-in. In this respect reference may be made in particular to the documents mentioned in points 40 and 59 of the decision from which it is quite apparent that it must have been ICI which from the middle of 1982 was the undertaking working the most keenly to make the cartel function as it was intended. I therefore conclude that the Commission's assessment of ICI's role in the cartel must be accepted.

ICI claims that the Commission's assessment of the product market was marked by several errors (which are not referred to elsewhere in this Opinion). ICI states that the relatively insignificant quantities of polypropylene which, according to point 7 of the decision, are imported into the Community may have a highly significant negative effect on market prices. According to ICI, it was only in the second half of 1983 (and therefore not in 1982) that a reasonable balance was reestablished between supply and demand on the West European market. Finally, ICI contends that the Commission underestimated the problems facing the industrial sector in question by failing to take account of the substantial drop in prices after 1977 of products that were substitutable for polypropylene, including other plastic products. The threat that demand would turn to other materials thus made it difficult to hold up polypropylene prices which further aggravated the undertakings' problems.

As the Commission rightly points out, it is hard to see how any errors of assessment in those respects can affect the lawfulness or correctness of the decision. The question whether the impact on prices of a modest volume of imports is great or small is altogether irrelevant in assessing liability and the gravity of the infringement. That it was difficult to hold up polypropylene prices was indeed plain, yet ICI's views are tantamount to claiming that the undertakings should be rewarded for venturing upon an infringement of the law which was not easy to commit with any success because the task in itself was a difficult one. The fact that balance was possibly not restored to the market until some time later than the participants supposed cannot, in my view, alter the fact that the participants' disposition to continue the infringements at a time when they believed, rightly or wrongly, that the market was in balance must if anything be regarded as an aggravating factor — see the third paragraph of point 37 of the decision and Section III, B, below.

M — Montedipe (Case T-14/89)

Like ICI, Montedipe essentially acknowledges the factual circumstances on which the Commission's decision is based. As the Commission points out, the case put forward by Montedipe against the decision is essentially based on the view that the conduct in question must in effect be regarded as lawful in the particular circumstances obtaining in the polypropylene sector during the material period. Montedipe's arguments are characterized by their great originality and can be summarized as follows.

First of all, the activities of the cartel must, in Montedipe's view, be regarded as lawful because its object was to prevent the undertakings concerned from undercutting prices, which would have constituted unfair competition. And since Article 85(1) does not protect unfair competition, the cartel was not incompatible with that provision. As the Commission rightly observes, the applicant's argument certainly cannot support the theory that the cartel may have been lawful. One of the aims of the competition rules is to prevent one or more undertakings which hold a dominant position on the market from, individually or by agreement, keeping prices artificially low in order to prevent other undertakings from entering the market.

In the present case, however, the situation was the reverse. Both the established undertakings and the newcomers were seeking to keep prices artificially high and in any event there can hardly be any question of unfair competition where the (low) price which is obtainable is accepted. It is also notable that, in answer to a question at the hearing, Montedipe stated that an undertaking acting on its own would hardly have been able to out-compete the new producers.

As noted in the introduction to this Opinion, commercial undertakings, too, have the right to meet and jointly safeguard their interests. Nobody has disputed that right. According to Montedipe, however, the Commission's decision constitutes a breach of freedom of opinion, freedom of information, freedom of assembly and freedom of association. Clearly, it will always be necessary to be vigilant against overzealous public officials who see in every meeting between businessmen a place where unlawful agreements or unlawful concerted practices are forged. In spite of the obvious difficulties which may confront the Community authorities in gathering evidence in cases of this kind, consideration of the evidence must never develop into pure speculation. On the other hand, as mentioned above, it must to some extent be permissible, without undermining any basic freedoms, to deduce from an undertaking's participation in a long series of meetings at which something unlawful occurred that it took part in the unlawful acts. It is in fact a matter of keeping the assessment of the evidence within generally acceptable bounds. I am satisfied that in the present cases the evidence is in itself sufficiently cogent to refute any theory that the decision entailed a de facto breach of any basic freedoms.

I think that the remainder of Montedipe's argument, which is not considered elsewhere, can be summarized in a few words: Montedipe believes that considerations of necessity may be invoked to argue that the cartel was lawful. Using a striking analogy, it explains that its conduct may be compared to that of a group of shipwrecked people who all follow the call of one of their number to swim to land, this being the only rational course. Montedipe asks whether that call is a proposal for an agreement or just a simple statement of the only way to salvation. In terms of competition law, the answer must categorically be that it constitutes a call to conclude an agreement. It is certainly not the object of the competition rules, as they operate under Article 85 of the EEC Treaty, to prevent anyone from drowning. The competition rules under the EEC Treaty contain no reference to any solidarity principle, as exists within the context of the ECSC Treaty, and in any event it is not up to undertakings — which are expected to compete — to try to introduce such a principle. Montedipe further alleges that the cartel had particularly beneficial effects since production, sales and consumption all increased while imports were reduced. As the Commission states, it may be highly debateable whether the cartel did have or could have had such effects. In any event, they are not established and the prohibition laid down in Article 85(1) applies regardless of whether some undertakings may have been able to increase their sales through cartel agreements.

Montedipe's leading role is less well documented than that of ICI. On the other hand, it is not disputed that Montedipe acted as chairman of the group until the middle of 1982 and there is no reason to believe that chairmanship of the group entailed less important tasks before it was assumed by ICI.

N — Chemie Linz (Case T-15/89)

According to Article 1 of the decision, Chemie Linz took part in the cartel from November 1977 until it came to an end.

Chemie Linz has claimed that it in any event did not take part in the meetings of the cartel from its inception. It has pointed out that its participation is documented only from the beginning of 1981. Otherwise it says that it is no longer in a position to determine from what date it began to take part in meetings. In its answer to the Statement of Objections, it stated in this connection that its participation in the meetings from the beginning is most improbable in view of its weak position in the common market. It further points out that ICI's information regarding the participation of other undertakings does not contain any further details regarding the period in which its own participation took place. Chemie Linz also essentially claims that it took part merely in order to obtain information.

The Commission for its part points out that Chemie Linz must have taken part in the quota arrangement for of 1979 and that, according to ICI's information, Chemie Linz regularly took part in the meetings, without any limit in time.

