The market for wood pulp

The product

The producers

The customers

Announced prices and transaction prices

Trends on the pulp market

Procedure and decision

I — Procedural submissions

A — The decision embodies complaints which are not referred to in the Statement of Objections

(1) The Statement of Objections did not refer to concertation on transaction prices

(2) Duration of concertation

(a) Duration of the infringement until the date of the Statement of Objections

(b) Duration of the infringement established for certain producers

(3) Identity of the producers involved in the concerted practice

(4) St Anne's participation in Fides

B — The evidence on which the decision is based

(1) Transaction prices

(2) The documents relating to Fides

C — The refusal to organize a joint administrative hearing

D — The failure to consult the Advisory Committee on Restrictive Practices and Monopolies

Conclusion

II — ‘General’ concertation on announced prices

A — The concept of concerted practices

B — Evidence of a concerted practice on announced prices for wood pulp

(1) Parallel conduct of the producers

1.1. Parallelism of announced prices

1.2. The system of price announcements

1.2.1. The system of price announcements created artificial transparency on the pulp market and constituted a system for exchanging information between producers

(a) The role of customers in the dissemination of information on announced prices

a.1 Integrated producers

a.2 The role of important customers

a.3 Common customers

a.4 Horizontal communication between purchasers

(b) Common agents

(c) The trade press

1.2.2. The system of quarterly announcements

(a) The quarterly announcements mechanism is said not to be necessitated by objective market conditions

a.1 Quarterly basis of price announcements

a.2 The use of the US dollar by non-United States producers

(b) Simultaneous price announcements

1.3. Parallelism of announced prices: evidence of concertation on the pulp market?

1.3.1. The Commission's analysis

1.3.2. The expert's report ordered by the Court

(a) The expert's report is said to oversimplify the Commission's reasoning

(b) The economic model used by the experts

(c) The alleged failure to examine the level of prices

(d) The period from 1975 to 1977

(e) The different aspects of price uniformity

e.1 Uniform prices for the same grade of pulp

e.2 Uniform prices for different qualities of pulp

e.3 Uniform prices for different customers buying differing quantities

e.4 Uniform prices in US dollars for customers in different countries

(f) Concertation or market conditions?

f.1 Undertakings that were not involved in the procedure

f.2 Changes in market shares

(g) The pulp market before 1975 and after 1981

g.1 The market before 1975

g.2 The market after 1981

Conclusion

1.4. The statement of reasons in the decision

(2) Direct and indirect exchange of information

III — Concertation within KEA

IV — Concertation within Fides

V — The clauses prohibiting export and resale

VI — Effect on trade between Member States

VII — The alleagation of discrimination

VIII — The undertaking

IX — The fines

(1) Discrimination

(2) The amount of the fines

X — Costs

1. 

Last but not least... For several reasons, that is the expression which most readily comes to mind when delivering my Opinion in this, the ‘wood pulp’ case. First of all, because it is the last competition case brought by undertakings to be decided by the Court at first and last instance: the criterion adopted for the allocation of cases pending before the Court of Justice at the time of the establishment of the Court of First Instance meant, in view of the stage of the proceedings reached at the time, that this case would be heard by the Court of Justice. Secondly, because this Opinion and the Court's judgment will constitute the last stage of a long drawn-out procedure whose salient features were, in particular, a hearing in January 1988, an initial judgment of the full Court on 27 September 1988 rejecting the submission relating to the extraterritorial scope of Community competition law, the hearing in 1990 of the first ‘accountancy’ experts followed, in November 1991, by the hearing at which every aspect raised in the applications was considered and in which the ‘economics’ experts were heard. However, neither this case nor this last stage are by any means of little account. The quite vast dimensions of this case, in terms of both the complexity and the sheer volume of the arguments put forward, as well as the number of applicants, make it anything but ordinary.

2. 

Inevitably, I have had to resign myself, in order to take account of manifold objective constraints and of the concern to keep this Opinion within reasonable limits, to neglecting certain aspects of those arguments where the reasons for the solution which I shall propose to the Court would seem not to be affected thereby.

3. 

Before I proceed to examine a case which has itself helped to stimulate demand for wood pulp in the Community as a result of the sheer volume of documents produced, it is necessary to recall the characteristics of the market for wood pulp, followed by the main features of the procedure and of the contested decision.

4.

The contested decision is concerned with bleached sulphate pulp used in paper manufacturing and offered on the open market (‘market pulp’). There are other processes for the production of pulp (chemical or mechanical) but sulphate pulp is best in quality. It is obtained by the chemical processing of cellulose in a two-stage process. To begin with, the first stage leads to the production of unbleached (brown) sulphate pulp which is used in the manufacture of packaging material (bags, boxes etc). Next, in order to be used in the manufacture of finer papers, (newsprint, writing paper) requiring, in particular, a high degree of brightness, the pulp is bleached chemically. That process has the advantage of consuming less energy and causing less pollution than other processes. In addition, practically all types of wood can be used in the manufacture of bleached sulphate pulp. Consumption of such pulp has grown in recent years and, according to the decision, accounted for approximately one-third of the market in 1980.

5.

There are several types of bleached sulphate pulp whose classification depends on the type of tree from which it comes. The main distinction is between softwood pulp and hardwood pulp. The two categories have different properties: softwood pulp has longer and more resistant fibres. Hardwood pulp, which has shorter fibres, is used for the manufacture of soft paper, which is not so strong. Softwood pulp is considered to be of better quality than hardwood pulp. Within each category, moreover, there is a difference according to whether the pulp was produced in northern or southern latitudes. Generally speaking, hardwood and softwood from northern latitudes are considered better in quality than those from southern latitudes. Since 1978 there have been four price levels for those products, the highest prices being paid for- northern softwood and the lowest for southern hardwood.

6.

In view of the properties of the different categories of pulp, paper manufacturers generally use a mixture of types in order to obtain paper of the specific grade required.

7.

Within any pulp category, pulp is very largely interchangeable. In contrast, there is less interchangeability between the different categories in the short term. A switch in category is likely to be very costly for the paper manufacturer; it is a relatively long process requiring trials and possibly even conversion of equipment.

8.

Bleached sulphate pulp is produced by some 800 producers scattered in over 30 countries. A major part of the pulp is used by the producers (or their subsidiaries). The pulp which they sell on the market is known as ‘market’ pulp, it should be recalled, and it is this which forms the subject-matter of the contested decision. According to the decision, total output of bleached market pulp amounted to approximately 18 million tonnes in 1981. Canada was the leading producer (with more than 6 million tonnes), followed by the United States (with more than 4 million tonnes), Sweden (with 2.5 million tonnes) and Finland (with 1.6 million tonnes).

9.

According to the decision, there are more than 50 firms which sell market pulp in the Community. For their part, some of the applicants emphasize that pulp is sold in the Community by 115 firms and that it comes from 18 countries.

10.

As far as total turnover is concerned, the United States and Canadian producers are larger than the Swedish and Finnish producers, who are the traditional suppliers of the European market to which they sell twothirds of their output. The market was penetrated by Canadian and United States producers in the 1950s. Exports from the United States grew substantially in the 1970s. The Community market does not have the same importance for the North American producers as it has for the Swedish and Finnish producers. The main outlet for Canadian and United States producers is the United States domestic market. From their point of view, therefore, the Community market is used primarily to offset cyclical fluctuations on the United States market.

11.

The tendency towards vertical integration is particularly marked in the pulp industry: many pulp producers also manufacture paper or hold sizeable shares in paper mills. ( 1 )

12.

The cost structure ( 2 ) varies according to the areas in which production is located. Generally speaking, wood costs are substantially higher in Finland and Sweden than in the United States and Canada. However, transport costs are said to be higher for pulp from the latter countries than they are for pulp from Scandinavia. Furthermore, the Commission and some of the applicants consider that fixed costs are high whereas the second group of experts, relying on the figures set out in the decision, ( 3 ) have instead concluded that the percentage of those costs is relatively small in relation to variable costs.

13.

The construction of a new pulp mill requires very large capital investments and a long period for its completion. Pulp mills are installed in the vicinity of forestry resources.

14.

The decision emphasizes that between 1975 and 1981 the capacity utilization ratio and level of stocks differed widely not only betwen different countries but also between the producers of one country.

15.

The Finnish producers are members of Finncell, ( 4 ) an association founded in 1918 whose object is to sell in its own name the wood pulp manufactured by its members. Finncell fixes the export prices and divides orders amongst its members.

16.

During the period covered by the decision, all the United States applicants, ( 5 ) with the exception of Bowater, were members of the Pulp, Paper and Paperboard Export Association, commonly known as the ‘KEA’, corresponding to the initials of its former name (Kraft Export Association). It was established under the Webb Pomerené Act, a United States statute which permits the formation of associations for the promotion of exports without infringing United States antitrust legislation. The Act authorizes producers, in particular, to exchange information on the marketing of their products abroad and to agree on export prices.

17.

As a rule, a single producer supplies some 50 customers in the Community. Finncell is said to have the largest number of customers (290).

18.

The producers market the pulp in the Community through their subsidiaries, sales offices, branches or agents. Some agents represent several producers.

19.

In the first place, it should be noted that the price of the pulp accounts for approximately 50% to 70% of the cost of paper. According to the decision, the Community market is the most important market for bleached sulphate pulp and in 1981 it accounted for one-third of global sales. In the same year Community pulp production amounted to approximately 700000 tonnes, whilst total sales in the Community amounted to some 6 million tonnes. The main purchasers were Germany (approximately 2 million tonnes), France (approximately 1.3 million tonnes) and the United Kingdom (approximately 1.1 million tonnes).

20.

Bleached sulphate pulp is bought by more than 800 paper manufacturers operating in the Community. According to the figures given by some of the applicants, ( 6 ) which have not been contradicted by the Commission, in the mid 1970's just 17 purchasers accounted for 50% of pulp purchases and 41 of them accounted for 80%. Purchasers of pulp are often linked with pulp producers whose headquarters are located outside the Community.

21.

The majority of pulp sales are made on the basis of long-term supply contracts which are concluded for periods ranging from three to five years and frequently contain an automatic renewal clause (known as ‘evergreen renewal provisions’). Those contracts generally provide that the seller is to reserve a particular tonnage of pulp for the purchaser on a quarterly basis at a price which is not to exceed the price announced in the period prior to the commencement of the quarter. It is common ground that the customer does not necessarily purchase the entire quantity reserved for him.

22.

The pulp sold on the spot market accounts for a negligible proportion of consumption.

23.

In order to obtain the desired mixture and to ensure availability of supplies, purchasers place their orders with several producers, who may be located in different areas. Furthermore, purchasers of pulp linked to a producer are led to purchase certain types of pulp from other suppliers.

24.

According to a well-established practice, a few weeks or, at times, a few days before the beginning of a quarter the producers communicate the prices which they wish to obtain for the types of pulp marketed by them. Those prices, which include the cost of transportation to European ports, are generally fixed in different ways according to whether they are for deliveries to ports in North-West Europe (zone 1) or to Mediterranean ports (zone 2). Those announcements are made to the producers' customers, potential customers, agents and sales offices. For the entire period covered by the decision after 1976, those prices have been quoted in United States dollars by all the producers concerned. The prices are published forthwith in the specialist press.

25.

Announced prices are to be distinguished from transaction prices, that is to say the prices actually invoiced to customers. Transaction prices may be identical to announced prices or may differ (where rebates, payment concessions or other discounts are granted).

26.

The pulp market is largely dependent on demand for paper and board, which in turn is influenced by changes in the general economic situation.

27.

The table set out below ( 7 ) shows changing patterns of consumption and importation of bleached sulphate pulp in the Community.

TABLE 2

Consumption and importation of bleached sulphate pulp in the EEC

(1 000 tonnes)

28.

The very strong demand for paper pulp in 1973 — when there were shortages — began to fall at the end of 1974 and slumped in 1975. During that year the capacity utilization ratios of the United States, Swedish and Finnish producers declined significantly. The same was true in the case of those Canadian producers whose production was affected by strikes. Still in 1975, the Swedish Government introduced a stock incentive scheme as a result of which Swedish producers' stocks grew to colossal proportions. In 1976 demand for pulp in the Community recovered to almost the 1974 level.

29.

During 1977 consumption stagnated (+0.7%) and there was a very slight decline in imports (-1.5%). In 1978 consumption and imports began to rise appreciably, a trend which continued throughout 1979 and peaked in 1980. In 1981 there was a slight fall in demand which continued into 1982.

30.

According to the decision, the trend with regard to announced prices on the Community market may be summarized as follows: 1975 ( 8 ) and 1976 were characterized by complete stability until the third quarter of 1977 ( 9 ) which witnessed the beginnings of a decline that became more pronounced in the following quarter and the first two quarters of 1978. That was when announced prices fell to their lowest level during the period covered by the decision. In the third quarter of 1978 announced prices recovered slightly, and rose steadily in the fourth quarter. Subsequently, in each of the following quarters until the second quarter of 1980, announced prices continued to increase. They then remained stable until a further increase in the third and fourth quarters of 1981, which, according to the Commission, was not ‘accepted’ in all the Member States.

31.

It should be pointed out at this juncture that the Commission considers that during the period in question the producers almost simultaneously announced similar prices for similar periods and invoiced identical transaction prices, a finding which, in its view, can be explained only by the existence of a concerted practice. However, the Commission states that in 1977 and 1978 transaction prices differed significantly from announced prices, since according to the decision, weak demand led to greater competitive pressure which made it impossible for the undertakings to charge unrealistic concerted prices.

32.

It should be noted that in April 1978 the Commission initiated an antidumping procedure concerning imports of pulp originating in Canada, the United States, Finland and Sweden, ( 10 ) which was terminated at the end of the same year. ( 11 )

33.

In 1977, after carrying out investigations under Article 14 of Regulation No 17, ( 12 ) the Commission stated that it had discovered the existence in the wood pulp industry of a number of restrictive practices and agreements which had not been notified under Articles 4 and 5 of that regulation.

34.

On completion of those investigations, on 29 July 1981 the Commission commenced on its own initiative a proceeding under Article 3(1) of Regulation No 17 against 57 producers or associations established in the United States, Canada, Finland, Norway, Sweden, the United Kingdom, Spain and Portugal. On 4 September 1981 the Commission served its Statement of Objections on them. They were alleged to have participated in price fixing by way of concerted practices, in decisions by associations, in common organizations, in joint and common agency agreements, in agreements on sales conditions and in exchanges of information. I have deliberately confined myself to reproducing the terms of the letter accompanying the Statement of Objections since the contents and scope of the latter are the subject of dispute between the parties.

35.

The hearing of the producers was held by the Commission in March/April 1982. On 1 September 1982 the Commission requested the addressees of the Statement of Objections to provide additional information under Article 11 of Regulation No 17.Taking the view that it was necessary, in the light of the observations made by the producers in reply to the Statement of Objections and at the hearing, to update and complement the information relating to the facts on which the proceeding was based, ( 13 ) the Commission asked the producers, in particular, to submit for the period from 1974 to the second quarter of 1982 the invoices and documents concerning transaction prices that differed from announced prices. The producers accordingly forwarded over 100000 invoices to the Commission.

36.

During the first six months of 1982, the producers and the Commission entered into negotiations concerning an undertaking by the producers to change their conduct on the Community market, particularly as regards the choice of the currency in which prices were to be announced or invoiced and the period of validity of price announcements. The Commission's aim was to reduce what it describes as ‘artificial transparency of the market’ in order, still according to the defendant, to make concertation on prices more difficult between the producers. ( 14 ) That undertaking was initially intended to serve as a basis for an amicable settlement of the matter (as the Commission acknowledged at the hearing in reply to a specific question in that regard). The discussions continued into 1983 and 1984, raising difficulties concerning, it would seem, both the actual content of the undertaking and the fact that certain producers consistently refused to sign it, in particular because they rejected any charge that they were engaging in a concerted practice.

37.

At the beginning of December 1984 the Swedish and Finnish producers signed the undertaking in the terms desired by the Commission, which included ‘currency’ clauses that were rejected by the North American producers at the time. According to those producers, on 11 December 1984 the Commission advised them that an amicable settlement was out of the question unless they gave an undertaking identical to that signed by the Scandinavian producers, no later than the following morning. Some of those producers thereupon agreed to give the undertaking, but were allegedly informed on 12 December that on the previous evening it had been resolved to adopt a decision.

38.

On 15 December 1984 the press ( 15 ) named the producers concerned and gave details of the fines that would be imposed on them by a forthcoming Commission decision. In imposing those fines, the Commission was said to have taken into account the undertaking signed by the Swedish and Finnish producers with regard to their future conduct.

39.

On 18 December 1984 the North American applicants were informed by the Commission that it would not go back on its resolve to adopt a decision and that the normal procedure would be followed leading to the notification of the decision. ( 16 ) The producers were also given until 18.00 hours (Belgian time) on the same day to agree to an undertaking identical to that which had been agreed by the Swedish and Finnish producers in order to qualify for a reduction in the amount of the fines in the same proportion. That period seems to have been extended by few hours. All the applicants but three (Bowater, St Anne and International Pulp Sales) agreed to the undertaking and accordingly had their fines reduced by 90% in the decision which was adopted on 19 December 1984 in relation to 43 of the addressees of the Statement of Objections. ( 17 ) Fines of between ECU 50000 and ECU 500000 were imposed on 36 of those addressees. ( 18 ) Six addressees of the decision have their registered offices in Canada, eleven in the United States, twelve in Finland, eleven in Sweden, one in Norway, one in Portugal and one in Spain. The Norwegian, Portuguese and Spanish addressees, as well as four Swedish producers, two Finnish producers and a United States producer, were not fined.

40.

The Commission found that the addressees had committed a number of infringements. First, it was found that practically all of them had concerted on prices for bleached sulphate wood pulp announced for deliveries to the Community ( 19 ) during the whole or part — depending on the producer — of the period from 1975 to 1981, and on actual transaction prices for the same pulp charged on the markets of five Member States in the whole or part — depending on the producer — of 1975, 1976, and 1979 to 1981. ( 20 ) Secondly, the addressees who are members of KEA were found to have concerted on announced prices and on actual transaction prices for deliveries of pulp, and to have exchanged individualized data concerning prices for those deliveries. ( 21 ) Thirdly, certain addressees (Swedish producers, Finncell, other European producers and a Canadian producer) were found to have concerted on announced and transaction prices and to have exchanged within the framework of Fides ( 22 ) individualized data concerning prices for deliveries of bleached sulphate hardwood pulp to the Community from 1973 to 1977. ( 23 ) Finally, an infringement was established against some of the addressees of the decision inasmuch as they had applied, in contracts for the sale of wood pulp to customers in the Community, clauses prohibiting export or resale of wood pulp purchased by the latter. ( 24 )

41.

The undertaking is annexed to the decision which specifies that the fines imposed on the producers who have signed it have been substantially reduced.

42.

So far as the decision is concerned, it is necessary to recall certain aspects of the Commission's findings regarding ‘general’ concertation on announced prices and transaction prices, which is referred to in Article 1(1) and (2) of the decision.

43.

The operative part of the decision does not specify precisely the producers between whom concertation took place. The producers mentioned therein are said to have engaged in a concerted practice in certain specified years. The Court asked the Commission to give details of its findings, which it did by producing lengthy tables that are set out in the addendum to the Report for the Hearing.

44.

It would appear that the ‘Commission concluded that concertation had taken place at least between those producers who announced (or sold at) a common price for a given product in a given area for a given quarter.’ ( 25 )

45.

In that regard, the details of the Commission's findings relating to prices are set out in Tables 6 and 7 annexed to the decision.

46.

Table 6 gives the prices that were allegedly announced by the producers concerned for each quarter of the period covered by the decision.

47.

Table 7 gives the transaction prices. The version published in the Official Journal does not mention any names. Each producer was sent a copy of the decision in which Table 7 omitted the names of its competitors and gave only that producer's own prices. The Court had received during the written procedure the table setting forth the transaction prices of all the producers.

48.

In April 1985, 28 of the 43 addressees of the decision brought between them a series of ten actions for annulment. ( 26 )

49.

In view of the fact that on 21 February 1989 The Mead Corporation (Case 114/85) withdrew from the proceedings and that, by judgment of 27 September 1988, the Court declared the decision void in so far as it related to KEA (Case 114/85), the number of applicants has been reduced to 26, namely one association (Finncell) and 25 producers. The Swedish producers did not bring proceedings.

50.

The applicants seek the annulment of the decision adopted by the Commission against them or, failing that, a reduction in the fines imposed. In addition, some of the applicants have asked the Court to annul or discharge them from the undertaking.

51.

It should be borne in mind that in its judgment of 27 September 1988 the Court rejected the submission alleging that the Community lacked jurisdiction in relation to producers, such as the applicants, established outside the Community, on the ground that the conduct which they are alleged to have engaged in took place within the common market.

52.

By order of 25 November 1988 the Court decided to obtain an expert's report relating essentially to an analysis of the documents concerning the announced prices and the transaction prices; on 16 March 1989, the Court appointed the firm of Moret and Limperg to draw up the report together with Mr Whitehouse.

53.

I refer to the Report for the Hearing for the questions formulated by the Court and the written conclusions of the experts, who were heard by the Court on 5 June 1990.

54.

The Court subsequently asked Mr Fishwick and Mr Cockram to prepare an expert's report designed, in particular, to describe and analyse the characteristics of the wood pulp market during the period covered by the decision and to give their opinion, in the light of those characteristics, on whether and for what reasons the natural operation of the wood pulp market should lead to a differential price structure or to a uniform price structure. I shall deal at some length with the experts' conclusions, which are set out in the addendum to the Report for the Hearing, in my analysis below.

55.

As a preliminary remark, let me point out that some applicants raised an objection under Article 91(1) of the Rules of Procedure at the same time as they instituted these proceedings. They have asked the Court, in substance, for an order restraining the Commission from using or invoking in these proceedings the invoices communicated to the latter after the producers were heard. According to those applicants, the Statement of Objections did not contain any allegation of an infringement involving concertation on transaction prices.

56.

In their view, the Commission based the findings in its decision relating to concertation on transaction prices exclusively on an analysis of the invoices forwarded by the addressees after they had been heard. During the administrative procedure they were denied access to those documents by the Commission. The latter cannot remedy that irregularity by giving the producers an opportunity to consult those documents before the Court. That is not the function of the Court, which would be obliged to consider complex facts and rule on matters which the Commission ought normally to have examined during the administrative procedure on the basis of the observations submitted by the producers. The defendant cannot therefore make use of that evidence before the Court.

57.

The Court has reserved for the final judgment the decision on that interlocutory application which the Commission considers inadmissible on the grounds that it constitutes in fact a submission on the substance of the case, and is, moreover, irrelevant, and not a procedural issue or an objection. Although I do not consider it essential to state my views in that regard, let me point out that the applicants' reasoning in connection with the interlocutory application is substantively identical to that relied upon in support of the main applications themselves inasmuch as the latter seek the annulment of the decision with regard to concertation on transaction prices on the ground that the rights of the defence have been infringed. I would therefore suggest that the Court begin by considering the applicants' submissions concerning the irregularities in the administrative procedure which relate to transaction prices.

58.

I shall examine in turn:

59.

All the applicants have criticized the procedure followed by the Commission in relying on complaints which, on examination, let it be said at once, reveal certain manifest irregularities affecting much of the contested decision.

60.

Before we examine those irregularities, it should be noted that Bowater has pointed out that the letter notifying the decision bears a date on which the signatory was no longer the Commissioner responsible for competition. The Commission explains that anomaly by the Secretariat General's heavy workload at the time, in particular as a result of the large number of decisions despatched at the end of 1984.

61.

It is scarcely satisfactory for the Court to be confronted with a document which was not signed on the date which it bears. However, that finding, which is only to be deplored, does not seem to me to have any effect on the validity of the decision itself. That, it seems to me, is the conclusion which must be drawn from the Court's case-law, according to which ‘irregularities in the procedure for notification of a decision are extraneous to that measure and cannot therefore invalidate it.’ ( 28 ) The Court undoubtedly takes the view, however, that such irregularities are capable of preventing the period for lodging an application from starting to run. But it cannot be disputed in this case that the applicants ‘had full knowledge of the text of the decision and... exercised [their] right to institute proceedings within the prescribed period’. ( 29 ) Accordingly, ‘the question of possible irregularities concerning notification ceases to be relevant’. ( 30 )

62.

I shall examine in turn the irregularities alleged with regard to the existence of complaints embodied in the decision and not referred to in the Statement of Objections,

63.

The applicants found to have concerted on transaction prices ( 31 ) (that is to say the prices actually charged) claim that the Statement of Objections was concerned exclusively with concertation on announced prices. In embodying in its decision a distinct infringement relating to concertation on transaction prices, therefore, the Commission failed to observe the rights of the defence guaranteed in Article 4 of Regulation No 99/63 and in Article 19 of Regulation No 17, which provides that the only complaints which may be taken into account are those on which the undertakings concerned have been given an opportunity to make their views known. According to the producers, it was only in October 1984, that is to say two months before the adoption of the decision, that the Commission informed them that it would also make a finding of infringement relating to concertation on transaction prices.

64.

The Commission counters those arguments primarily by reference to the wording of the Statement of Objections, which contains passages concerning concertation on both transaction prices and announced prices.

65.

Before considering the passages in question, there is a point I wish to make. The defendant is obliged to rely both on passages in the ‘facts’ part of the Statement of Objections and on passages in the part headed ‘applicability of Article 85(1) of the EEC Treaty’. In the absence of a formal operative part, I consider that at least the latter part of the Statement of Objections should specifyin detail the infringements with which the addressees are charged.

66.

In the light of those remarks, the fact that certain passages in the ‘facts’ part of the Statement of Objections refer to transaction prices ( 32 ) is, in my view, of only limited significance in determining whether concertation on those prices was envisaged.

67.

As for the part headed ‘applicability of Article 85(1) of the EEC Treaty’, it refers to ‘concerted price-fixing practices pursued inter alia through the system of price announcements’. ( 33 ) Even though the expression ‘inter alia’ is underlined in the body of the text, it would require a considerable effort of the imagination to regard it as a reference to concertation on transaction prices. In any event, it is very much an implicit reference, to say the least.

68.

Paragraph 54 et seq. of the Statement of Objections, in which the Commission analyses concerted price-fixing practices between the producers, call in question essentially the practice of price announcements and ‘other contacts between producers’. The only reference which may be relevant ( 34 ) in support of the Commission's contention is to be found in paragraph 66 which states that ‘the North American producers charged until 1978... the same price as the Scandinavian producers, except during the first half of 1977, when they granted rebates and increased their market share’. In any event, a reading of the Statement of Objections does not reveal any express and unequivocal reference to support the conclusion that a complaint was formulated regarding concertation on transaction prices.

69.

However, the Commission maintains in its rejoinder that it is clear from the replies of certain producers to the Statement of Objections or at the administrative hearing that they had understood that concertation on transaction prices was meant.

70.

I do not believe that the passages relied upon by the Commission in that regard conclusively bear out its argument. Admittedly, KEA itself considered that ‘the Statement of Objections is wrong in assuming that actual transaction prices did not fluctuate in accordance with the market’, and criticized that document, moreover, for stating that ‘actual transaction prices were the same as announced prices or so-called list prices’. It is also true that the Finnish applicants maintained that competition was extremely vigorous and had led to the fixing of widely differing transaction prices. Nevertheless, those assertions do not in any way establish beyond dispute that the Statement of Objections envisaged concertation on transaction prices. The argument that those prices differed from announced prices may constitute a defence to a charge of concertation on announced prices alone. Nor does the Commission gain more leverage by relying on passages in which, for instance, the Canadian producers state that ‘each of the Canadian respondents is charged with a concerted practice to fix price’ or Bowater denies participating ‘in any price fixing or price setting’. It would still be necessary to establish beyond doubt that the expression ‘to fix price’ covers concertation on transaction prices, and that is precisely what has to be proved.

71.

Admittedly, the Commission's view that to distinguish between announced prices and transaction prices is not tantamount to considering that two wholly distinct concerted practices are involved would not seem to be altogether illogical.

72.

However, is the uncertainty not the result primarily of the course of action taken by the Commission itself? In that regard, the decision found that concertation on prices charged and concertation on announced prices constituted two distinct infringements. Thus, for 1977 and 1978, no concertation on transaction prices was found whereas, for the same years, an infringement was established with regard to announced prices. In the decision, St Anne was found to have taken part only in concertation on announced prices, whilst other producers were alleged to have taken part in concertation on transaction prices alone. Furthermore, it is clear from the explanations of the Commission itself that the fines were determined by taking into account concertation on announced prices separately from concertation on transaction prices.

73.

It must be stated that in the decision the Commission did indeed take the view that there were two distinct infringements, whereas in the Statement of Objections there was no express charge concerning the complaint regarding concertation on transaction prices.

74.

There is a further consideration which tips the scales decisively in favour of the applicants' argument. In Annex VI to the Statement of Objections there is a table which in fact refers only to announced prices, as the Commission confirmed, moreover, at the hearing in reply to a specific question on that point. The Statement of Objections does not contain any annex comparable to Table 7 of the decision which sets out transaction prices. That difference between the Statement of Objections and the decision is in itself extremely revealing. Admittedly, the Commission maintains that in the Statement of Objections it had quoted the announced prices on the ground that those prices mostly corresponded to the transaction prices. ( 35 ) However, that argument runs counter to the Commission's line of reasoning. In support of its contention to the effect that concertation on transaction prices was envisaged in the Statement of Objections, the Commission points in particular to certain passages thereof which mention a difference for certain periods between announced prices and actual prices. ( 36 ) Annex VI to the Statement of Objections contains no mention of transaction prices for those periods. To accept that the Statement of Objections envisaged concertation on transaction prices, it would still have been necessary for the table in question to show, clearly at the very least, the periods in which, according to the Commission itself, they had differed from the announced prices.

75.

As the Court has declared, the purpose of the Statement of Objections ‘is to enable those concerned to put forward their arguments in the context of proceedings initiated against them’. To that end, it must be possible for the addressee actually‘to take cognizance of the objections held against him’. ( 37 )

76.

That function of the Statement of Objections would be largely disregarded in this case if the Commission's view were to prevail. The Court considers that the Statement of Objections satisfies the requirements laid down in Article 4 of Regulation No 99/63 where ‘it sets out, summarily indeed but clearly, the principal facts upon which the Commission relies’. ( 38 ) That requirement of clarity, which the Court has consistently recalled and verified, ( 39 ) is quite obviously not complied with when laborious explanations are needed to discern in the Statement of Objections the existence of a specific charge. It is essential for complaints which are brought to the attention of an undertaking to be precisely formulated ( 40 ) in order to enable the latter to present a full defence. In areas characterized by the complexity of the facts, the volume of information and the consequences of the procedures in question, the Court cannot allow those requirements to be satisfied by stretching the terms of the Statement of Objections which, in this regard, does not in any case contain an expressly formulated charge regarding concertation on transaction prices.

77.

I would therefore propose that the Court hold that, by making a finding in its decision of an infringement regarding concertation on transaction prices which had not been clearly, precisely and expressly envisaged in the Statement of Objections, the Commission has failed to comply with its obligations under Article 4 of Regulation No 99/63 and Article 19 of Regulation No 17. That infringement should, in my view, lead to the annulment of Article 1(2) of the decision, as requested by the applicants concerned. In any event, my examination of the legality of the procedure in regard to the evidence on which the Commission relied in establishing concertation on transaction prices will lead to the same conclusion.

78.

A number of applicants claim that the duration of the infringement as found in the decision was longer than the period referred to in the Statement of Objections.

79.