In my view, an examination of the meeting reports and notes produced as evidence in this case, together with ICI's account of the participation of Chemie Linz in the cartel, shows that Chemie Linz took part on the same footing as most of the other small undertakings with regard to target prices, quotas and account leadership. Finally, it is clear from the meeting reports that by means of price instructions corresponding closely to the respective meetings of the cartel, Chemie Linz sought to achieve the target prices. Apart from the applicant's own observations, there is nothing in the case to weaken the credibility of the evidence of its taking part in meetings from Autumn 1979, as appears from a review of the evidence.

I therefore consider that the evidence supports the conclusion that Chemie Linz took part in the cartel from Autumn 1979 until around November 1983.

III — Penalties

A — The Commission's fining policy

Pursuant to Article 15 of Regulation No 17/62 the Commission can impose fines on undertakings for intentional or negligent infringements of the competition rules. The fines can constitute up to 10% of each undertaking's turnover in the preceding business year. Under Article 15 regard must be had both to the gravity and to the duration of the infringement.

In the cases before the Court fines were imposed which at first sight seem very high in relation to the level of fines in other competition cases. It is probably true to say, as was submitted in this case, that the sum of the fines in the polypropylene cases is greater than the sum of all fines imposed in previous cases. Comparison with, for example, the survey contained in Bellamy and Child at pages 498 to 500 also gives the impression that the level of fines is high. It is not and cannot be disputed, however, that the fines in the cases now before the Court constitute a relatively small proportion of the maximum level of 10% of the undertakings' total turnover laid down in Article 15(2) of Regulation No 17/62.

All the undertakings have nevertheless argued that the fines are excessive. The applicants' objections are of two kinds: objections concerning the general level of the fines and objections directed more specifically at the situation of the individual applicant.

It is not directly stated in the decision, but the magnitude of the fines in comparison with fines previously imposed and the general part of the Commission's defence suggests that even though there may not have been any real increase in the general level, the Commission has in any event imposed fines which reflect an upward tendency. It may therefore be appropriate to examine the attitude of the judicial authorities towards the Commission's fining policy as such.

In 1979, in the Pioneer decision, the Commission significantly increased the level of fines in respect of infringements which involved an established practice or were otherwise considered by the Commission to be particularly serious. In the Commission's view, the level of fines was not sufficiently high to have a preventive effect in relation to undertakings which might reckon on obtaining such significant advantages from unlawful activity that it could be worthwhile running the risk of a relatively small fine. In its Thirteenth Report on Competition Policy the Commission gave a more general statement of its more severe attitude towards fining policy (p. 56 et seq.).

In that report the Commission stated that after about 20 years' experience of enforcing the competition rules, during which time it had imposed relatively light fines, it had found that fines of that size were not proving adequate to deter companies from continuing to commit even quite clear-cut infringements. The Commission went on to state that in a decision taken at the end of 1979 it had indicated that it intended to reinforce the deterrent effect of fines by raising the general level thereof in cases of serious infringements, that is to say in particular those for which fines had been imposed in the past and had been confirmed in judgments of the Court of Justice, such as export bans, market partitioning and horizontal and vertical price fixing, in the realm of restrictive agreements, and in the sphere of abuses of dominant positions, refusals to supply, price discrimination, exclusive or preferential long-term supply agreements and loyalty rebates.

The Commission went on to state that the complexity of the factors to be weighed meant that the assessment of fines, rather than being a mathematical exercise based on an abstract formula, involved a legal and economic appraisal of each case on the basis of the above principles.

In its judgment in the Pioneer case the Court of Justice approved the new, more severe attitude on the part of the Commission. The Court of Justice held inter alia (at pages 1905 et seq.):

In that judgment a number of issues which were also discussed in the cases now before the Court were in fact decided. As I understand it, the Court for the first time provided unmistakable support for the Commission's view that as a matter of principle it is responsible for formulating penalties policy in relation to the Community's competition rules. Indeed, that point of view is well founded. It is the Commission that has direct contact with the relevant area of the law and it is the Commission that deals with the great bulk of cases, whereas which cases come before the Court of First Instance and the Court of Justice is in the nature of things a somewhat random matter.

It follows, in my view, that in normal circumstances the Community judicial authorities should not pursue their own penalties policy but — within the limits of Regulation No 17/62 — leave it to the Commission to set the general level of fines. In that respect I am thus of the same view as Mr Advocate General Warner, who stated in the BMW case that the Court's unlimited review jurisdiction under Article 17 of Regulation No 17/62 (see Article 172 of the Treaty) does not mean that the Court should in every case substitute its own assessment of what constitutes an appropriate fine for the Commission's.

In my view, there is only reason to intervene where the Commission, without giving reasons, departs from a relatively well-established level of fines in a single case and thus acts contrary to the principle of equal treatment. However, in my view there are no grounds for concluding that the Commission has done so in the cases now before this Court, as the applicants submit. In the parallel decisions in the PVC and LdPE cases the fines also appear high, particularly when it is borne in mind that in assessing the fines in those cases due regard was had to the fact that fines had been imposed on most of the undertakings for their participation in the polypropylene cartel. In the most recent somewhat comparable decisions the level of fines is also high: ECU 7000000 ECU each for Solvay and ICI in Case IV/33.133-A, ECU 3000000 for Solvay and ECU 1000000 for Chemische Fabrik Kalk in Case IV/33.133-B, ECU 20000000 for Solvay in Case IV/33.133-C and ECU 10000000 for ICI in Case IV/33.133-D.

Since there is thus no basis for the conclusion that in the polypropylene decision the Commission departed from the general level of fines which it has applied since the Pioneer judgment, this Court should take as its point of departure the level of fine which the Commission has applied in the cases now before it.

Even if it is not decisive, I consider this point particularly valid inasmuch as the polypropylene cartel continued to exist notwithstanding the publication of the Pioneer decision on 5 March 1980, when the applicants were put on notice that it could become very expensive for them to infringe the competition rules.

B — Has the Commission taken all relevant factors into account f

Now that the general level of fines has been established, it is necessary to examine whether the Commission took all relevant factors into account in determining the amount of the fines. That must, of course, be apparent from the decision.