The decision establishes the existence of concertation until the date of the Statement of Objections, namely September 1981. Table 6 annexed to the decision shows the prices announced for the four quarters of 1981, whilst Table 7 shows the transaction prices for the first three quarters of that year.

80.

The Canadian applicants and Bowater claim that the Statement of Objections covered a period which did not go beyond the second quarter of 1980.

81.

The Commission maintains that the Statement of Objections covers the whole of the period from 1973 until the date of its notification, namely September 1981. The description of the market and the conduct of the addressees of the Statement of Objections, as well the assessment of that conduct, relate to the whole of that period. There is not a single passage in the Statement of Objections which confines to a shorter period the description of the facts relating to the involvement of each applicant in the infringements.

82.

Let me point out in the first place that although the table set out in Annex VI to the Statement of Objections goes no further than the second quarter of 1980, there is a note expressly stating that prices were the same for the third and fourth quarters of 1980. At the very least, therefore, 1980 was covered in its entirety. But what about 1981? In my view, having regard to the information provided by the aforesaid table, a clear reference was necessary specifying that the infringement continued after the end of 1980 since prices for 1981 were not set out in Annex VI to the Statement of Objections. In maintaining that the Statement of Objections was never confined to a period shorter than that ending in September 1981, the Commission is, it seems to me, inverting the terms of the issue. The absence from Annex VI of any reference to 1981 should logically have been such as to exclude that period from the duration of the infringement, unless the Commission expressly stated otherwise in the Statement of Objections itself.

83.

According to the defendant, however, the document ‘clearly’ calls in question the addressees' ongoing conduct.

84.

I must voice serious reservations in that regard. There is not a single passage in the Statement of Objections which the Commission can properly rely upon in support of that contention, with the exception of paragraph 21 which reads ‘the table in Annex VI shows the prices announced by the... producers from 1974 to 1981’. The Court will gauge the weight to be attached to that statement from the fact that the table in question does not in fact contain any reference to prices after 1980. Furthermore, that sentence immediately precedes a long passage describing price trends for wood pulp over the period from 1974 to 1980. Accordingly, the statement in question in no way corroborates the Commission's contention.

85.

The Commission also denies that it was raising a fresh issue by extending in its decision the duration of the infringement until the date of the Statement of Objections. The Commission simply found that the addressees of the decision had continued to engage in the same conduct until the date of notification, pointing out that the prices announced and charged between the third quarter of 1980 and the third quarter of 1981 were the same as those announced and charged during the second quarter of 1980. There is no rule of Community law which requires the Commission to use only the information at its disposal when it draws up the Statement of Objections. Furthermore, in submitting the invoices and documents relating to 1981 requested by the Commission in its letter of 1 September 1982 pursuant to Article 11 — covering, it should be borne in mind, the period from 1974 to 1982 — the applicants are said to have commented on the Commission's allegation that they had concerted on prices. The Commission's letter referred expressly to the Statement of Objections and to the administrative hearing, yet none of the applicants refused to submit invoices for the period after the second quarter of 1980 on the ground that that period was not covered by the Statement of Objections.

86.

Let me point out first of all that the letter of 1 September 1982 seeking ‘to update and complement the information on the facts on which the... proceeding is based’ cannot in any event be equated with a Statement of Objections. Moreover, it seems to me particularly dangerous to infer from the attitude of the applicants, who did not express any reservations with regard to the transmission of the documents in question, implied confirmation that the Statement of Objections covered 1981.

87.

As certain producers point out, they also submitted in their replies documents relating to the period subsequent to the Statement of Objections — third quarter of 1981/second quarter of 1982 — which, by definition, cannot have been covered by the Statement of Objections.

88.

Besides, in response to the Commission's argument to the effect that ultimately the applicants should be taken to have been aware that the period subsequent to 1980 was covered by the Statement of Objections on the ground that they did not refuse to submit documentation in respect of that period, the producers maintain that they were trying to be cooperative on the assumption that the information would be used, in particular, for comparative purposes.

89.

The Court has held that the duration of an infringement constitutes an ‘essential factor’ in the infringement, which the Commission must indicate ‘on the basis of the information available to it at the time when it formulates the Statement of Objections’, ( 41 ) although ‘the Commission may extend the period thus stated if supplementary information obtained during the administrative procedure so justifies, provided that the undertakings have had an opportunity to make their views known in that respect’. ( 42 )

90.

In this case, by claiming the right to use information obtained during the administrative investigation, the Commission would thus seem to acknowledge that it was that investigation which yielded the information for the period subsequent to 1980. It was still necessary for the producers in question clearly to be given an opportunity to state their views in that regard, something which has not specifically been established in the Commission's observations to the Court.

91.

I would therefore propose that the Court uphold the application submitted by the applicants concerned so far as concerns 1981, unless it adopts a purely exegetic argument based on the covering letter accompanying the Statement of Objections, according to which ‘it is proposed to require the firms concerned to terminate such infringements’. ( 43 )

92.

IPS claims that the Statement of Objections envisaged infringements by IPS only in the period from 1974 to 1978 (with the exception of the first quarter of 1977), whereas the decision concerning IPS covers the period from 1975 to 1981.

93.

In my view, that argument cannot be upheld. With regard to the period commencing in the third quarter of 1978, Annex VI to the Statement of Objections refers to the prices of ‘producers US’, and it is undisputed that IPS is a United States producer. Accordingly, the can to my mind be no denying the fact that IPS was alleged to have taken part in concertation on prices in the period subsequent to 1978. On the same grounds, the Court should reject the argument in which Bowater, also a United States company, denies that the Statement of Objections called in question its conduct for the period subsequent to the first quarter of 1978.

94.

On the other hand, the Court should uphold Bowater's arguments with regard to 1975. Annex VI to the Statement of Objections does not mention Bowater's prices for that year, although it does specify those of the other United States producers for the same period, who are designated by name.

95.

Bowater's absence from the list set out in Annex VI for 1975 is not disputed by the Commission. The Commission also acknowledges that it did not contradict Bowater when the latter stated, at the administrative hearing, as regards the period from 1975 to 1976, that it was not involved in a ‘process covering a period first of all where there is no complaint, or none against us separately’. The defendant considers, however, that a possible irregularity cannot lead to the annulment of the decision in that respect since, in view of the existence of convincing evidence for the period in question, it does not seem that the Commission's finding would have differed in substance if the complaints relating to 1975 had been couched in more specific terms in the Statement of Objections.

96.

In my view, that suggestion is unacceptable: Bowater's conduct in 1975 was not in any event referred to in the Statement of Objections, if it is borne in mind that Annex VI does not give any of Bowater's prices for that year. Moreover, that annex, as we have seen, relates to announced prices whilst Bowater was found only to have participated in concertation on transaction prices in 1975. The applicant pointed out at the administrative hearing that it did not consider its conduct for that period to have been called in question. Since the Commission acknowledges that it refrained from challenging it in any way at the time, there cannot be any justification now in maintaining that the procedural irregularity is ‘covered’, as it were, in view of the evidence against Bowater for the period in question.

97.

The Commission seems to be suggesting here that the Court should apply the doctrine of breach of essential procedural requirements.

98.

It is true that in its judgment in Distillers ( 44 ) the Court considered it unnecessary to examine the procedural irregularities alleged by the applicant, pointing out that ‘the position would be different only if in the absence of those irregularities the administrative proceedings could have led to a different result’. ( 45 ) But it is necessary to specify what irregularities were at issue in that case. They related, in particular, to the failure to notify to the Advisory Committee the minutes of the applicant's hearing and matter supplementing the applicant's answer to the Statement of Objections, and to notify the complaint in full. The Court held that those procedural defects could not in any event have affected the decision inasmuch as it refused to grant exemption under Article 85(3) on the ground that as the price terms had not been notified to the Commission, they could not in any event be examined with a view to exemption.

99.

The difference between the gravity of those irregularities and that of an infringement resulting from an impairment of the primary function of the Statement of Objections — namely to indicate precisely the complaints made against an undertaking in order to enable it to present its defence — is too obvious for there to be any need for me to draw attention to it here. Let me simply point out that the contention put forward by the defendant in these proceedings could ultimately lead to the formulation of complaints against an undertaking being deemed to serve no purpose if the Commission considered that sufficient evidence existed against that undertaking. That is precisely the consequence to which the Commission's position leads on the specific issue of the complaint that Bowater engaged in concertation in 1975, and I strongly advise the Court to reject it.

100.

IPS considers that the Statement of Objections envisaged only its involvement in a concerted practice with the Scandinavian producers; in the decision IPS is alleged to have concerted on announced prices during certain periods with the Canadian producers and with other United States producers. In addition, so far as transaction prices are concerned, IPS claims that in the decision it is charged with concerting on southern hardwood pulp with producers whom the applicant ‘assumes’ are from the United States.

101.

The Commission emphasizes that by calling in question ‘concerted practices between Scandinavian producers and North American producers’, the Statement of Objections also envisaged concertation within each of those two groups of producers. According to the defendant, if the Statement of Objections had called in question only concertation between Scandinavian producers on the one hand and North American producers on the other, the result would have been a type of concertation so extraordinary that it could have been described as fanciful. It would mean interpreting the Statement of Objections as covering a series of distinct concerted practices, each involving producers from two continents separated by thousands of kilometres, but never involving more than a single competitor from the same region, the same country or the same continent.

102.

It is necessary to ascertain, therefore, whether the words ‘concerted practices between Scandinavian producers and North American producers’ covered only concertation between those two groups of producers, without calling in question concertation between producers within each of those groups.

103.

In my view, the Commission's standpoint, outlined above, would seem to be appropriate. It clearly exposes the ‘unreasonable’ consequences of IPS's argument regarding the identity of the producers involved in the concertation established in the Statement of Objections. It is self-evident, however, that the words ‘concerted practices between Scandinavian producers and North American producers’ are not entirely free from ambiguity. But it is clear from the actual contents of the Statement of Objections, in my view, that both concertation between regional groups and concertation within those groups were meant.

104.

Thus the following passage is to be found, as the Commission rightly emphasizes, under the heading in question: ‘C. Concerted practices between Scandinavian producers and North American producers. The relevant documents (...) indicate the existence of direct contacts between the Canadian firm MacMillan and US producers Weyerhaeuser and Georgia Pacific. The contacts resulted in exchanges of information on KEA prices and on sales quantities’. That passage establishes beyond doubt that concertation within the group of North American producers was covered by the Statement of Objections. The Court should therefore reject IPS's arguments in that regard.

105.

St Anne, a Canadian producer, claims that the Statement of Objections did not call in question its participation in Fides, whereas the decision established that the applicant had committed an infringement on that account.

106.

The Commission acknowledges that Annex VII to the Statement of Objections, which lists the members of Fides, does not mention St Anne, and that the references in other passages to concerted practices between ‘Scandinavian producers and European producers’ to describe such practices within Fides are ‘unfortunate’ since they do not envisage St Anne, a Canadian producer. However, in view, in particular, of the telex mentioned in the factual part of the Statement of Objections referring to St Anne, the defendant considers that it is for the Court to determine whether St Anne should have regarded the Statement of Objections as alleging its participation in concerted practices within Fides.

107.

Let me point out first of all that in its legal assessment the Statement of Objections does not refer to St. Anne in connection with concertation within Fides. Nor is that producer included on the list in Annex VII to the Statement of Objections headed ‘Principal Members of the Hardwood Pulp Section of Fides’.

108.

Next, the factual part of the Statement of Objections sets out the text of a telex from a third party reporting on a meeting in which various producers, including St Anne, exchanged information on prices. In my view, however, that reference cannot constitute a sufficient ground for alleging that St Anne was involved in concertation within Fides. Another undertaking mentioned in the same telex as having taken part in the same meeting, Celbi, was not one of the addressees of the Statement of Objections and has not been involved in these proceedings. Accordingly, if the reference to Celbi in that document did not result in the Commission addressing the Statement of Objections to that undertaking, I do not consider that the telex could in itself have sufficed to formulate a complaint against St Anne regarding participation in Fides.

109.

Furthermore, it is unacceptable, in my view, for the Commission to justify in this case the failure to formulate a complaint against St Anne on the ground, relied upon on in the defence, that ‘it would have taken too long to enumerate, each time, all producers involved’. It was necessary to specify clearly, at least once, that St Anne was alleged to have taken part in concertation within Fides. In addition, the Commission emphasizes that St Anne is not mentioned in Annex VII to the Statement of Objections because that list is not exhaustive and concerns only the principal members. That argument, it seems to me, leads to the opposite conclusion to that which the Commission purports to draw. If St Anne is not regarded in the Statement of Objections as a sufficiently ‘important’ member of Fides to be included as such in the list of producers annexed to the Statement of Objections, the only logical consequence which can be drawn therefrom is that no complaint was made against St Anne in that regard. The interpretation which the Commission apparently seeks to derive from the fact that the list is not exhaustive flouts the most elementary requirements regarding the precision and clarity with which complaints must be formulated. Failure to comply with those requirements, in connection with St Anne's participation in Fides, must entail the annulment by the Court of Article 1(4) of the operative part of the decision in so far as it affects St. Anne.

110.

To begin with, let me remind the Court that earlier I came to the conclusion that the Statement of Objections did not in any event set out the allegation of concertation on transaction prices clearly, precisely and expressly, and therefore in conformity with the requirements stemming from the rights of the defence.

111.

Next, in order to make it easier to follow the applicants' reasoning, it is necessary to recapitulate briefly the chronological sequence of events in the proceeding. After hearing the producers in March/April 1982, the Commission notified them on 1 September 1982 of its intention ‘to update and complement the information on the facts on which the above proceeding is based’. With that end in view, it requested them to forward the invoices and documents relating to the period from 1974 to 1982 in order to establish that the transaction prices differed from their announced prices, an argument put forward by the applicants at the administrative hearing. In response to that request, made under Article 11 of Regulation No 17, approximately 100000 invoices were forwarded to the Commission, as I pointed out earlier.

112.

The applicants found to have concerted on transaction prices dispute the legality of the procedure followed by the Commission. ( 46 ) They maintain that the Commission relied exclusively on a comparative analysis of the documents in question as a basis for the findings in its decision with regard to transaction prices. Accordingly, they consider that the rights of the defence have been infringed since the defendant allegedly failed to inform them before the adoption of the decision of the results of that analysis. As regards the invoices communicated by the other producers, they emphasize, relying on the case-law of the Court and, in particular, on its judgment in Hoffmann-Laroche, ( 47 ) that if the confidential nature of those documents precluded their being communicated to the applicants, the Commission was not entitled to use the documents against them. In any event, the defendant should have asked the producers, before adopting the decision, whether they had any objection to their invoices being forwarded to their competitors or to their competitors' lawyers.

113.

The Commission denies in the first place that it relied exclusively on documents produced after the administrative hearing. It points out that in the course of its investigations in 1977 it asked the producers concerned for representative invoices showing that the prices actually charged duly corresponded to the ‘announced’ prices. The invoices produced after the administrative hearing merely confirmed that those prices were the same. The Commission also points out that it relied exclusively on the documents submitted by each applicant for confirmation that the transaction prices of each did not generally differ from its announced prices. Where those prices were broadly the same as the transaction prices, it was unnecessary to draw any conclusions from a comparison with the documents of other producers which were inaccessible to the applicants. In its view, therefore, the applicants had in no way substantiated their allegations to the effect that the Commission had relied entirely on conclusions of that kind or exclusively on documents submitted after the hearing. It was not necessary for each producer to consult the other parties' invoices to assert that its transaction prices differed from its announced prices, since those invoices had played no part in the Commission's assessment of that producer's involvement in concertation on transaction prices.

114.

Finally, the defendant contends that it was under no obligation either to inform every applicant of the results of its analysis of the documents produced by the latter or to give every applicant an opportunity to submit its observations on the Commission's evaluation. Each producer was given an opportunity to state its views on the facts alleged and on the legal assessment of those facts set out in the Statement of Objections, and to defend itself. It did not have any right to be informed whether or not the Commission considered its defence persuasive, or to discuss the position taken by the Commission before a decision was adopted.

115.

Let us begin by considering the applicants' claim that the Commission based it findings regarding concertation on transaction prices exclusively on documents produced after the administrative hearing.

116.

It should be borne in mind that the Court asked the ‘accountancy’ experts whether the documents which were available to the Commission when it drew up the Statement of Objections had already enabled it to conclude that the transaction prices offered by the producers to customers were identical to the prices announced by them.

117.

The experts' answer to that question was that the documents in question were not sufficient to enable the Commission to conclude that the transaction prices offered by the producers to their customers were the same as their announced prices.

118.

In reaching that conclusion, the experts pointed out, on the basis of an analysis of the documents that were at the Commission's disposal when it drew up the Statement of Objections, that

119.

Those conclusions, which are amply reasoned, establish beyond doubt that at the date of the Statement of Objections, the Commission had no objective evidence whatsoever that the announced prices and the prices charged were the same for all the producers. Even on the assumption that the sample available to the Commission at the time showed that the announced prices were the same as the prices charged, those invoices related only to a very small number of producers and transactions. Accordingly, the documents relating to transaction prices which the Commission had on file at the date of the Statement of Objections did not by any means constitute objective evidence concerning transaction prices against the vast majority of the producers.

120.

Accordingly, the evidence on which the finding of concertation on transaction prices is based — namely the documents used to compile Tables 7 and 8 annexed to the decision — was obtained in the main after the administrative hearing. Furthermore, a reading of paragraph 25 of the decision would seem to dispel any doubts in that regard. ( 49 )

121.

That finding thus raises the question whether the Commission was under an obligation to draw up a new Statement of Objections.

122.

According to the case-law of the Court, ‘the Commission must take into account the factors emerging from the administrative procedure in order... to supplement and redraft its arguments both in fact and in law in support of the objections which it maintains’ and ‘the decision is not necessarily required to be a replica of the Commission's notice of objections’. ( 50 ) The Court has also held that the Commission ‘has the right and where appropriate the duty to institute fresh inquiries during the administrative procedure if it appears from the course of that procedure that additional investigations are necessary’ ( 51 ) and that the Commission is not systematically obliged to draw up an additional Statement of Objections following those investigations. The Court has made it quite clear, however, that an additional Statement of Objections would be required if the result of the further inquiries led the Commission ‘to take new facts into account against the undertakings or to alter materially the evidence for the contested infringements’. ( 52 )

123.

Let us pause to consider the reasons which seem to have led the Court to lay down that requirement. The Statement of Objections constitutes the ‘measures stating the final attitude of the Commission concerning undertakings’. ( 53 ) On the one hand, it informs undertakings of the conduct which constitutes an infringement of the rules of competition, that is the complaints against them. On the other hand, however, it also brings to their attention the principal evidence in the Commission's possession, that is the evidence on which they may be charged with one or more infringements. When there is a very considerable alteration in the evidence as a result of investigations carried out during the administrative procedure, the initial Statement of Objections can no longer correctly describe the basis on which the complaints are founded. In other words, the Commission is then required formally to approach the undertakings concerned, specifying the substantial alteration in the evidence henceforth at its disposal.

124.

That, in my view, is precisely the step which the Commission was required to take in this case if it had reached the conclusion that the 100000 or so invoices ‘confirmed’, in the Commission's contention, the existence of concertation on transaction prices.

125.

The most, elementary common sense shows that there had been, at the very least, a material alteration in the evidence of the infringements: it is sufficient to note that the Commission did not have at its disposal, by the date of the Statement of Objections, the information to compile Tables 7 and 8 annexed to the decision. In addition, the substantial administrative task — emphasized by the Commission itself — of analysing the invoices involved deepening and widening its investigations, which in themselves provided a very revealing picture of the substantial extension in the range of evidence of concertation on transaction prices.

126.

Applying the Court's case-law to this case, the Commission would have been required to draw up a new Statement of Objections after processing the 100000 or so invoices and credit notes. That Statement was all the more essential since, in my view, as I mentioned earlier, the actual complaint regarding concertation on transaction prices was not referred to in a manner consistent with the requirements stemming from the rights of the defence in the Statement of Objections of 4 September 1981. That is my first conclusion, which would therefore lead to acceptance of the contention that the Commission unlawfully based its finding of concertation on transaction prices largely on documents submitted after the administrative hearing, when it was required to issue a new Statement of Objections.

127.

However, I shall extend my investigation to two further questions:

128.

It should be borne in mind, in the first place, that the Commission did not give each producer the opportunity, before the decision was adopted, to discuss the conclusions which it had reached concerning that producer's own invoices and credit notes. In that regard, a letter dated 12 November 1984 states as follows:

‘It was the respondents in their replies who alleged that although the announced prices were similar the transaction (applied) prices widely differed. In raising this defence and submitting relevant documents the respondents have therefore already commented on the finding of the Commission that their transaction (applied) prices were broadly identical. Each respondent had the opportunity “to make known its views on the truth and relevance” of the invoices, credit notes, etc. submitted by him’. ( 54 )

129.

Let me state first of all that, as the Canadian applicants have righdy pointed out, the administrative hearing could not ex hypothesi relate to the conclusions reached by the Commission in its decision to the effect that transaction prices were identical since, as we have seen, it was essentially the invoices received subsequently which made it possible to substantiate that conclusion.

130.

Secondly, I fail to grasp the significance which the Commission attaches to the statement that the producers had the opportunity to make known their views on the correctness and relevance of the documents in question at the same time as they forwarded them. The crucial issue here is whether each producer was entitled to be notified before the adoption of the decision of the individual conclusions resulting from the Commission's analysis of the documents forwarded to it by that producer.

131.

In support of an affirmative answer, it should be borne in mind that the Court stated in its judgment in AEG that

‘it must be observed that the important point is not the documents as such but the conclusions which the Commission has drawn from them’ ( 55 )

and excluded certain documents which emanated from the company in question itself but which it had been unable to react to before the adoption of the decision.

132.

However, the AEG case differed from the present cases in one respect. Here, the documents were submitted by the producers themselves to the Commission, whilst in that case it was not AEG which had communicated the documents in question to the Commission. AEG was not in a position to know that the documents would be used in the decision. In this case, on the other hand, the producers quite clearly could not rule out that possibility.

133.

It is plain from the Court's decision, however, that knowledge of the document ‘as such’ is not necessarily sufficient to ensure compliance with the rights of the defence. It is indisputable that in these proceedings, regard being had in particular to the volume of documentation and the need to assess the competitive nature of rebates or discounts, the conclusions drawn from the documents presupposed an in-depth and detailed analysis, as is borne out by the discussions before the Court with regard to the first expert's report.

134.

However, the producers were never given an opportunity to discuss the Commission's conclusions before the decision was taken, precisely because at the time of the administrative hearing the invoices in question were not in the defendant's possession. It is obvious, therefore, that they were found to have concerted on transaction prices even though they had not been given an opportunity to state their views on the Commission's analysis of their own invoices, on which the complaint was based. As for the importance of discussing conclusions drawn from documents which, I repeat, were produced after the Statement of Objections, it is sufficiently apparent from the fact that they make up the total contents of Tables 7 and 8 annexed to the decision, on which the finding of concertation on transaction prices is based. That, in my view, is a further ground on which the Court should annul the findings in the decision regarding concertation on transaction prices, if it adopts my opinion that it was necessary to communicate to each producer the conclusions from the study of the invoices submitted by that producer.

135.

Second question: Was the Commission entitled to refuse to communicate to the producers the invoices from the other producers before the adoption of its decision?

136.

The defendant's refusal is based on two considerations:

137.

Let us consider the latter justification first. The Commission contends that a comparison between the producers' invoices was unnecessary. It was sufficient for it to establish that each producer's transaction prices did not as a rule differ from its announced prices. However, a reading of the decision reveals precisely findings as to the similarity or identical nature of the producers' transaction prices.

138.

Thus, in the following passage, the similarity between transaction prices is referred to in these terms: ‘The firms concerned allowed such discounts on very similar terms. They usually did not exceed 3% and never amounted to more than 7%’. ( 56 ) Moreover, the decision itself describes the conduct of two producers in certain periods as regards the prices charged in support of the view that it was possible to differentiate between the prices for bleached sulphate pulp. In the case of the first producer, Domtar, the decision states that ‘prices ranged according to customer from USD 420 to USD 565 per tonne for northern softwood, whereas the firms to which this decision is addressed were uniformly charging USD 545 per tonne’. ( 57 ) As for Bowater, its conduct is described in the following terms: ‘whereas, in the latter years, its transaction prices corresponded to both announced prices and transaction prices of its competitors, its transaction prices in the first two years generally differed widely from those of itscompetitors’. ( 58 ) Finally, is it not in fact the aim of Table 7 annexed to the decision precisely to draw attention to the identical nature of transaction prices and that of Table 8 to point up the similarity between the discounts or allowances granted? Nevertheless, no producer was able to consult the documents furnished by its competitors which enabled those tables to be drawn up. Contrary to the defendant's contention, the rights of the defence cannot be exercised in full except in so far as undertakings are able not only to ascertain whether the analyses of their own price structures are correct but also, and above all, to see for themselves whether their own prices are identical to those of their competitors. The Commission's position in this case irresistibly calls to mind those plans of prisons in which the warder alone is able to see all the inmates but they cannot see each other.

139.

Moreover, as regards transaction prices at any rate, the producers were never given an opportunity before the adoption of the decision to ascertain the precise identity of those of their competitors with whom they allegedly concerted on transaction prices over a given period.

140.

Therefore I cannot accept the Commission's argument that the invoices and credit notes of other competitors were unnecessary for the purposes of each individual producer's defence. I would also point out that the Commission itself, in a letter subsequent to the adoption of the decision, acknowledged that full access to the documents in question was ‘psychologically more satisfactory’. ( 59 ) In my view, such access is surely no less important from the point of view of the legal requirements.

141.

What course of action should the Commission have taken, however, in view of the fact that the documents in question were regarded as business secrets, and that as such they were confidential?

142.

In that regard, the Court has consistently stated that the Commission cannot ‘use to the detriment of the undertakings involved in a proceeding referred to in Regulation No 17, facts, circumstances or documents which it cannot in its view disclose if such a refusal of disclosure adversely affects that undertaking's opportunity to make known effectively its views on the truth or implications of those circumstances, on those documents or again on the conclusions drawn by the Commission from them’. ( 60 )

143.

I have had occasion to discuss in some depth the difficulties raised by the need to reconcile observance of confidentiality and the rights of the defence in connection with antidumping proceedings; I pointed out, in particular, that:

‘If the European Commission of Human Rights declares inadmissible applications directed against national decisions enacted pursuant to a Community act, the main reason is that, through its successive judgments, the Court has established the principle that it reviews the Community institutions' observance of fundamental rights. It is therefore far from unimportant to avoid conspicuous discrepancies between the construction this Court puts on the right to a fair trial and the requirements already laid down by the European Court of Human Rights.

On this point there seems to be no doubt that the antidumping proceeding, although conducted by an administrative authority, must meet the needs of a “fair hearing”, which implies that an “equality of arms” must prevail between the parties. Furthermore, observance of the principle of hearing arguments from both sides demands that the party or his representative have the opportunity of consulting and criticizing the case documents, and in particular the evidence on which the decision was based’. ( 61 )

144.

After recalling the fundamental character of the principle that the rights of the defence must be observed, the Court emphasized that:

‘Consequently, in performing their duty to provide information, the Community institutions must act with all due diligence by seeking, as the Court stated in its judgment of 20 March 1985 in Case 264/82, Timex Corporation v Council and Commission [1985] ECR 849, to provide the undertakings concerned, as far as is compatible with the obligation not to disclose business secrets, with information relevant to the defence of their interests, choosing, if necessary on their own initiative, the appropriate means of providing such information. In any event, the undertakings concerned should have been placed in a position during the administrative procedure in which they could effectively make known their views on the correctness and relevance of the facts and circumstances alleged and on the evidence presented by the Commission in support of its allegation concerning the existence of dumping and the resultant injury’. ( 62 )

145.

In the light of those principles, ( 63 ) therefore, in establishing the existence of concertation on transaction prices on the basis of confidential documents, the Commission should either have specified the procedures for communicating the information necessary for the exercise of the rights of the defence or have refrained from using the invoices even though its findings are based on them.

146.

It is clear from the correspondence exchanged between the Commission and the applicants after the adoption of the decision that various solutions were examined by both sides in order to find ways of enabling the producers to consult the documents in question. I do not consider it necessary at this juncture to assess the merits and drawbacks of those solutions. Allow me simply to point out that the applicants have referred to the practice applied by the Commission in other competition cases for safeguarding business secrets whilst observing the rights of the defence. In any event, however, communication subsequent to the adoption of the decision is too late to satisfy the requirements stemming from those rights if the infringements found are based specifically on the documents at issue. As we have seen, the infringement involving concertation on transaction prices was established essentially on the basis of documents and invoices submitted after the administrative hearing.

147.

The Court has categorically stated that:

‘in order to respect the principle of the right to be heard the undertakings concerned must have been afforded the opportunity during the administrative procedure to make known their views on the truth and relevance of the facts and circumstances alleged and on the documents used by the Commission to support its claim that there has been an infringement’. ( 64 )

Those requirements have manifestly not been observed in this case. That irregularity therefore affects the decision in so far as it establishes concertation on transaction prices, and it must once again lead to the annulment of that finding of infringement.

148.

The ‘status’ of the documents at issue in the proceedings before the Court remains to be defined, by way of reply, in particular, to the objection raised by certain applicants under Article 91 of the Rules of Procedure. In that regard, since I am proposing that the Court should annul the decision in so far as it concerns concertation on transaction prices, the issue concerning the use of invoices in support of that complaint becomes devoid of purpose. Furthermore, it seems reasonable to me for the Court to consider that the Commission cannot rely in these proceedings on the conclusions drawn from the documents submitted after the administrative hearing in support of its findings relating to announced prices, having regard to the procedural irregularities which I have detected.

149.

Finncell claims that the Commission's findings regarding the former's participation in concertation within the framework of Fides are based on documents which were not communicated to it, in particular written and oral opinions from third parties, and on a memorandum of a meeting said to have taken place in Zurich on 31 March 1977. Those documents, which are set out in paragraphs 57 to 60 of the decision, were produced either in reply to the Statement of Objections ( 65 ) or at the administrative hearing, ( 66 ) or else were forwarded in response to a request for further information made by the Commission after that hearing. ( 67 )

150.

The Commission does not deny that those documents were not brought to Finncell's attention, but it maintains that there were other documents set out in the Statement of Objections which in any event established Finncell's participation in the infringement in question.

151.

In my view, the information and the documents mentioned in paragraphs 57 to 60 of the decision cannot be relied upon against the applicant since it was unable to consult them before the adoption of the decision.

152.

It will be necessary, therefore, when the substance of the case comes to be considered, to verify whether this finding of infringement against Finncell is proven by the other evidence which is referred to in the decision and on which the applicant was properly given an opportunity to state its views.

153.

All the applicants, with the exception of St. Anne and IPS, claim that the Commission infringed the rights of the defence by refusing to organize a joint hearing of the producers or, at the very least, joint hearings on a regional basis. According to the producers, such hearings were necessary because the Commission was charging them with engaging in a concerted practice. They therefore had to be acquainted with the statements made when the other parties were heard in order to organize their own individual defence. They conclude, therefore, that they should have been allowed to attend the hearings of the other producers. Moreover, the Commission might have made improper use of separate hearings by pitting some of the ‘codefendants’ against the others or by extracting confidential information. Accordingly, even though Article 9(3) of Regulation No 99/63, under which persons are to be heard separately, does not provide for a joint hearing, it should be interpreted as meaning that it is for the producers themselves to decide whether it is necessary to hold a joint hearing rather than separate hearings. Furthermore, the Commission allegedly failed to keep its promise to hear the producers subsequently on any new facts or arguments which might arise in a hearing of one or other of them. Finally, certain producers state that they did not receive a copy of the record of the hearing until the end of 1985, that is say two months after the adoption of the decision.