Under Article 15(2) of Regulation No 17/62, the basic criteria for determining the amount of fines are — in addition to the limits on the amount — the gravity and the duration of the infringement.

In its decision the Commission states first (point 107, second paragraph) that the infringement was deliberate. In the light of the evidence I think there can be no doubt that the Commission is correct to state that the undertakings were fully aware of both the unlawful nature of their conduct and the serious penalties of which they ran the risk. These are undoubtedly, therefore, deliberate infringements, which must in every case be dealt with much more severely than negligent infringements.

With regard to the latter criterion to be applied under Article 15(2), the Commission states that the infringement was of relatively long duration.

According to the assessment of the matter I have set out above, the period during which the cartel can be held to have been in existence ran from Autumn 1979 until Autumn 1983, that is to say about four years, and not about six years, the period the Commission took as the basis for its decision and thus for fixing the amount of the fines. It should be observed that liability to fines in respect of the infringements committed by ICI, Montedipe, Shell and Hoechst between Autumn 1977 and Autumn 1978, discussed above in Section I G, must be held to be time-barred.

Since the general level of fines applied by the Commission must, as set out above, be taken as a basis, there must be a certain reduction in the fines as a result of the fact that the duration of the infringements cannot be held to have been as long as the Commission maintains. The reduction should not, however, be proportional to the length of the period which is to be left out of consideration, since the organization of the cartel became increasingly professional in the period after Autumn 1979, whereas previously, in the Commission's view, it took less serious forms and was assessed accordingly in the decision. I think a reduction in the fines of about 10 to 15% would be appropriate in this respect.

In connection with the discussions regarding the duration of the cartel a number of the applicants argued that its duration should be determined solely on the basis of the periods in respect of which the Commission in its decision states that unlawful practices were committed: for example, that it is the total duration of the individual price measures and not the length of the period during which there were meetings that should be taken into account. Even though, as I have already said, I do not entirely agree with the Commission that a framework agreement was involved in these cases, I do not think the applicants strictly mathematical view of the duration of the cartel is right. The duration must be determined on the basis of the entire period during which there were activities directly connected with the unlawful practices.

No further justification is required for the conclusion drawn by the Commission (point 107, fourth paragraph, of its decision) that this was a particularly serious infringement. The applicants had the very clear intention of seeking to achieve a price level above that of the market, and in any event with regard to the period after the middle of 1982 it is difficult to take seriously the applicants' strongly expressed protests of reasonable and worthy motives for the infringement. Regardless of whether or not the assessment of those attending meetings that the market was more or less in balance at that time was in fact correct, the comment reported in a meeting note referred to in point 37, third paragraph, of the decision, to the effect that it was the participants and not the market that should determine the price level even if supply and demand were in balance, shows in any event that the purpose cannot have been to protect production capacity which might in the long term be viable under normal conditions of competition against a crisis. The purpose was quite simply to obtain a higher payment for goods than could have been obtained without infringing Article 85(1). As the Commission points out, the Court of Justice, too, has regarded price agreements as particularly serious infringements of Article 85, holding that agreements which prevent the supply of goods to consumers at the most favourable prices are particularly serious, and the Commission is justified in strictly exercising its power to impose penalties.

In my view, the gravity of the infringement cannot be directly assessed on the sole basis of the extent of its actual and detected harmful effects. First of all, part of the definition of the offence laid down in Article 85 takes no account of the effects at all. Agreements or concerted practices whose purpose is to interfere with competition can constitute just as serious infringements as the infringements regarded as equivalent to them whose effect is to restrict competition in an unlawful manner. Where it must be concluded that conduct did not have the object of interfering with competition but had that effect, the unlawful effect must at least constitute negligence on the part of the offender before a fine can be imposed. Cases in which there is only the infringement consisting of an agreement or concerted practice having as its effect... involve only the lowest grades of intention, since more serious intention — having the object — constitutes a part of the actual description of the offence in the other part of the definition of the offence contained in Article 85(1).

I think that an infringement which, for one perhaps quite fortuitous reason or another, does not have the desired effects but where the intention was of the highest degree is in reality more serious than an infringement with more far-reaching effects which were not, however, directly aimed at in the same way as in the cases now before us.

It is not simply the actual harmful effects but largely the potential harmful effects which must be emphasized. If we were to suppose that Article 85 created simply a result offence with an attached provision on attempts, the assessment of the gravity of an attempt would presumably not differ drastically from the assessment of the completed offence, where the attempt demonstrated a fixed intent to take every conceivable step to realize the unlawful purpose. I therefore agree with the Commission that the gravity of the infringement must largely be determined on the basis of the evidence presented to us concerning the cartel's intention to try to influence the market.

With regard to the cartel's intention to seek to undertake joint action, I do not think the case can give rise to much doubt. The fact that in practice it was often impossible to carry through the planned price initiatives makes no difference in that regard. In the period with which we are concerned the cartel in no way appeared dilettantish or unstructured. On the contrary, its organization appears to have been entirely professional, which in my view must be given considerable weight in the assessment of the gravity of the infringement. I therefore think that all the submissions challenging the Commission's assessment of the gravity of the infringement must be rejected.

However, the Commission based its decision on the fact that the cartel had a certain effect. Consequently, the calculation of the fines must undoubtedly be changed if the Commission was not correct in its assumption. As the Commission has described its assessment of the cartel's effects in points 72 to 74 of its decision, the actual influence on prices can even in its view have been relatively restricted. The central issue, says the Commission, is that the agreed or coordinated prices served as the basis for the negotiation of prices with customers. I think that is right, and the agreements or concerted action thus clearly had a certain influence on price formation. We have had the impression in this case that prices for individual consignments of goods were to a large extent set on an individual basis and it was certainly an advantage for the firms' sales forces to have a defined common price to guide them.