154.

In reply to those arguments, the Commission considers that it alone has the power to decide whether or not it is necessary to organize a joint hearing. In this case, it took the view that the producers had to be heard separately. It was foreseeable that the principal matters dealt with at such hearings would concern the pricing policy of the different producers and the production and sales position. Information of that kind is covered by business secrecy and cannot be divulged. Hence most of the producers themselves had emphasized that the hearings should be held separately. In any event, it was for each producer to state its views, so far as it was concerned, on the Commission's conclusions regarding its own price structure. There is no general right for each party to a proceeding to state its views on each and every argument put forward by the other parties to that proceeding in relation to their own competitive conduct. In any event, the reasons stated in the decision were not based on any information disclosed in a hearing of one of the producers which was used against other producers.

155.

In my view, the applicants' argument is unfounded. Even though it is not expressly provided for in Article 9(3) of Regulation No 99/63, it is for the Commission, the authority entrusted with responsibility for laying down the detailed rules of the administrative procedure, and not for the producers, to decide whether it is necessary to organize a joint hearing.

156.

However, those powers must be exercised subject to observance of the rights of the defence and, in support of its decision, the Commission may not therefore use information gathered at the hearing of one of the producers against another producer, without having shown the information to the latter first.

157.

It is by no means apparent from a reading of the contested decision that the Commission based its findings of concertation on information gathered at the hearings. ( 68 ) Moreover, the applicants have failed to demonstrate that the decision is based on information so obtained.

158.

The Court should therefore reject their arguments in that regard.

159.

According to Bowater, the Commission is guilty of an irregularity in failing to consult the Advisory Committee on Restrictive Practices and Monopolies after certain producers had given the undertaking — not signed by Bowater — thereby obtaining a 90% reduction in the amount of the fines imposed on them.

160.

Consultation of the Advisory Committee would have enabled the Commission to recognize the illogicality regarding the amount of the fine imposed on Bowater. Bowater emphasizes that it was never a member of KEA and is charged with taking part in a concerted practice for periods shorter than those established for many other producers, yet one of the two largest fines was imposed on it even though the Commission has acknowledged elsewhere that it had cooperated during the procedure. The Advisory Committee would undoubtedly have drawn the defendant's attention to that situation and the Commission would then have reduced the amount of the fine imposed on Bowater.

161.

The Commission points out that the fines were initially fixed on the basis of the opinion given by the Advisory Committee on 26 September 1984. It concedes that the Advisory Committee was not consulted again after the undertakings were agreed but before the adoption of the decision. However, it rejects Bowater's arguments regarding the consequences of the failure to consult the Committee again after certain producers had agreed to the undertaking. It considers that any illogicality could have been taken into account only with regard to the proposed reduction in the fine but could not have affected the basic amount of the fine imposed on Bowater.

162.

It does not seem to me that, so far as Bowater is concerned, the procedure is unlawful on account of the failure to consult the Advisory Committee. Since it did not agree to the undertaking, Bowater's own position remained quite unchanged compared to what it had been at the time of the Advisory Committee's initial opinion. Accordingly, so far as concerns Bowater, the Commission was entitled to adopt the decision on the basis of that opinion. The Court will not, in any event, have to consider whether it was necessary to consult the Advisory Committee again with regard to the reduction in the amount of the fines imposed on the other producers, since they have not raised the issue.

163.

It is clear from consideration of the allegations relating to procedural irregularities that the Court must uphold a number of the submissions put forward by the applicants, the foremost of which relates to the finding in the decision of concertation on transaction prices (Article 1(2)), which I strongly urge the Court to annul, as requested by the applicants. In view of the reasons tor my attitude and the extensive discussion called for by the examination of the present applications, I do not propose to consider the merits in that regard, not even on a subsidiary basis. ( 69 )

164.

It is necessary, in my view, to begin with a discussion of the concept of concerted practices. I do not share the defendant's view that ‘the case-law extracted from continental systems (referred to by certain applicants) merely serves to underline the proposition that the problem in cases of alleged concerted practices is one of proof rather than [of] definition’. ( 70 )

165.

On the contrary, as MacMillan rightly points out, a satisfactory approach in examining the concerted practices alleged in this case presupposes that that concept is defined first, before the question whether the Commission has furnished proof of such infringement of competition rules is considered.

166.

In prohibiting concerted practices, Article 85(1) of the EEC Treaty did not define that concept. ( 72 ) The Court has had the matter raised before it on numerous occasions ( 73 ) and has therefore been able to provide guidance of some value in a number of noted cases. However, I cannot dissemble the fact that there is still, in my view, some uncertainty regarding the definition of concerted practices. Quite clearly, it is necessary to proceed with caution in order to avoid general forms of words which are too constricting in areas in which the law is faced with changing and complex economic realities. Courts must not engage in academic theorizing. However, one is not succumbing to the temptation to theorize if one seeks to clarify certain concepts where the requirements of legal certainty are at stake. Traders must have a clear frame of reference to ascertain which practices are prohibited by competition rules. Besides, who would deny that the effectiveness of the law depends on its precision? Let us therefore now attempt to identify the elements which make up the concept of concerted practices.

167.

One point may already be considered established: a concerted practice is to be distinguished from a formal agreement between undertakings. In its judgments in Dyestuffs, the Court made it clear that

‘[although] Article 85 draws a distinction between the concept of “concerted practices” and that of “agreements between undertakings”... the object is to bring within the prohibition of that article a form of coordination between undertakings which, without having reached the stage where an agreement properly so-called has been concluded, ... substitutes practical cooperation between them for the risks of competition’. ( 74 )

A concerted practice ‘by its very nature, then, does not have all the elements of’ an agreement ( 75 ) and therefore constitutes a legal category ‘sui generis’, independent of the latter. In the words of one writer, that is an ‘initial delimitation of the concept of concerted practices: it presupposes a common intention which does not amount to an agreement’. ( 76 )

168.

Far more delicate is the question which then arises; what is the substance of the concertation required between undertakings? In its judgment in Suiker Unie, the Court stated that

‘the criteria of coordination and cooperation..., which in no way require the working out of an actual plan, must be understood in the light of the concept inherent in the provisions of the Treaty relating to competition that each economic operator must determine independently the policy which he intends to adopt on the Common Market’. ( 77 )

The Court added that

‘this requirement of independence... however strictly preclude [s] any direct or indirect contact between... [competitors], the object or effect whereof is either to influence the conduct on the market of an actual or potential competitor or to disclose to such a competitor the course of conduct which they themselves have decided to adopt or contemplate adopting on the market’. ( 78 )

169.

Those indicators are very broad in scope. They cover every form and method of contact between undertakings. From that point of view, the Court has quite rightly drawn the conclusions from the informality that is implicit in the very concept of a concerted practice. Similarly, the Court has considered that contacts for the joint determination and fixing of future conduct are unnecessary, once again in accordance with the specific nature of that concept in relation to that of an agreement. However, certain commentators have taken the view that the Court had widened the concept of concerted practices by including unilateral action by undertakings not capable of constituting concertation which presupposes joint discussions or the exchange of views. ( 79 )

170.

I doubt whether it is possible to restrict to those two specific situations the instances of prohibited concertation. The form of the contacts between undertakings matters little. On the other hand, it is clear to me that by definition concertation requires reciprocation of communications between competitors. ( 80 ) Article 85(1) cannot apply to unilateral action by undertakings. But that is what would happen if the requirement of reciprocal communications, ( 81 ) by whatever means, was not laid down. In its aforesaid case-law the Court does not seem to me to have laid sufficient emphasis on that requirement. There are, however, some indications in more recent judgments that seem to postulate such reciprocity.

171.

Thus, in its judgment in Zücbner, the Court stated that a national court

‘must consider whether between the banks conducting themselves in like manner there are contacts or, at least, exchanges of information’. ( 82 )

Clearly, such reciprocity of communication between competitors is something that must be defined in each individual case. At this stage, therefore, it is apparent that concertation presupposes reciprocal communications between undertakings, whatever the form. That being so, what must their purpose be?

172.

As I have said, it is not necessary that by means of such communications competitors jointly determine their future conduct. It is apparent from some of the Court's decisions that it is necessary, though sufficient, that by virtue of their communications the competitors have removed the uncertainty as to their future market conduct. Thus, in its judgment in Dyestuffs, the Court stated that

‘it is contrary to the rules on competition contained in the Treaty for a producer to cooperate with his competitors, in any way whatsoever, in order to determine a coordinated course of action relating to a price increase and to ensure its success by prior elimination of all uncertainty as to each other's conduct regarding the essential elements of that action, such as the amount, subject-matter, date and place of the increases’. ( 83 )

In its judgment in Ziichner, the Court referred to an identical requirement in connection with the exchange of information on

‘the rate of the charges actually imposed for comparable transfers which have been carried out or are planned for the future’. ( 84 )

173.

In other words, concertation must at the very least give the participants an assurance as to the conduct to be expected of their competitors; each of them can foresee the future actions of the others.

174.

It follows from those observations that the concept of concerted practices refers to reciprocal communications between competitors with the aim of giving each other assurances as to their conduct on the market. As one writer has pointed out, ( 85 ) the definition proposed by the Commission in the Dyestuffs case synthesizes those elements perfectly:

‘it is enough that they (the competitors) let each other know beforehand what attitude they intended to adopt, so that each of them could regulate his conduct, safe in the knowledge that his competitors would act in a similar fashion’. ( 86 )

175.

It is further necessary that the competitors' knowledge of each other's conduct stem from communications between them, and not simply from monitoring the market.

176.

In its case-law, the Court has been at pains to point out that

‘every producer is free to change his prices, taking into account in so doing the present or foreseeable conduct of his competitors’. ( 87 )

It has also stated that

‘this requirement of independence does not deprive traders of the right to adapt themselves intelligently to the existing or anticipated conduct of their competitors’. ( 88 )

The Court there therefore recognizes that each undertaking is entitled to align itself independendy on the conduct of its competitors, if knowledge of such conduct is obtained solely by monitoring the market.

177.

That is a vital factor having regard to the principles of economic theory. ( 89 ) Parallel conduct is not necessarily the result of prior concertation. It can be explained, or even dictated, by the very structure of certain markets. As a summary outline, let me refer to the two examples most frequently mentioned in that regard. The first situation involves a concentrated oligopoly, in which undertakings are interdependent: each undertaking must take account in its decisions of the conduct of its rivals. Alignment on each other's conduct constitutes a rational response, independently of any concertation. ( 90 )‘Price leadership’ constitutes the second situation: undertakings align themselves on a ‘price leader’ on account of the latter's power on the market. Mention may also be made of spontaneous alignment on a price leader which acts as a barometer, with its decisions reflecting changes in market conditions for reasons linked, for instance, to its previous knowledge of that market. ( 91 )

178.

In view of the findings made in the contested decision, two further sets of remarks are called for.

179.

In the first place, the question may arise whether in itself monitoring the conduct of competitors on the market falls in any event outside the scope of the rules of competition. What is the position where the decisions of each undertaking operating on the market immediately or very rapidly come to the knowledge of its competitors? In the 1970's the United States antitrust authorities instituted proceedings against conduct known as ‘price signalling’ ( 92 ) with a view to prohibiting in certain cases indirect ‘discussions’ in public between competitors or even public price announcements.

180.

In my view, the greatest caution must be exercised in calling in question the practices known as ‘price signalling’. To begin with, challenging public price announcements even when made in advance seems to me to call for very serious reservations. ( 93 ) It must be borne in mind that such announcements do meet quite legitimate and normal business requirements. Who would deny, in particular, that purchasers may in some cases need to know prices some time in advance so as to be able to determine their own costs and their own prices in order to notify those prices to their own customers? In a market economy, information which a producer communicates to his customers, whether actual or potential, is of fundamental importance. One writer has pointed to the ‘absurdity’ ( 94 ) of ignoring the fact that prices by nature have to be communicated.

181.

That, in my view, is the basic consideration which must be borne in mind in connection with ‘price signalling’. However, it would be irrational to deny as a matter of principle that concertation between competitors may take the form of exchanges of information which, whilst being public, are nonetheless reciprocal and are such as to remove any uncertainty regarding the conduct of rival undertakings. From that point of view, the difficulty clearly hes in eliciting the criteria for demarcating ‘permissible’ public price announcements from a public exchange of information which is open to criticism. In that regard, reference has been made to ‘complex, unusual and artificial’ ( 95 ) practices which, lacking any commercial justification, in fact establish a public dialogue between undertakings by giving mutual assurances as regards each other's conduct. ( 96 )

182.

If it were possible to identify the reciprocal nature of the communications in quesion, I would see no reason not to treat them as an element of concertation just because it is exchanged publicly, when the same exchanges recorded in the minutes drawn up at the end of a meeting held behind closed doors would constitute a breach of the competition rules. Let me emphasize, however, that a situation of that kind is not connected in any way with ‘mere’ price announcements which are, in principle, unilateral acts on the market and cannot therefore by themselves constitute a concerted practice.

183.

It remains to be determined, and this is my second point, what is the scope of a finding of identical conduct on the market, that is to say parallel conduct, in relation to the concept of concerted practices.

184.

In other words, is similar conduct by undertakings included as an ‘objective’ element within the concept of concerted practices? Some of my predecessors have answered that question in the affirmative. ( 97 ) However, the view has also been taken ( 98 ) that it is wrong to include parallel conduct within the actual concept of concerted practices. It is in the field of evidence that parallelism of conduct should be situated: it may point to the existence of concertation if it cannot be explained other than by concertation.

185.

The first view may be supported by reference to the actual wording of the Treaty: Does the use of the term ‘practice’ not seem to refer to actual conduct on the market? However, attention must be drawn to the scheme of Article 85(1) which prohibits not only concerted practices having the effect of restricting competition but also those whose object is to achieve that result. If parallel conduct is included within the concept of concerted practices as an ‘objective criterion’, ( 99 ) how can the prohibition in Article 85(1) extend to types of concertation which, albeit not implemented, nevertheless pursued a restrictive object? Accordingly, to take the view that de facto identical conduct forms part of the concept of concerted practices would lead to a particularly restrictive conception of the Treaty, which is contrary to the scheme of Article 85(1).

186.

The major obstacle here was clearly identified by Advocate General Mayras when he pointed out that:

‘There can be no doubt that it would seem curious for a concerted practice which has not had any material effect on the competitive situation, despite the intention of the participants and because of circumstances beyond their control, to escape the application of Article 85’. ( 100 )

Yet that is the conclusion which follows from the inclusion within the concept of concerted practices of de facto common conduct which was the view held by my predecessor. In order to overcome the resultant obstacle, he suggested that

‘in such a case merely to attempt or to initiate execution would be enough to justify the application of Article 85(1)’. ( 101 )

Hence the need to have recourse to categories of criminal law, which are alien to the concepts of Community competition law, reveals in my view the impasse resulting from the inclusion of an ‘objective criterion’ derived from the parallelism of undertakings' conduct in the definition of a concerted practice.

187.

No doubt the ambiguity of the term ‘practice’ lies at the root of this discussion. I believe, however, that, in the light of the scheme of Article 85 the scope of the concept of concerted practices is quite clear: a concerted practice does not refer to identical conduct by undertakings. In fact, within that concept, in my view, the Treaty distinguishes between and contrasts de facto concertation or concertation in practice, on the one hand, and formal concertation resulting from an agreement, on the other. ( 102 ) Ultimately, the Treaty precludes cooperation amongst traders in whatever form: ‘Article 85 does not rule out uniformity of conduct but only certain ways by which it is achieved’. ( 103 )

188.

I therefore share the view that ‘mere concomitant conduct does not constitute a concerted practice but may at best point, on the basis of further evidence, to the existence of an agreement between the parties concerned’. ( 104 )

189.

No doubt, certain forms of words used by the Court in its case-law might lead to the assumption that the Court regards parallel conduct as constituting in itself an objective criterion for establishing a concerted practice. ( 105 )

190.

In Dyestuffs, however, the Court seemed to consider, at least as regards only the principles which it laid down, that it was in connection with evidence that the scope of parallel conduct had to be assessed, stating that although such conduct

‘may not by itself be identified with a concerted practice, it may however amount to strong evidence of such a practice if it leads to conditions of competition which do not correspond to the normal conditions of the market, having regard to the nature of the products, the size and number of the undertakings, and the volume of the said market’. ( 106 )

191.

In any event, it must be emphasized that parallel conduct does not in itself ( 107 ) constitute proof of a concerted practice. From parallel conduct it is possible to deduce evidence of concertation where it cannot have arisen as a result of individual and rational decisions taken by the undertakings concerned having regard to the characteristics of the market in question. ( 108 )

192.

Those principles, it may be worth noting, are ultimately no different from the solutions which seem to emerge from the development of United States case-law. Although the exceptional wealth of that case-law precludes a detailed exposition of it here, ( 109 ) I should like to draw attention to one crucial stage in its development. In 1948, relying it would seem on certain dicta of the Supreme Court, ( 110 ) the Federal Trade Commission expounded the view that where several undertakings engage in identical conduct in the knowledge that they are all acting in the same manner, there is an ‘agreement’ between them. In other words, the infringement of antitrust legislation was established by the mere fact of ‘conscious’ parallel conduct.

193.

The Supreme Court unequivocally rejected that analysis in the following terms:

‘To be sure, business behavior is admissible circumstantial evidence from which the fact finder may infer agreement... But this court has never held that proof of parallel business behavior conclusively establishes agreement or, phrased differently, that such behavior itself constitutes a Sherman Act offense. Circumstantial evidence of consciously parallel behavior may have made heavy inroads into the traditional judicial attitude towards conspiracy, but “conscious parallelism” has not yet read conspiracy out of the Sherman Act entirely’. ( 111 )

194.

Whilst reaffirming ( 112 ) the possibility of establishing a ‘conspiracy’ on the basis of circumstantial evidence, the Supreme Court quite clearly refuses to regard parallel conduct in itself as proof of an infringement of antitrust legislation. The probative value of parallel conduct falls to be determined in each individual case by a specific assessment.

‘“Conscious parallelism” is not a blanket equivalent of conspiracy. Its probative value in establishing the ultimate fact of conspiracy will vary case by case. Proof of agreement, express or implied, is still indispensable to the establishment of a conspiracy under the antitrust law.’ ( 113 )

So far as concerns Article 85 of the EEC Treaty, parallel conduct cannot be equated with a concerted practice either. On the other hand, however, such conduct may serve as evidence of concertation.

195.

In that regard, since recourse to indirect forms of evidence is involved, ( 114 ) caution ( 115 ) must be exercised in establishing the existence of a concerted practice. According to the Court's case-law, the evidence must be ‘sufficiently precise and coherent... to justify the view that the parallel behaviour... was the result of concerted action’. ( 116 ) In other words, it is necessary to establish a degree of certainty that goes beyond any reasonable doubt. In accordance with the principles governing the burden of proof, it is for the Commission to demonstrate that: the burden of proof cannot be shifted simply by a finding of parallel conduct. Unless the court can be satisfied by a set of presumptions having a solid basis, concertation is not established. In any event, if there is a plausible explanation for the conduct found to exist which is consistent with an independent choice by the undertakings concerned, concertation remains unproven.

196.

That is the principle which can be elicited from the Court's judgment in CRAM and Rheinzink. ( 117 ) The Commission had established the existence of a concerted practice by those two undertakings which had interrupted in quick succession their deliveries to a single producer who had re-exported the products concerned to the German market, contrary to the terms of the contracts of sale concluded with the suppliers concerned. CRAM and Rheinzink explained the reasons which allegedly prompted each of them individually to interrupt its deliveries and the Court held that

‘it is sufficient for the applicants to prove circumstances which cast the facts established by the Commission in a different light and which thus allow another explanation of the facts to be substituted for the one adopted by the contested decision’. ( 118 )

197.

Of course, the existence of direct evidence of concertation may be conclusive by itself and, in circumstances in which it is sufficient, it may even render devoid of purpose recourse to indicia drawn from parallel conduct. ( 119 ) In this case, however, the Commission treats the analysis based on parallel prices as primary evidence, since the documents referred to in the decision play a secondary role, merely providing, as the defendant puts it, ‘independent support’.

198.

The foregoing observations were designed to pave the way for the examination of the facts of the case.

199.

It must be borne in mind from the outset that, according to the grounds stated in the decision, concertation took place either between all the addressees of the decision, or between addressees located in the same country or the same continent, or between individual addressees. ( 120 ) I have already stated that since there is no express reference to this point in the operative part, the Court has been obliged to ask the defendant to specify its exact findings of concertation. On the basis of the lists communicated by the Commission, its approach can be summarized as follows: the Commission has concluded that a producer participated in concertation where he announced (or charged) a price identical to that of some of his competitors for a given product in a given area and for a given quarter. That is why the concertation which the Commission regards as established seems to be something of a ‘floating’ phenomenon. The Commission states, moreover, in its rejoinder, that the question is not whether intermittent participation in a concerted practice falls within the scope of Article 85(1), which it clearly does, but whether intermittent parallel conduct by undertakings constitutes sufficient evidence of concertation. The Court must therefore bear in mind that the fact that a given producer announced a price identical to that of his competitors constitutes the primary evidence on the basis of which his participation in concertation has been established in the decision.

200.

Further, I wish to express certain doubts regarding the connection between the findings of ‘general’ concertation on announced prices (Article 1(1) of the decision) ( 121 ) and those set out in Article 1(3) and (4). In paragraph 109 of the decision, the Commission considers that the exchange of information between certain United States producers within the framework of KEA and between producers of hardwood pulp within the framework of Fides was not only part of the concertation on prices but also constituted independent infringements of Article 85(1). That analysis prompts a number of observations.

201.

In the first place, the Court will find that the principle of non bis in idem seems to be disregarded as regards concertation between the producers who are members of KEA. Article 1(3) of the decision charges them with concerting on announced prices and transaction prices for bleached sulphate wood pulp within the framework of KEA without specifying, however, the period when such concertation took place. In paragraph 35 of the decision, however, the Commission points out that for most of the period from 1975 to 1981 the prices announced by the members of KEA were identical, as were those announced prices and the transaction prices of those producers for part of that period. If reference is made to the lists submitted to the Court by the defendant in describing its findings of ‘general concertation’ in Article 1(1) and (2) of the decision, on the other hand, the Court will find that concertation amongst members of KEA also seems to be at issue in that regard. In other words, concertation on prices between KEA producers seems to have been envisaged twice.

202.

Is there any need to emphasize that self-evidently the same facts cannot in any event lead to the same infringement being established twice over?

203.

With regard to Article 1(4) of the decision, the English version, which alone is authentic, states as follows: ‘Addressees 4, 29, 34, 36 and 40 to 43 by concerting on announced and actual transaction prices and exchanging within the framework of Fides individualized data concerning prices for deliveries of bleached sulphate hardwood pulp to the European Economic Community from 1973 to 1977’. There is already an ambiguity, to which I shall return, as to whether only concertation within Fides is meant in that section of the operative part. Even if that is the case, however, does not such concertation overlap, at least in part, in the case of the producers concerned, with some of the findings in Article 1(1) and (2) of the operative part?

204.

In the light of those observations, I now turn to the parties' arguments. The Court will recall that the defendant's position, as is apparent from the decision and from the further information provided at the Court's request, is as follows: evidence of general concertation on prices stems primarily from the producers' parallel behaviour which can be explained only by a concerted practice between them (1). That evidence is corroborated by various kinds of exchanges of information, which are mentioned in the decision (2).

205.

Paragraph 83 of the decision calls in question both similar price announcements for similar periods, at short intervals or even simultaneously, and identical transaction prices in support of the rinding that parallel conduct of that kind can be explained only by a concerted practice on the part of the addressees. Let me point out first of all that under this heading I propose to consider only those aspects of the decision relating to concertation on announced prices. I therefore propose to ‘isolate’ parallel conduct relating to those prices.

206.

I shall tackle the question as follows: first, I shall set out my position on actual parallelism of announced prices (1.1). I shall go onto deal with the aspects of the system of price announcements (1.2). I shall then attempt to determine what guidance can be elicited from the discussion concerning the economic analysis of parallel prices in the light of the expert's report commissioned by the Court (1.3). Finally, I shall explain why, in my view, the contested decision is in any event seriously defective as regards the provision of an individual statement of reasons for the infringements which each addressee of the decision was found to have committed (1.4).

207.

The Commission sets out its findings on parallelism of announced prices in paragraph 22 of its decision, stating that the addressees of the decision announced the same prices for northwest Europe (zone 1) and almost the same prices for southern Europe (zone 2) for the following periods:

208.

However, those findings are qualified significantly by three footnotes to the paragraph in question. The first footnote states that ‘in the second quarter of 1979 and in the second quarter of 1980 the uniform prices quoted by the Canadian firms were USD 5 higher than the uniform prices quoted by the US firms. In the third quarter of 1979 the difference was USD 10’. That footnote is, however, contradicted by Table 6: in the second quarter of 1979, the prices announced by the Canadian producers are not USD 5 higher than the prices of the United States producers but lower (softwood, zone 1). Even if they do not substantially affect the Commission's findings, such errors can but be deplored.

209.

According to the second footnote, ‘in the first quarter of 1975 the prices announced by Finncell were the same as those announced by the American firms. From the second to the fourth quarter of 1975 Finncell announced an alternative price of USD 450’. ( 122 ) However, it is appropriate to supplement that footnote by reference to paragraph 112 of the decision which states that Finncell ‘in the last three quarters of 1975 announced a price in US dollars and in Swedish kronor, the dollar price being at times higher and at times lower than that announced by Swedish suppliers in Swedish kronor...’. ( 123 ) That reference thus qualifies significantly the Commission's general statement in the second indent of paragraph 22 that from the first quarter of 1975 to the second quarter of 1977 the prices announced by the Swedish and Finnish producers were the same.

210.

The third and final footnote to paragraph 22 reads as follows: ‘In the third quarter of 1979 and in the second quarter of 1980 only the prices announced by the Canadian and European firms were the same. The US firms uniformly quoted prices that were respectively USD 10 and 5 (i. e. 1 to 2%) lower’, which, moreover, is borne out by the figures in Table 6.

211.

A detailed study of that table reveals, however, that many other footnotes would have been necessary in order to give an accurate reflection of the data set out therein. For example, again as regards the fourth quarter of 1979, the prices announced by the Canadian producers are USD 10 lower than the prices quoted by the Swedish and Finnish producers (softwood, zone 1). The same applies for the first quarter of 1980. In addition, from the second quarter of 1980 to the second quarter of 1981, the prices announced by the United States producers are USD 5 lower than the prices announced by the other producers for northern softwood (zone 1). In the third quarter of 1981, the difference between the Canadian prices and the Swedish and Finnish prices was USD 45 (zone 1, softwood), as is clear from Table 6. Accordingly, the Commission's assertion in the third indent of paragraph 22 of its decision that prices were absolutely identical from the third quarter of 1979 to the third quarter of 1981 by no means tallies exactly with its own findings.

212.

Therefore, while both the existence of the footnotes, which are moreover incomplete or inaccurate, and the analysis of Table 6 annexed to the decision, show a ‘single’ parallelism between all the producers for certain periods (particularly from the first quarter of 1976 to the third quarter of 1977 for softwood, zone 1), for other periods they reveal a number of price parallelisms whose pattern may be described as ‘complex’.

213.

The Commission has informed the Court that the primary evidence of a given producer's participation in concertation stems from the finding that he announced a price identical to that of one or more competitors in a given quarter. That approach therefore explains the statement in paragraph 81 of the decision that ‘this concertation took place either between all addressees of this decision, or between addressees located in the same country or the same continent [or] between individual addressees’. Let me point out at this juncture, however, that the operative part of the decision does not contain any indication of the precise methods of such concertation.

214.

I shall shortly consider the arguments of some of the applicants who challenge, for their part, the findings in Table 6. I must point out beforehand that both the first expert's report and the explanations given by the experts when they were heard by the Court on 5 June 1990 confirm, with one exception, the existence of parallelism as described by the Commission in that table ( 124 ) annexed to its decision; that exception stems, as stated on page 21 of the expert's report, from the fact that the information relating to the prices announced for the beginning and the middle of 1978 seems to be missing. In reply to a written question from the Court, the experts stated that there was no information regarding announced prices at the beginning of 1978 and in the summer of that year in the Commission's documentation.

215.

Apart from that qualification, the experts confirmed at the said hearing that they supported the conclusion, set out in paragraph 22 of the decision, which the Commission draws from the information in Table 6 regarding the existence of ‘parallelism’, the configuration of which, as we have seen, is variable.

216.

The Commission's findings regarding the existence of such ‘parallelism’ are not fundamentally called in question by the omissions and inaccuracies in Table 6 which were revealed during the written procedure as a result of the objections raised by certain applicants. Nevertheless, the evidence for the prices announced by certain individual producers for certain periods gives rise to serious misgivings.

217.

For instance, the members of Finncell claim that the Commission's investigation concerning Finncell's announced prices related only to the period from 1974 to 1977. The prices set out in Table 6 for the period after 1977 were obtained simply by imputing to the members of Finncell the prices established for the Swedish undertakings.

218.

The answer given by the first experts to a written question from the Court on that point is quite clear: the Commission's documentation did not contain any invoice or telex concerning the prices announced by the members of Finncell for the period after 1977.

219.

Accordingly, one can but surmise on what basis the Commission determined the prices announced by the members of Finncell for the latter period. The Commission, in its rejoinder, refers to the accounting documents, without further details, forwarded by the members of Finncell in response to the Commission's request under Article 11 of Regulation No 17, dated 1 September 1982, and to the telex messages listed in paragraph 61 et seq. of the decision.

220.

It is apparent on reading the aforesaid paragraphs of the decision that those telex messages all relate to the period prior to 1978. As for the Commission's request for information of 1 September 1982, it relates essentially to the prices charged, even though clarification is sought on any differences between those prices and the announced prices.

221.

It is worth noting, however, that whilst for the last three quarters of 1977 the members of Finncell relied on Table 6, which shows prices different from those of the other producers, the Commission has produced, in an annex to its defence, telex messages from Finncell showing that the figures in that table were the result of a ‘misunderstanding’ and that the members of Finncell had in fact announced the same prices as their competitors in respect of that period.

222.

It would have been helpful if, for the period subsequent to 1977, the Commission had likewise produced the documents on which it relied in determining the prices announced by Finncell. The figures in Table 6 are not in themselves sacrosanct; they merely serve as an indication. Since they are disputed by the applicants, it is for the Commission to produce before the Court the documents on which they are purportedly based, in order to demonstrate the accuracy of its previous findings.

223.

The Commission explains, in the general part of its defence, ( 125 ) which comprehensively rejects the arguments put forward by all the applicants, that it relied, in gathering the data recorded in Table 6, partly on telexes and other documents submitted to it during the investigation under Article 14 of Regulation No 17, partly on information given by the addressees of the decision in their replies to the Statement of Objections, and partly on articles published in Pulp and Paper International (PPI) and in PPI Newswire, and that as regards the first two sources the applicants had an opportunity to make their views known. According to the Commission, the information from the trade press was used only to rebut the applicants' argument that they had not announced prices for a given period. Hence the Commission would seem to acknowledge by implication that the last-mentioned information was not communicated to the applicants during the administrative procedure.

224.