There is some doubt, on the other hand, about the extent to which the unlawful conduct in fact influenced prices. It is possible that the Commission went somewhat too far in seeing a direct causal relationship between the target prices which were fixed (points 90 and 91 of the decision) and the actual prices, in the sense that the former influenced the latter, whereas as a matter of principle the causal relationship could very well have been the reverse, as submitted by the undertakings. But even if the latter hypothesis is true, the fact that the prices actually achieved were lower than the agreed prices led to the adoption of new, adjusted target prices which the parties were to seek to achieve. From the documentary evidence in the form of meeting reports it is unmistakably clear what was meant by agreed or coordinated target prices. These, as the name itself indicates, expressed a price level that was to be sought to be achieved on the market, and were invariably higher prices than those which had been previously achieved. In the course of these proceedings we were shown studies prepared by Professor Albach which for part of the market in question sought to simulate the prices which might be assumed to have prevailed in the absence of agreements or concerted practices. The results of those studies were, as I have already mentioned, that the market would to all intents and purposes have behaved in the same manner without any agreements or concerted practices, and that the assessments of the participants in the meetings themselves of the results of their many endeavours were incorrect.

Quite aside from the methodological problems associated with such price simulations, and as the Commission pointed out at the hearing, Professor Albach expressly acknowledged, however, as the Court will recall, that the market may very well have been influenced to a certain extent.

In the light of those factors I do not think that there is any basis for reducing the fines on considerations relating to the cartel's actual effects, even though the Commission's view of the causal relationship may have been expressed a little too emphatically. In any event, in the light of the clear intention I think considerations about the actual effects of the cartel should be given relatively little weight in the calculation of the fines.

For the rest, I agree with the Commission that the factors enumerated in point 108 of the decision (with the exception of the last indent) all support the view that the infringement was a serious one.

C — Mitigating circumstances

In the last indent of point 108 the Commission states that it recognized the following as mitigating circumstances:

I do not think that enumeration of mitigating circumstances can be criticized.

In the applicants' opinion, however, the factors set out were not given nearly enough weight in the calculation of fines, and certain applicants even suggest that no fines should have been imposed, since the significant losses which were incurred may in their view be regarded as especially mitigating circumstances warranting the non-imposition of fines.

The applicants' point of view cannot be upheld. As the Court of Justice held in its judgment in IAZ v Commission, an obligation on the part of the Commission to take into account an undertaking's economic difficulties would be tantamount to conferring an unjustified competitive advantage on undertakings least well adapted to the conditions of the market.

When it is an entire industrial sector which is involved, to take excessive account of losses resulting from structural problems would in fact entail the de facto legalization of any attempt to prevent market forces from ensuring that the necessary structural adaptation was carried out. It is self-evident that any such result would be contrary to the market economic thinking on which Article 85 is based.

It is possible that there may in any event have been good intentions in seeking to maintain production capacity, jobs and so on for a certain period until demand increased in a sector which undoubtedly had a future. But in the system established by the Treaty departures from the fundamental rules of Article 85 are not a matter for private initiatives and the applicants' submissions in this respect must therefore be rejected.

It is also possible, as ICI in particular argued, that during the relevant period there was a degree of permanent or temporary structural adjustment in the polypropylene sector. However, as a matter of principle unlawful activities should be assessed as they stand, in isolation. The fact that the persons who committed the infringements also took lawful and undoubtedly reasonable measures is in no way unexpected or abnormal, and does not in my view constitute a reason to take a more lenient view of the infringements.

Finally, it is possible that State aid has distorted competitive relationships on the polypropylene market to some extent. As the Commission correctly pointed out, however, Article 85(1) is applicable in any event. It is striking, moreover, that virtually all the undertakings in the relevant sector were implicated in the infringements, both those that received State support and those that did not. We are thus not dealing with a cartel established by a number of undertakings with the purpose of defending themselves against unreasonable competition from State-supported undertakings. The applicants' submissions on this point must therefore be rejected.

In the last paragraph of point 109 of the decision the Commission states that a few of the undertakings cooperated with the Commission's investigations, at least once most of the incriminating evidence had been discovered. At the hearing we were told that the undertakings in question were mainly ICI and Hercules, which to a certain extent provided the Commission with material which could be useful in its conduct of the case. As the Commission observes, however, the undertakings were in any event under an obligation to provide the documents mentioned, and I therefore agree with the Commission that that provides a basis for only a small reduction in the fines. According to the Commission's explanations, the difference between the fine of ECU 10000000 imposed on ICI and that of ECU 11000000 imposed on Montedipe must be seen in the light of ICI's cooperation with the Commission.

I have already given my view of the effects of the cartel in relation to the gravity of the infringement. Finally, I agree with the Commission that even higher fines would have been called for if the cartel had in addition taken steps to impose actual measures of constraint on undertakings which did not comply with the agreements and such like which had been entered into.

In conclusion, my view is that the fines are not out of proportion, seen in relation to the very serious infringement in question, and that proper account was taken of the mitigating circumstances which can and should reasonably be taken into account.

D — Reduction of the fines as a result of the annulment of certain parts of the decision

As I have stated in Section B, the fines must be reduced if it is found that the cartel lasted for a shorter time than the Commission asserts. As regards the period thus determined (Autumn 1979 until Autumn 1983), it must be held that in so far as there is insufficient evidence for the Commission's findings in regard to certain points in the decision there should be a further reduction.

On the basis of what I have stated above, there is no sufficient evidentiary basis to confirm what the Commission asserts was an unlawful arrangement in 1981 and 1982 in relation to volume control, and there is not an entirely sufficient basis for the conclusion that there was a quota system in 1983. Furthermore, I do not think there is any basis for the finding that the undertakings unlawfully diverted goods from the territory of the Community in order to create an artificial shortage on the European market.

The annulment of the decision, in so far as it concerns the period from 1977 until Autumn 1979 as well as the aspects I have mentioned above regarding 1981, 1982 and 1983, should, I think, lead to a general reduction in the fines of about 20%.

E — Must the Commission draw up a catalogue of fines f

During this case there was a lively debate on the question to what extent the Commission should be obliged to give a detailed explanation of the criteria which are taken as a basis for the calculation of the individual fines.

The question is related to the series of factors enumerated by the Commission in the first paragraph of point 109 of the decision, which were decisive in the calculation of the individual fines. These factors are (1) the individual undertaking's role, (2) the duration of each undertaking's participation, (3) the sales of each undertaking in the Community and (4) the individual undertaking's total turnover. It is undisputed that those factors both can and must be taken into account.

Of the factors mentioned, the last two are entirely quantifiable, provided that it is possible throughout to obtain reliable figures for polypropylene sales and total turnover. The duration is also quantifiable, but as I have already said it cannot simply be viewed proportionally, since the cartel's activities took increasingly serious forms. The extent or intensity of an individual undertaking's participation, on the other hand, cannot be quantified so as to provide a basis for the mathematical calculation of fines, but an estimate of it must enter into the fixing of fines.

It is self-evident that it is extremely difficult to explain in any reasonable manner the relative weight to be given to different factors only some of which are entirely quantifiable. The decision must to a large extent remain a matter of discretion. Indeed, later on, in its answer to the written questions put by the Court and at the hearing, the Commission denied that in assessing the fines it should have carried out precise calculations on the basis of the quantifiable factors.

I do not think there is any basis for believing that the Commission did not give a correct explanation of how it arrived at the amount of the individual fines. The Commission insisted that the decision was adopted after an overall assessment. As I have indicated, I do not think that the Commission can be criticized for that, and in my view there is no basis in case-law or elsewhere for the conclusion that the Commission should be obliged to draw up specific calculation models.

In addition, as the Commission has also observed, from the point of view of general deterrence it may be dangerous to draw up an actual catalogue of fines in a legal area such as the one in issue here, where considerations of the economic benefits and possible drawbacks of a contemplated infringement of the law are clearly an important factor in an undertaking's decision whether or not to go ahead with an infringement.

Those points are also applicable in relation to the statement in Section B. II of the decision of the reasons for the fixing of the fines. Most of the applicants have argued that the decision did not contain an adequate statement of reasons since there was no individual statement of the reasons for the fine imposed on each undertaking. In my view, the Commission provided the reasoning which in the nature of the case it was possible to give. Where it is not possible to quantify all the factors entering into the calculation of fines, it is difficult to see what further elements could be added to those mentioned by the Commission.

The above considerations ultimately have repercussions on the assessment of the Court's review function with regard to the calculation of fines. As was made clear in particular at the hearing, in the absence of any precise knowledge of the weight given by the Commission to the various factors in relation to each other, it can be difficult to carry out a review of the calculation of fines. But if it must be acknowledged, as I think it must, that in this area the Commission should have a broad discretion to determine fines according to its assessment of all the circumstances of the case, that problem is no different from that of any other area of the law in which the administration has a more or less broad discretion.

In Community law, however, the difference lies in the fact that the Court has unlimited jurisdiction with regard to the calculation of fines and thus can, if it wishes, substitute its own assessment for that of the Commission. What takes place is thus in reality an independent assessment on the part of the Court. In my view, however, the Court should display some caution in this regard and only step in when it comes to the conclusion that the Commission has been guilty of a clear error of assessment. That might for example be thought to be the case where it appeared that the Commission had taken incorrect turnover or sales figures as a basis for its assessment.

Finally, we may reject the view that was put forward to the effect that a principle of equal treatment was infringed by the fact that the Commission imposed no fines on BP and AMOCO, even though certain applicants took the view that they were involved in the cartel's activities. I think that point of view is based on a fundamental misconception and confuses the issues of proof and penalty. Once it is established that BP and AMOCO could not be brought into the case, for lack of evidence, it is entirely out of the question that the conduct of those undertakings should play any role in the assessment of the fines imposed on the other undertakings.

A last factor which should perhaps be mentioned in this section is that, as the Commission explained in more detail at the hearing, the reference in the second paragraph of point 107 of the decision to the fact that BASF, Hoechst and ICI had previously been involved in infringements of the Community competition rules does not mean higher fines were imposed on grounds of recidivism. As the Commission explained at the hearing, the purpose of the references was simply to provide further evidence that in respect of those undertakings the infringements were deliberate. The Commission explained that the previous offences did not result in higher fines because they had been committed more than 15 years previously, and because the general level of fines had increased significantly in the meantime. There is no reason to reject that explanation; nor, in my view, can the fact that the fines were not increased in that respect be held to constitute a disregard of a principle of equal treatment or any other principle.

F — Extent or intensity of the individual undertaking's participation

1. Extent as stated in the Commissions decision

In my view, there can be no reasonable doubt that the Commission was correct to conclude that the four largest producers, Montedipe, Hoechst, ICI and Shell, constituted the nucleus of the arrangements which were introduced and that they formed a separate leadership group. That is shown by the subject-matter of the separate meetings held by those undertakings (see table 5 of the decision) and the further indications of the four large undertakings' role to be found in the evidence. The Commission is also therefore correct to conclude that the four large producers must bear a large part of the responsibility for what took place. Even if it cannot be said that there is proof that they were in fact the instigators, the four large undertakings clearly — I could almost say naturally — played a central role. That is true in particular of Montedipe and ICI, each of which took on the leadership of the group for a certain period. That provides substantial grounds for imposing severe fines on those undertakings.

It is not apparent from the case why Shell did not take part in the regular plenary meetings. I agree with the Commission, however, that the fact that it did not take part in meetings with all the producers cannot in itself be regarded as a mitigating circumstance. Shell's participation must therefore be assessed on the same footing as that of the other undertakings in so far as there is an evidentiary basis for its involvement. That is described in more detail above in point II J.

As described above in point II F, the role of Hercules seems to have been surrounded by a degree of ambiguity. I agree with the Commission, however, that the fact that Hercules did not provide the other undertakings with information on its own sales figures cannot be regarded as a mitigating circumstance, since in other respects that undertaking participated in the cartel and benefited or sought to benefit from it.

According to the sixth paragraph of point 109 of the decision, the Commission does not accept that any substantial distinction can be made between the other undertakings, that is to say other than the four large producers. I agree that there is no basis in the evidence for any confident assessment as to which undertakings were more or less persistent in ensuring effective cooperation. The general impression is that all of them were clearly interested, and that the main differences lie in which of them were most optimistic. As mentioned above, at a meeting in May 1982 Solvay made the reasonable suggestion that the meetings should be discontinued because supply and demand were in balance. That could be seen as reflecting a somewhat lesser degree of commitment, were it not for the fact that Solvay continued to participate in the group after that time.