It must be stated that, even though the information from the trade press served merely to rebut the argument based on the absence of price announcements, the use of such information nevertheless constitutes part of the evidence on which the Commission relied in support of its standpoint. It seems, moreover, that such information was also used to determine the announced prices, as is clear from the words ‘with effect from 3rd quarter 1978, the prices given have been taken from PPI’ in Annex VI to the Statement of Objections, at page 20, ( 126 ) and from the experts' reply to the written question from the Court, to the effect that the Commission did not have any telexes in respect of the prices announced by Finncell after 1977.

225.

That is borne out by the fact that whilst the defence to the action brought by the members of Finncell is silent in that regard, there is an annex to the defence to the action brought by the members of KEA containing six passages from trade journals which refer in places to the changes in Finncell's announced prices. It is possible, on the basis of an analysis of those passages in particular to verify the accuracy of Table 6 so far as concerns the prices announced by Finncell for the first and second quarters of 1980 and the third quarter of 1981. But what about the other periods? It would certainly have been of value for the Court to be able to carry out the same verification. It must be noted that, in any event, the information relied on by the Commission in establishing the prices announced by Finncell after 1977, which does indeed seem to be from the trade press, has not been communicated to the Court. Accordingly, there remain serious doubts regarding the relevant data in Table 6.

226.

Other shortcomings in the decision as to the evidence for the prices announced by certain applicants have been acknowledged by the Commission. Thus, so far as concerns Bowater, the defendant conceded that after rechecking the documentation, it was unable to trace any written evidence for the finding in Table 6 that in 1977 Bowater announced a price of USD 415. ( 127 ) Moreover — a point to which I shall return — the Commission was unable to contradict Bowater's statement concerning the dates of its price announcements for several quarters in the period from 1979 to 1981, which do not correspond to those cited in Table 6. Admittedly, it is conceivable that, having gathered its information from the trade press, the Commission found itself confronted with imprecise details concerning the date on which a given producer or group of producers intended to announce a price which, for its part, is recorded correctly. ( 128 ) The fact remains that given the failure to produce any documents whatsoever, including mere extracts from trade journals, doubts persist — yet again — with regard to the information on which the Commission relied in determining the prices announced by that producer for the period from 1979 to 1981.

227.

It would have been helpful, moreover, if the Commission had replied to Weldwood of Canada's allegation — which, in my view, is well documented — to the effect that the price of USD 415 for the first quarter of 1975, which is set out in Table 6, is incorrect. In its application, Weldwood claims to have demonstrated during the administrative procedure, in particular by producing the relevant letters and telexes, that the price of USD 425 was maintained throughout the first and second quarters of 1975. The Statement of Objections refers, moreover, for the first quarter of 1975, to a price of USD 425 in so far as concerns Weldwood. That must no doubt be a clerical error in the decision.

228.

Although the members of KEA do not formally challenge the figures set out in Table 6, they call in question the very existence of parallel prices in so far as they are concerned for the period from 1978 to 1981.

229.

It must be borne in mind that the concerted practice which should be at issue here is that allegedly established between the members of KEA and the other producers — in particular Canadian or European producers — not forming part of that association, since concertation on announced prices within the framework of KEA constitutes a separate infringement which is referred to in Article 1(3) of the decision. However, it is apparent on reading Table 6 that there is practically no quarter when the prices announced by the members of KEA differed from those announced by all the other producers. Thus, although after 1979 the prices announced by the members of KEA frequendy differ by USD 5 or 10 from those announced by the other European and Canadian producers (for instance, for softwood, zone 1: in the second quarter of 1979, USD 425 for the members of KEA, USD 420 for the Canadian producers and USD 435 for the Scandinavian producers; in the third quarter of 1979, USD 425 for the members of KEA and USD 435 for the Canadian and Scandinavian producers; in the fourth quarter of 1979, where the price of only one member of KEA is mentioned, it is USD 5 lower than the Canadian price and USD 15 lower than that of the Scandinavian producers; see also the second and third quarters of 1980), in respect of the same periods the members of KEA generally quote the same prices as those which Table 6 ascribes to Bowater, ( 129 ) a United States producer not forming part of that association (for instance, in the second and third quarters of 1979, USD 410 for southern softwood, zone 1, and USD 425 for zone 2; the same holds true for the second and third quarters of 1980: USD 530 for southern softwood, zone 1).

230.

The applicants' reasoning with regard to the difference between their announced prices and those announced by the Scandinavian or Canadian producers tallies with my previous findings that, for certain periods, the parallelism outlined by the Commission in Table 6 covers several parallelisms having variable configurations, according to the quarters in question.

231.

The members of KEA point out, moreover, that they have absolutely no idea where the Commission obtained its information and that, in any event, such information was not communicated to them during the administrative procedure.

232.

On those two points, the Commission explained in the general part of its defence, ( 130 ) as I said earlier, that it relied, in compiling Table 6, on telexes and other documents communicated during its investigations, on information given by the producers in their replies to the Statement of Objections, and finally on information from the trade press which, let me repeat, was used, according to the Commission, only to rebut the argument based on the absence of price announcements.

233.

In their reply, the members of KEA do not seem specifically to repeat their criticisms of the Commission's failure to communicate the information which enabled the amount of their announced prices to be determined. In any event, since they do not actually challenge the announced prices set out in Table 6, the arguments adduced here would scarcely seem to apply so far as concerns proof of the existence of parallel prices.

234.

Without calling in question the data in Table 6, St. Anne points out that in fact the prices referred to in that table show that, for the third and fourth quarters of 1976, it announced a price appreciably lower than that of its competitors (USD 395 for hardwood, zone 1, when the members of KEA were announcing a price of USD 405, and the Finnish and Swedish producers a price of USD 410). The Commission replies that, on the other hand, for zone 2 (Mediterranean ports), St. Anne announced the same prices as the United States producers, although it acknowledges that St. Anne's sales in Italy accounted for no more than 5.5% of Community imports in 1976. That fall in St. Anne's sales in the southern zone should, moreover, have led it, according to the Commission, to lower its prices for that zone.

235.

In Article 1(1) of the operative part of the decision, the Commission charges St Anne with concerting on announced prices for 1975 and 1976. Having regard to the data in Table 6, it follows that the alleged concertation for the last two quarters of 1976 applies only to the southern zone, given the absence of truly identical prices for the northern zone.

236.

Finally on this point, it is self-evident that if the data in Table 6 cannot be substantiated in relation to one producer, that would, in my view, ipso facto be a ground for annulling the decision in relation to that producer with regard to the period in question since the Commission considers that it is the existence of identical announced prices that constitutes the primary evidence of a producer's participation in general concertation on price announcements.

237.

In any event I consider that the Court will be required to draw the appropriate conclusions from the absence of proof or from the errors in the decision in that regard only if it decides not to follow my Opinion, expounded below, as to the inadequacy of the reasons stated in the decision relating to each applicant's individual participation in a concerted practice.

238.

Let me begin with a brief outline of the mechanism of quarterly price announcements. In the weeks before the beginning of a quarter, the producers announce the price which they wish to obtain in the following quarter. Long-term supply contracts entered into with customers generally provide for the fixing of prices on a quarterly basis. Those prices are communicated to the producers' customers or agents. It is common ground that the trade press swiftly publishes the prices announced by the producers. It is true, according to the producers, that they quickly learn of the prices announced by their competitors either through their customers or from the press.

239.

The first question which arises concerns the function of announced prices. Already in my introduction I observed, as was also stated in the Report for the Hearing, which has not been contested on this point by the Commission, that announced prices constitute a maximum or ceiling price throughout the period in which they are valid. A number of applicants emphasize that price announcements are designed to give purchasers the assurance that prices will not exceed the announced level during the quarter in question whilst individual prices will be negotiated on that basis. In its defence, however, the Commission states that the argument that announced prices merely serve as a ceiling for actual prices and as an opening bid in the bargaining with each customer is clearly without foundation, and points out that for the third quarter of 1979 some United States producers announced a price of USD 425 for northern softwood although the product was sold at USD 435, being the price announced by their Canadian and European competitors.

240.

In that regard, I wish to make two points. In the first place, having analysed the producers' prices covering a period of some seven years, it strikes me as extremely unconvincing to deny that announced prices serve as a ceiling price on the basis of the findings relating to a given quarter for a few producers. Secondly, as I have pointed out, it is apparent from the statements made by some pulp purchasers which are now before the Court that, in their view, announcements serve as a ceiling and that costs are thus assured for the period in question. ( 131 ) Those documents have not been discussed by the Commission in its pleadings before the Court.

241.

The decision is ambiguous with regard to the actual assessment of the system of price announcements. In places, the Commission seems to consider that the mechanism of quarterly announcements constitutes an aspect of concertation on prices. Thus, in paragraph 108 the Commission states that it constituted ‘at the very least’ an indirect exchange of information on future market conduct. ( 132 ) Elsewhere, it seems to regard the details of the system and, in particular, the simultaneity of announcements, as evidence of concertation between the producers at an earlier stage. ( 133 ) The Court did indeed request the Commission on two occasions to make up its mind whether it regarded the system of quarterly announcements as an aspect of concertation on announced prices or as evidence of concertation at an earlier stage. The information provided by the Commission has by no means dispelled all the uncertainty. In its initial reply, the Commission essentially stated that the system of announcements was both an important part of the concertation ‘tending to reinforce the concertation on the common announced price’ and, on account of its nature, timing and form, evidence of concertation between producers at an earlier stage. The Court then asked the Commission to indicate the precise scope of that analysis. In its second reply, the Commission maintains that the system of price announcements constituted evidence of an infringement for two reasons. First of all, certain aspects of the system (in particular, simultaneous announcements) were not dictated by market conditions. Secondly, it formed part of the pattern of exchange of price information, ensuring that the announced prices of each producer — ‘and thus any departures from the common announced price’ — would be known to all other producers before they were implemented. Hence the system of quarterly announcements constituted an ‘important part’ of the concertation on announced prices by ensuring artificial transparency whereby each of the producers concerned could ensure that the others were abiding by the ‘common’ price.

242.

It is precisely the term ‘common’ price which masks all the ambiguity inherent in the defendant's position. Is that ‘common price’ the result of concertation between the undertakings prior to the announcements themselves, whereby the system of announcements constitutes the means for monitoring compliance with such concertation? Or does the system of price announcements constitute the machinery of concertation for fixing that common price?

243.

The lack of clarity in the Commission's position is unfortunate. It was not possible to resolve it at the hearing since the Commission referred both to concertation prior to the announcements, suggesting that an earlier agreement on prices made it pointless to settle on which undertaking was to be the first to announce its prices, and to an ‘unspoken understanding’. ( 134 )

244.

In any event I have been unable, notwithstanding the many hours spent in studying this bulky file, to grasp clearly the precise function played, according to the Commission, by the system of price announcements within the framework of concertation. The defendant's position in this case has the consistency of mercury. Just as one is about to grasp it, it eludes one, only to assume an unexpected shape.

245.

I do not propose to analyse further the structure of the Commission's position regarding the features allegedly exhibited by concertation in this case. I must simply emphasize that the Court's assessment of the findings in the decision will be severely hindered as a result. In any event, it seems to me to be more appropriate for the sake of clarity to discuss that aspect of the position taken in the contested decision from the following angles:

246.

It is appropriate to begin by calling to mind the whole of the Commission's analysis which is contested by the applicants. According to the decision, although the large number of competitors should theoretically have made concertation between producers more difficult, that difficulty was neutralized by the transparency of prices on the market. However, there was nothing inevitable about that transparency. It was the result of behaviour deliberately chosen by the producers who were in constant direct or indirect contact with one another. The market was made artificially transparent simply by the fact that the prices quoted by the producers were made known, published by the trade press or passed on to agents acting for a number of producers so early that other producers were able to react even before those prices came into force. Still according to the decision, if the producers had disclosed their prices only to the potential customers actually concerned, it would have been impossible for them, given the large number of customers and the wide range of products on the market, to obtain a rapid and accurate picture of the prices quoted by their competitors. That system constituted in itself at the very least an indirect exchange of information on future market conduct. That applied particularly where prices were made known by the producers themselves by being given to the trade press for immediate publication or by being passed on to agents acting for other producers. In such cases, the producer could expect the prices he announced to reach his competitors, just as he himself would expect to be given details in the same manner of his competitors' prices. In transmitting that information, therefore, the producers made use of third parties (such as the press, agents etc.). The fact that prices were published well in advance of their entry into effect at the beginning of a new quarter guaranteed that producers had sufficient time to announce their own — corresponding — new prices before that quarter and to apply them from the beginning of that quarter. In any event, the producers were given the information in time.

247.

That analysis is contested by the producers. They claim, in the first place, that the transparency of the pulp market stems from certain features which the Commission has disregarded and that, in particular, the Commission underestimated the role played by customers in the dissemination of information. They also observe that the defendant wrongly described the communication of prices to common agents or their publication in the trade press as an exchange of information between competitors.

248.

Before I consider those arguments, there are a number of observations I wish to make.

249.

In the first place, it must be stated that in no circumstances can the behaviour of undertakings on a market, which comes to the knowledge of their competitors, constitute in itself an exchange of information amounting to concertation. That would be tantamount to calling in question by means of Article 85 the transparency of certain markets. However, undertakings cannot be prevented from having recourse to public and reciprocal exchanges of information as regards their future conduct. In that respect, in view of the requirements of legal certainty, a criterion must be laid down enabling situations of that kind to be categorized.

250.

The concept of artificial transparency of the market adopted in the decision strikes me as being far from ‘reliable’ from that point of view. Emphasis must be laid, in my view, less on the artificial nature of transparency than on the artificial nature of the conduct of the undertakings themselves. In other words, as I said earlier, if they engage in conduct which is justified by commercial needs, I do not see on what grounds it can be challenged.

251.

If, on the other hand, transparency is the result of practices which do not correspond to rational needs in the light of the characteristics of a given market, the possibility cannot be ruled out that the undertakings' conduct may cease to constitute behaviour on the market directed at customers and may amount to an exchange of information. In that regard, it should be borne in mind that what is at issue in this case are ‘simple’ price announcements.

252.

Secondly, I would observe that the fact that prices are known in advance seems, according to the Commission, to constitute one of the aspects which enables an exchange of information to take place. As I have pointed out, however, prior price announcements correspond to specific commercial needs. Purchasers may, as I have said, wish to know in advance the cost of the product concerned for the purpose of making their own cost forecasts, determining their future prices, and communicating them to their own customers. Clearly, those needs will assume all the greater significance where the product concerned accounts for a major proportion of the purchaser's costs.

253.

That is precisely the case on the wood pulp market. In the first place, that product accounts for 50-70% of the cost of papermaking. The need for purchasers to know prices in advance is clearly borne out by some of the documents before the Court. For instance, according to a purchaser who describes himself as the most important individual purchaser in the Community, ‘it is certainly necessary for the papermakers that the increases in the price of pulp be announced well in advance, so as to have sufficient lead time to implement their own price increases on their own market’. ( 135 ) According to others, ‘we also look for agreement of the pulp price well ahead of its application, to provide a basis for planning our paper sales strategy’ ( 136 ) or ‘any change in pulp prices should be announced by the supplier long in advance’. ( 137 ) Those statements are quite unambiguous. In my view, they are of fundamental importance: the fact that prices are announced in advance does not a priori show that there is anything abnormal about the wood pulp market. That leads me to view with considerable caution the Commission's appraisal, according to which price announcements constitute in themselves a system for exchanging information.

254.

I now turn to the issue of the transparency of the market which the Commission regards as the ‘artificial’ result of the system of price announcements.

255.

According to the Commission, it should be recalled, the communication of announced prices to potential customers alone would not have enabled the producers to gain a rapid and accurate picture of those prices, in view of the range of products and the number of customers. In its view, especially in times of rising prices, ( 138 ) customers have no interest in informing producers of their competitors' prices. ( 139 )

256.

Consideration of the applicants' observations leads me to conclude that the Commission underestimated certain characteristics of the market for wood pulp, to which I know turn.

257.

The producers have drawn attention to the importance of vertical integration in the wood pulp and paper industry. Several of the producers involved in this case belong to groups which also manufacture paper and purchase pulp from competing producers for their own needs. That aspect has, moreover, been referred to in the decision.