Some of the minor producers were described by the others on several occasions as disruptive and aggressive. In this context, that can be taken to mean that they displayed somewhat less commitment than the others. It is, however, to be noted that in any event they sought to further their interests within the framework of the cartel. The fact that some producers managed to win themselves a not insignificant market share during the period of the cartel's existence can therefore be taken not as meaning that their commitment was less than that of the others but rather that they were adept at using the cartel for their own purposes, that is to say to achieve a better market position in a well ordered manner.

Finally, the Commission states in the seventh paragraph of point 109 of its decision that three undertakings took part for a shorter period than others. It should be observed here that, according to what I have been able to conclude from the evidence, Petrofina took part only from March 1982 onwards, and not from 1980, as the Commission alleged.

The undertakings can therefore be divided into two groups, the four large producers and the others.

2. The four large producers

Among the four large producers it is apparent from the decision that the Commission imposed larger fines on Montedipe and ICI because of the leading role that each of those undertakings played for a certain period, and that ICI's fine, as has been explained, was reduced because of the extent to which it cooperated in the investigation. The fines on both undertakings appear to have been assessed more or less uniformly in relation to their turnover in the polypropylene sector. It should be observed that the Commission did not take into account the objections to the effect that only the undertakings' external sales of polypropylene should be considered. The Commission seems instead to have relied on turnover figures corresponding to what the undertakings themselves mutually regarded as their market share. In my view, that is a good criterion for calculating fines. Leaving aside the fact that the undertakings total turnover was not discussed in the case, there is little to indicate that the Commission laid any particular weight on that in calculating the fines, which is hardly surprising since in any event these are very large undertakings whose strength on the market and whose ability to pay a fine of the magnitude in issue is beyond doubt. There thus seems to be a good balance between the fines of ECU 11 and 10 million imposed respectively on Montedipe and ICI. The same is true of the fines of ECU 9000000 imposed on Shell and Hoechst, which were clearly considered to have formed part of the nucleus of the group without, however, having played the leading role of Montedipe or ICI.

3. The minor producers

The minor producers can again be divided into the producers that took part in the cartel for the whole period of its existence and those that joined or left the cartel in the course of that period.

With regard to the first group, the level of fines seems again to be fairly constant in relation to their sales in the Community, which a priori it should do, moreover, where no distinction was made according to the intensity or extent of their participation and, as it appears, no decisive weight was placed on their total turnover. As a percentage of sales the fines were generally somewhat lower than in respect of the large producers. With regard to BASF, the parties disagree as to how far the sales figures for the production from Rheinische Olefinwerke GmbH, a 50/50 joint venture between BASF and Shell, should be included in the basis for calculation of the fine. BASF asserts that it acted only as sales agent for the part of the production attributable to Shell. The question is entirely unresolved and should have been investigated more closely by the Commission at an early stage in the procedure. The result, I think, must be that be that greater weight should be placed on BASF's own figures and an appropriate reduction made in the fine imposed on that undertaking. That reduction should probably be about 10%, so that the fine on that undertaking should be reduced by a total of 30%.

Somewhat lower fines were imposed on the last group of undertakings, that is to say ANIC, Petrofina and Rhône-Poulenc, as is stated in the decision. Leaving aside the fact that the fine must be further reduced in respect of Petrofina as a result of the shorter period for which its participation can in my view be considered to be proved, here too there seems to be a good balance both between the fines imposed on these undertakings and in relation to the fines imposed on the others. The fine imposed on Rhône-Poulenc should be reduced on the basis, first, of the fact that the period in relation to which there is evidence proving its participation is significantly shorter than the Commission assumed, and, secondly, of the fact that the period which must thus be disregarded was characterized by the fact that the cartel was in an introductory phase. Rhône-Poulenc's fine must therefore be reduced by a total of 40%. Petrofina's participation was of significantly shorter duration than is asserted by the Commission, and part of the period which should not be taken into account was after the cartel had taken on its definitive form. On that basis I propose that Petrofina's fine should be reduced by half. A particular problem is raised by the question to what extent the Commission included SIR's turnover in calculating the fine for ANIC. According to what we have been told, the basis of calculation was the turnover figures (equals market share) for 1982. There is no doubt, however, that the market share held by SIR in 1982 before it produced — for a short period — for ANIC's account was very small. Even if there are no grounds for giving greater weight to ANIC's view that the previous year's turnover figure should also be taken into account, I think that account should be taken of the fact that ANIC took part in the cartel only in the first half of 1982 and that the fine should therefore be reduced to a certain lesser extent. ANIC's fine should accordingly be reduced by a total of 30%.

Subject to what must follow from the foregoing, it is therefore my view that the Commission is not guilty of any error of assessment.

On that basis I therefore propose that the fines should be fixed as follows:

IV — Conclusion

In the light of all the foregoing I propose that the cases be decided as follows :

1 Original language: Danish.

2 See Baden Fuller in European Law Review 1979, p. 439.

3 SA Musique Diffusion Française and Others v Commission [1983] ECR 1825, at 1930.

4 Judgment of 14 July 1981 in Case 172/80 Gerhard Zūchner v Bayerische Vereimhank AG [1981] ECR 2021 (paragraph 14, at p. 2031).

5 Council Regulation No 17/62/EEC of 6 February 1962: First Regulation implementing Articles 85 and 86 of the Treaty (OJ, English Special Edition 1959-1962, p. 87). * In this caie the names Montedison, Montepolimeri and Montedipe, as well as the abbreviation Monte, appear inter alia. According to the documents in the case, until the end of 1980 the parent company of the Montedison Group, Montedison S. p. A., itself undertook polypropylene production but from 1 January 1981 production was transferred to the wholly-owned subsidiary Montepolimeri S. p. A. to which the Statement of Objections was addressed. As is apparent from the fourth paragraph of point 8 of the decision, in conjunction with Articles 3 and 5 thereof, an internal reorganization of the Montedison Group took place, which resulted in the decision being addressed to Montedipe S. p. A., which is the applicant in Case T-14/89. In this Opinion the undertaking will be referred to as Montedipe or Monte.

6 According to the note drawn up by Shell's representatives after the meeting, Annex 16 to Shell's application (Case T-11/89).

7 Judgment of 25 October 1983 in Case 107/82 AEG v Commiision [1983] ECR 3151.

8 Judgment of 29 October 1980 in Joined Cases 209 to 215/78 van Landewyck and Others v Commission (Fedelalf) [1980] ECR 3125 (paragraphs 79-81, p. 3248), and judgment of 7 June 1983 in Joined Cases 100-103/80 SA Musique Diffusion Française and Others v Commission (the Pioneer cases) [1983] ECR 1825 (paragraphs 6 to 8, p. 1880).

9 Case No 13258/87, M & Co. v Federal Republic of Germany.

10 See footnote 7.

11 Judgment of 21 February 1984.

12 See also Piiakos, Les Droits de la Défense et le Droit Communautaire de la Concurrence, Bruylant, Brussels 1987, p. 145 et seq.

13 On this point, see the judgment of the Court of 9 November 1983 in Case 322/81 NV Nederlandsche Banden-Industrie Michelin v Commission [1983] ECR 3461, at paragraphs 5 to 10, pp. 3498, 3499.

14 See footnote 7.

15 [1986] ECR 1899.

16 Judgment of 17 January 1984 in Joined Cases 43 and 63/82 VBVB and VBBB v Commission [(1984] ECR 19 (paragraphs 23 to 25, at p. 59)).

17 Michelin v Commission, cited above, footnote 12.

18 See the judgment of 25 October 1983 in Case 107/82 AEG v Commission [1983] ECR 3193.

19 See Schwarze, Europäisches Verwaltungsrecht II, p. 1294, note 77, according to which undertakings in cartel cases are at all events entitled on request to production of a copy of the relevant documents.

20 On this point, see the attitude taken by the Court of Justice in AEC v Commission, cited above at footnote 17.

21 On this point, see K. P. E. Lasok, The European Court of Justice, Practice and Procedure, p. 260, footnote 4.

22 Judgment of 15 July 1970 in Case 41/69 ACF Chemiefarma v Commission [1970] ECR 661, paragraphs 91-93, p. 691 et seq.; and judgment of 29 October 1980 in Joined Cases 209 to 215 and 218/78 van Landewyck v Commission [1980] ECR 3125, paragraphs 67-74, p. 3244.

23 The German text refers to fortgesetzte Zusammenarbeit, which can mean constant cooperation but which naturally conveys the idea of fortgesetzte Handlung in the criminal law sense.

24 Cited above in footnote 21.

25 Cited above in foocnote 7.

26 Judgment of 10 July 1980 in Case 30/78 [1980] ECR 2229.

27 Distillers v Commission, cited in footnote 25; see in particular [1980] ECR 2294.

28 Judgment in Case 44/69 [1970] ECR 733, at p. 753.

29 Judgment of 14 July 1972 in Case 48/69 [1972] ECR 619, paragraphs 27 to 32, at p. 651.

30 Sec Schwane, p. 1349 and note 84.

31 See the judgment in Chemiefarma cited in footnote 21 [(1970] ECR 6*1, parigraphs 76 and 77, at p. 689).

32 Judgment of 15 July 1960 in Joined Cases 36, 37, 38 and 40/59, Präsident Ruhrkohlen-Verkaufsgesellschaft GmbH and Others v High Authority of the ECSC [1960] ECR 423.

33 Judgment of 1 July 1986 in Case 185/85 Union Sidérurgique du Nord et de l'Est de la France (Usinor) SA v Commission [1986] ECR 2079, paragraph 20 at p. 2098.

34 Judgment of 15 October 1987 in Case 222/86, UNECTEF v Heylens, [1987] ECR 4112.

35 Sec U. Everling in Wirtschaft und Wettbewerb, 1989, p. 877.

36 See OJ 1989 C 215 of 21 Ausust 1989, p. 1, containing the Council Decision of 24 October 1988 establishing a Court of First Instance, as published in OJ L 319 of 25 November 1988 and amended by the corrigtndum published in OJ L 241 of 17 August 1989.

37 judgment of 13 July 1966 in joined Cases 56 and 58/64 Etablissements Consten S.à r.l. and Grundig-Verkau/s-GmbH v Commission [1966] ECR 299. The same approach was taken later, see for instance the judgment of 21 February in Case 6/72 Continental Can [1973] ECR 215 (at paragraph 6, p. 240) and VBVB and VBBB cited above in footnote 16.

38 See footnote 21.

39 See footnote 7.

40 Judgment of 2 March 1983 in Case 7/82 Gesellschaft zur Verwertung von Leïstungsschutzrechten GmbH (GVL) v Commission [1983] ECR 483.

41 Judgment of 21 February 1984 in Case 86/82 Hasstlblad (GB) Limited v Commission [1984] ECR 883 (at p. 915, right-hand column).

42 Judgment of 16 December 1975 in Joined Cases 40/73 et al. Coöperatieve Vereniging Suiker Unie UA and Others v Commmion [1975] ECR 1663 (paragraph 111, at p. 1921).

43 See footnote 7 (paragraph 32, p. 3236).

44 OJ 1989 L 74, pp. 1 and 21 (Decisions of 21 December 1988, IV/31.865, PVC, and IV/31.866, LdPE).

45 See most recently, Rutsch, Strafrechtlicher Durchgriff hei verbundenen Unternehmen?, Cologne 1987, and Lipowsky, Die Zurechnung von Wetthewerbiventößen, Munien 1987.

46 Cited above, in footnote 41.

47 Judgment of 28 March 1984 in Joined Cases 29 and 30/83 Compagnie Royale Asturienne des Mines S.A. and Rhetnzmk GmbH v Commission [1984] ECR 1679.

48 See lhe Chemie/arma case, cited above in footnote 21, paragraphs 106-116.

49 See the Sandoz case.

50 Cited above in footnote 36.

51 See Hildebrandt: Der Irrtum im Bußgeldrecht der Europäischen Gemeinschaften, 1990, p. 40; Dannecker/Fischer-Frilsch, Das EG-Kartellrecht in der Bußgeldpraxis, p. 15; see also Grabitz, Kommentar zum EWG-Vertrag, on Anide 85, No 28.

52 See, in particular the judgment in Consten and Grundig, cited above in footnote 36.

53 See footnote 41.

54 See footnote 3.

55 Sec, for example, Mr Advocate General Mayras, [1972] ECR 666, at p. 669.

56 See U. S. v Krnel (U. S. Supr. Court 1910) 173 Fed. 823, 218 US 601, Multiflex, Inc. v Samuéi Moore 6 Co, and Eaton Corp. (5th Circ. 1983), 1983-2 Trade Cases, 65, 507, and American Tobacco Co. et al. v U. S. (CCA-6 1944), 1944-45 Trade Cases, 57, 317 (p. 57, 587).

57 Cahien de Droit Européen 1974, p. 258.

58 See Joseph Ellis: Source Material for Article 85(1) of the EEC Treaty in Fordham Law Review, Vol. XXXII 1963, No 2, p. 247-278.

59 Judgment of 14 July 1972 in Cise 48/69 Imperial Chemical Industriei Ltd v Commission [1972] ECR 619, and judgments of the same date in Cases 49/69, 51/69, 52/69, 53/69, 54/69, 55/69, 56/69 and 57/69.

60 See Piriou in Cahiers de Droit Européen, 1973, p. 52, and Joliét, op. cit. p. 266, and the further references contained in both articles.

61 See footnote 41 above.

62 Judgment of 14 July 1981 in Case 172/80 [1981] ECR 2021, cited in footnote 3 above.

63 Judgment of 15 July in Case 41/69 ACF Chemiejarma v Commission [1970] ECR 661, riled above in footnote 21.

64 At p. 714, right-hand column.

65 In its Judgment the Court of Justice did not deal with the distinction between an agreement and a concerted practice.

66 See footnote 28 [1972] ECR 619.

67 Sec footnote 7.

68 Sec footnote 7, [1983] ECR 1825, at p. 1929.

69 Judgment of 11 January 1990 in Case C-277/87, published in summary form in [1990] ECR I-45.

70 See footnote 56.

71 La Politique de la Concurrence dans la CEĶ ]977, a publication of l'Institut Universitaire International, Luxembourg, Editions UGA.

72 Droit Européen dei Affaires, Themis, Droit 1990, p. 278, Presses Universitaires de France.

73 Journal du Droit International, 1973, p. 925 (p. 938).

74 Cahiers de Droit Européen, 1973, p. 50.

75 Revue du Marché Commun, 1973, p. 17.

76 Kartelrecht, 1986.

77 Handbuch des Europäischen Rechts, Article 85, Nos 17 and 18, p. 106 et seq.

78 Kommentar zum EWG-Vertrag, Article 85, No 28, p. 10.

79 Common Market Law of Competition, Third Edition 1987, para. 2-040, p. 60.

80 Clunet, 1977, p. 219.

81 Droit Matériel et Politique de la Communauté Européenne, 1986, p. 163.

82 Das Wettbewerbsrecht der Europäischen Wirtschaftsgemeinschaft, Article 85, No 23, p. 799.

83 See section (d) above.

84 See Hans-Dieter Lübben: Das Verbot, abgestimmten Verbaltem im deutschen und europäischen Kartellrecht, p. 90 and note 51.

85 See Schröter in Croeben/Botch, op. cit. no 13, p. 103.

86 On this point, sec the judgment delivered by the Court of Justice on 29 February 1968 in Case 24/67 Parke Davis & Co v ProbeĻ Reese, Beintema-Interpharm and Centrapbarm [1968] ECR 55.

87 [1972] ECRaip. 687.

88 Judgment of 30 June 1966 in Case 56/65 Société Technique Minière v Maschinenbau Ulm GmbH [1966] ECR 235.

89 Judgment of 2 July 1969 in Case 5/69 fram Võik v S.p.r.L Ets J. Vervaecke [1969] ECR 295.

90 See, for example, K. P. E. Lasok, The European Court of Justice, Practice and Procedure, 1984, p. 263.

91 See above, footnote 41.

92 See section A, 5 c.

93 See the US case Hunt v Mobil Oil Corp. (Supreme Court, 1977) 465 F. Supp. 195, 231, according to which it is not sufficient in order for unlawful conduct to be regarded as proven for an undertaking to have been present at the place where the unlawful act was committed and possibly even to have had knowledge of others' unlawful conduct.

94 Note in this regard that Van Damme distances himself on this point from Joliet's criticism of the ICI judgment: La Politique de la Concurrence dans la CEĶ p. 150.

95 Judgment of 3 July 1985 m Cast 243/83 SA Binon and Cie v SA Agence el Menageries de la Prene [1985] ECR 2015.

96 Judgment of 25 June 1976 in Case 51/75 EMI Records Limitedv CBS United Kingdom Limited [1976] ECR 811.

97 Regulation (EEC) No 2988/74 of the Council of 26 November 1974 concerning limitation periods in proceedings and the enforcement of sanctions under the rules of the European Economic Community relating to transport and competition, OJ 1974 L 319, p. 1.

98 Sec footnote 92; see also the Opinion of Advocate General Sir Gordon Slynn in SA Musique Difįuuon Française v Commission (cited in footnote 2).

99 Case 101/80 C. Melchers & Co. v Commission, one of the Pioneer cases, cited in footnote 7.

100 Common Market Law of Competition, see footnote 78.

101 Decision of 14 December 1979, OJ 1980 L 60, p. 21.

102 See footnote 7.

103 Judgment of 12 July 1979 in Joined Cases 32/78 and 36-82/78 BMW Belgium v Commission [1979] ECR 2435 at p. 2494.

104 See footnote 43.

105 Commission decision of 19 December 1990 in Cases IV/33.133-A: Soda-ash — Solvay, ICI; IV/33.133-B: Soda-ash — Solvay, CFK; IV/33.133-C: Soda-ash—Solvay; and IV/33.133-D: Soda-ash —ICI (OJ 1991 L 152 pp. 1, 16, 21 and 40).

106 See for example, Mr Advocate General Warner in the BM Uncase, cited in footnote 102, at page 2493.

107 See the judgment of 10 December 1985 in Joined Cases 240-242, 261, 262, 268 and 269/82 Stichting Siįarettenin-äuitrie v Commission [19851 ECR 3831, at paragraph 82, p. 3881.

108 Judgment of 8 November 1983 in Joined Cases 96-102, 104, 105, 108 and 110/82 IAZ International Belgium v Commission (ANSEAUNAVEWA) [1983] ECR 3369 at paragraph 55, p. 3417.

109 See Bellamy and Child, op. cit., p. 497, text to note 56.