258.

The Commission has not denied that although the number of potential purchasers is extremely large, the number of very important purchasers is relatively small. Thus, according to the figures supplied by the Canadian applicants — which were not disputed by the defendant — in the mid-1970's 17 purchasers accounted for 50% of the purchases made and 41 accounted for 80% of all transactions. Nor is it contested, or open to challenge at first sight, that those customers are very important to the producers, who have every reason to remain in constant touch with the papermakers concerned. ( 140 )

259.

The Commission has itself stated that paper mills diversify their purchases by buying from different producers. A single purchaser may therefore know the prices of several producers. In that regard, the figures set out in the common part of the application submitted by the producers from British Columbia are very instructive. ( 141 )

260.

On this point, it is sufficient to refer to the very vivid description given at the hearing by Mr Cockram. He pointed out, as the Court will recall, that direct contact concerning similar pulp products could take place between undertakings which were not necessarily situated in the same country. He also emphasized that information on any changes in price spreads very quickly through the industry and is passed on in a matter of hours rather than days. Mr Cockram pointed out that the purchasers also ‘talk’ to agents acting for the producers.

261.

At the very least, information relating to announced prices seems to spread very quickly and widely amongst the purchasers. The Court will find in that regard that transparency of announced prices for pulp is considered desirable by the purchasers themselves, if reference is made to the documents produced during the proceedings. ( 142 ) Although the Commission does not seem to have challenged those different characteristics, it has maintained in the proceedings that, particularly in times of rising prices, customers have no interest in letting a producer know the prices announced by his competitors. The applicants have vigorously challenged that contention on the ground that, even at such times, customers are likely to inform a producer in order to persuade him to announce a moderate price or in order to maintain good business relations.

262.

First of all, it should be noted that the Commission does not seem to challenge the view that when announced prices are falling, or even holding firm (which was the case from 1975 until the end of 1978), purchasers may inform producers of the prices announced by competitors.

263.

Secondly, there is clearly no doubt that paper mills which are linked to producers will disclose the prices announced by other pulp manufacturers, whatever the trend with regard to prices.

264.

Finally, and above all, documents were produced during the proceedings showing that even in times of rising prices a producer may find out his competitors' announced prices from his own customers. The file on the case contains documents which flatly contradict the Commission's assertions. Thus, IPS which, like the other applicants, claims to obtain from its customers information on the prices announced by its competitors, has annexed to its application ( 143 ) under the heading ‘A representative pricing decision’ ( 144 ) several telex messages from its sales offices ( 145 ) in Europe ( 146 ) which related precisely to a period of rising prices, namely the end of 1979. Those documents are quite unambiguous: during that period of rising prices, IPS was very rapidly apprised of some of its competitors' price announcements as a result of information supplied by customers to its own sales offices. If the Court refers to those documents, it will find that IPS was able to take cognizance of a rival producer's announcement within two days, one day, or even on the day it was made. What is more, in certain cases, a customer was able to disclose that a competitor's announcement was imminent, or would be made, for instance, on the following day.

265.

It cannot be said that too much weight was attached to those documents. The defendant, which, according to IPS, had nevertheless known about them since the Statement of Objections, did not dispute either their contents or their authenticity: they are not even referred to in its defence or rejoinder. It was for the defendant to demonstrate how it was able to reconcile the information yielded by the documents in question with its categorical assertion that ‘there is no valid explanation of how such information spread so rapidly (i. e. within a matter of days, in some instances, on the same day) between such a large number of firms’ (paragraph 89 of the decision). Accordingly, the statement in the decision that the disclosure of information only to customers would not permit the producers to gain a rapid and an accurate picture of the prices quoted by their competitors is, to say the least, far from proven.

266.

It remains to consider the parties' arguments relating to the role of common agents and the trade press.

267.

According to the decision, the communication by a producer to common agents of its announced prices contributed to making the market ‘artificially’ transparent and constituted a flagrant example of the indirect exchange of information between producers. The producers concerned claim that the Commission's argument is contradictory. It had initially taken the view in the Statement of Objections that the activities of common agents constituted restrictions of competition within the meaning of Article 85(1). It then stated that it intended to terminate the proceeding.

268.

The defendant disputes that argument and points out that the fact that it ‘has not proceeded with certain objections involving common agents does not mean that the Commission concedes that common agents cannot be used by the producers as a means of exchanging information as to their respective prices. It was, however, sufficient to bring an end to the concertation between the producers without putting an end to the legal activities of the common agents’. ( 147 )

269.

I am not swayed by that contention. The Commission considers that the activities of common agents may be lawful. The reasons which led it to take that view may be debatable. ( 148 ) It must be recognized, however, that if agents may, without infringing the rules of competition, act for several producers, those producers must evidently and necessarily communicate their prices to those agents in order to reach potential customers.

270.

Accordingly, the Commission cannot merely affirm that the communication of prices to common agents, which is justified by the actual sale of the pulp, constitutes an exchange of information between producers without demonstrating that such exchanges took place. From that point of view, the only telex mentioned in the decision which emanates from a common agency (paragraph 18) and concerns some of the applicants shows that Continental Cellulose, a Belgian agent for wood pulp, communicated the prices announced by KEA to Norrlands. ( 149 ) It is sufficient to state in that regard that, according to paragraph 36 of the decision, the agent in question did not represent either KEA or producers belonging to that association.

271.

In its decision, the Commission stated that announcements were made through the trade press. ( 150 ) Most of the producers claim never to have communicated their prices to the press. That is true, in particular, of Bowater, IPS, and the Finnish and Canadian applicants. St. Anne, which denies any systematic communications to the press, states that it has occasionally answered questions from the trade press. KEA producers claim, for their part, never to have made any public announcement to the press or to have taken any action to that end.

272.

In response to those arguments, the Commission contends in its defence that ‘the term “announcement” is used in the decision as comprising any form of disclosure or intimation of the future prices of a producer that enables these prices to be commonly known. It is therefore immaterial whether a producer contacted the trade press or whether the press contacted him. The relevant question is whether the prices would be published or in any other way passed on to other producers before they became effective and whether the producer disclosing them to a third person could count on this fact’. ( 151 )

273.

Those arguments call for two observations.

274.

In the first place, although the decision asserts that announcements are made to the press by the producers, which the producers deny, the Commission does not even seek to prove that assertion. In that regard, its arguments seem to be directed exclusively at those producers who acknowledged that they have occasionally replied to questions from the trade press. I find, therefore, that the Commission has failed to establish in this case a single instance in which prices were systematically communicated to the press by the producers. Accordingly, there is no need even to consider at this juncture whether such conduct would be justified by business reasons on the market for pulp.

275.

Secondly, it seems to me that the Commission's contention must be firmly rejected as it is so vague. The fact that a producer may know, in disclosing a price to a ‘third person’, that it would become known to his competitors is certainly not sufficient to constitute an exchange of information. In any event, the term ‘third person’ is fairly imprecise. Does it refer to customers? Does the Commission consider that customers should be bound by a duty of confidentiality? Does it refer to agents? Does the Commission consider that agents cannot communicate prices to customers if that renders the market transparent? Does it refer to the press? As we have just seen, there is no proof that the producers systematically communicated prices to the trade press.

276.

In other words, the Commission does not even attempt to demonstrate which aspects of the producers' conduct were artificial: the mere fact of announcing prices in full awareness that they will become known is treated as participation in an exchange of information.

277.

Besides, some applicants have pointed out that the size and importance of the pulp and paper industries in the economies of a number of countries lay at the root of the development of a highly effective specialist press and the appearance of specialist trade and investment analysts (the shares of several of their companies are quoted on the stock exchange). They have emphasized that a large number of qualified professionals thus ‘track down’ information which they subsequently spread. Those observations have the merit of drawing attention to the fact that the role of the press is not a purely passive one. The journals in question can no doubt obtain information from the producers, but what about the agents and the customers?

278.

That survey leads me to the conclusion that the Commission's assessment according to which the system of price announcements constituted an indirect exchange of information between competitors and created artificial transparency on the pulp market is supported by very inadequate evidence and is based on an analysis which seems to have disregarded certain features of that market.

279.

According to the Commission, the system of price announcements adopted by the producers is not necessitated by objective market conditions and it demonstrates that the parallel conduct of undertakings does not form part of the rationale behind independent competitive behaviour on the part of producers (a). It considers, more particularly, that the close succession or simultaneity of announcements constitutes evidence of concertation (b).

280.

According to the Commission, the practice of announcing prices for at least a quarter and of communicating them before its commencement is well established on the pulp market. However, it is not necessitated by objective market conditions. It was, for example, abandoned in 1982 and 1983. Moreover, the addressees of the decision declared their readiness to abandon that system by signing the unilateral undertaking annexed to the decision.

281.

Let me say at the outset that, as certain applicants have pointed out, the question whether the application of the system of price announcements is determined by the market is of no relevance whatsoever to the production of evidence of a concerted practice. The fact that the producers were able to adopt another system or even drastically amend their pricing policy as a result of the undertaking is altogether immaterial. Instead the fundamental issue is whether the details of the system of price announcements appear to be artificial. In that regard, two aspects of the system have in fact been called in question: the quarterly basis of the announcements and the use of the US dollar by producers from outside the United States. It should be noted that the undertaking entails abandoning those two features of the announcements mechanism, to which I now turn.

282.

To begin with, it is undisputed that for the period under consideration virtually all the producers generally announced their prices as applying for a given quarter. ( 152 )

283.

The applicants claim that the practice of fixing prices on a quarterly basis is well established on the pulp market. In their view, the purchasers favour stable prices in the long term whereas the sellers generally prefer prices to be fixed for shorter periods.

284.

Let me point out that most of the purchasers have expressed a preference for a system of fixing prices that remain valid for a long period of time and regard a quarter as the minimum duration of such a period. According to some papermakers, prices should be increased only every six months, or every year. ( 153 )

285.

From that, the Commission infers that ‘it is not impossible’, particularly when prices are rising, for a producer to offer to fix his prices for a longer period in order to attract new customers.

286.

That does not go far enough, in my view, if we accept that price announcements have served as maximum prices. By offering a longer period, the particular producer may in some cases attract fresh customers drawn by the prospect of stable prices in the longer term. At the same time, however, he runs the risk of losing for the whole of the period in question the benefit of a rise in prices. In other words, he has to decide whether in order to attract new customers he must forego the prospective benefit of a foreseeable increase in prices during the period in question.

287.

The possibility cannot automatically be ruled out of the producer taking the view that the profit realized on fresh sales may not necessarily be higher than that yielded by the foreseeable increase in prices during the period in question. Is it not possible for an individual and rational choice to prompt him to opt to exploit a prospective price increase to the full?

288.

In pointing out that the producers could have acted differently, the Commission in no way demonstrates in its decision that the practice of quarterly announcements, which, it acknowledges, is well established on the pulp market, necessarily constituted irrational conduct on the part of each individual producer.

289.

In its rejoinder, the Commission observes that the decision of non-United States producers to price pulp in US dollars was not due to some unalterable characteristic of the market, as shown by the fact that the Swedish and Finnish producers, at least until 1976, fixed their prices in Swedish kronor, and some of the Canadian producers (e. g. BCFP and St Anne) had earlier fixed their prices in Canadian dollars. In its decision the defendant had regarded as evidence that the addressees of the decision had ‘aimed to establish a uniform price level for their products independently of market forces’ the fact that the Swedish and Finnish producers had reverted to quoting their prices exclusively in US dollars (instead of Swedish kronor) at the beginning of 1976 in order to reestablish the parity of prices which had been affected by the devaluation of the US dollar in 1975.

290.

To treat the fact that the prices of the producers in question are the same as those of their competitors as evidence of concertation is one thing. To regard the use of the US dollar by non-United States producers as in itself an artificial practice on the pulp market is quite another.

291.

In any event, the Court will find that from 1975 at any rate, Finncell was using that currency at least in part. Table 6 annexed to the decision shows a price of USD 415 announced by Finncell for the first quarter of 1975 for softwood pulp, zone 1. And, according to paragraph 112 of the decision, in the last three quarters of 1975 Finncell announced a price in US dollars and in Swedish kronor, the dollar price being at times higher and at times lower than that announced by the Swedish producers in kronor.

292.

During both the proceedings before the Court and the administrative procedure, ( 154 ) moreover, the Finnish applicants explained why the US dollar was regarded as a ‘lead currency’ on the market for wood pulp, emphasizing, in particular, the European customers' preference for the use of the dollar. Negotiations between the Commission and the producers with regard to the undertaking gave purchasers, or at least some of them, the opportunity to make their views known quite clearly in that regard. Their position, often expressed in very forceful terms, is free from any ambiguity: there has to be a single currency, whether or not the US dollar, for pulp transactions.

293.

I do not intend to embark here on a detailed examination of the various letters sent by paper mills to the Commission during its discussions with the producers. ( 155 )

294.

Allow me simply to quote a passage taken from a letter from the Secretary General of the Association of German Papermakers: ‘Apart from that general remark, I consider it important that dealings in a raw material of world wide significance are carried out in a single currency. In that connection it is of no importance to the German paper industry whether that currency is the US dollar or another strong currency. We are concerned only that it should be a single currency because only that can provide the guarantee that all competitors who obtain that raw material purchase it at the same quoted price’. ( 156 )

295.

A passage from another letter reads as follows: ‘We have always found it valuable to use the dollar price as a reference point. The removal of that reference point would not, in our view, be beneficial to the buyers, nor lead to any ultimate benefit for the consumer. Changing the basis to a wide range of currencies would make any assessment of the true current price more difficult due to variation in currency exchange rates. This would be to the disadvantage of the buyer rather than the seller and could lead to buyers in particular countries being placed at a competitive advantage over others. Therefore our ability to ensure that we are in a fair competitive situation with competitors in other countries would be eroded’. ( 157 )

296.

It is quite clear from those documents that both the German and British purchasers in question are in favour of the system of using a single currency. Are those standpoints representative? It would in any event have been useful if the Commission had expressly stated its views ( 158 ) on those documents, the production of which during the proceedings makes it impossible, in my view, to establish with any degree of certainty that the use of the US dollar as such constitutes an artificial mechanism.

297.

The Commission considers that the rapid succession or even simultaneity of price announcements would not have been possible without a constant flow of information between the producers concerned (paragraph 107). It considers that there is no valid explanation of how such information spread so rapidly, that is to say, within a matter of days, or even on the same day, between such a large number of producers. The information supplied by customers passes from one producer to another in a multistage process (the producer — his agent or subsidiary — the customer — the agent or subsidiary of another producer — the producer) before it is disclosed to another producer who makes his own announcement.

298.

According to the decision, the simultaneity of announcements was particularly clear in some quarters, notably the first quarter of 1978 and the first three quarters of 1979. Producers in the same country generally announced their prices within a few days of one another.

299.

The applicants consider that the Commission's findings with regard to the dates on which announcements were made show that those dates were fairly close; however, those findings are often incomplete and they disregard such features as enabled the applicants to discover the prices announced by their competitors, and in particular the role played by customers.

300.

Let it be noted, first of all, that the Commission seems to have substantially underestimated the speed at which the information on announced prices given by the customers themselves was circulated. In that regard, I would again emphasize that it is clear from certain documents which were produced during the proceedings and whose authenticity has not been called in question by the defendant, that the possibility cannot be categorically ruled out, as the decision did, that producers are informed by customers of the prices announced by their competitors very soon after the latter have ‘brought out’ their prices, even in times of rising prices. It is also apparent from those documents that the customers are, in some cases, aware of the intentions of producers who have not yet disclosed their prices and that, once again, such information may be passed to competing producers by the customers themselves.

301.

The relevance of the Commission's contention that there is ‘no valid explanation’ (paragraph 89 of the decision) of how the producers are able to discover their competitors' prices so quickly must therefore be assessed in the light of the documents referred to earlier.

302.

It seems to me, therefore, that the proximity of the dates of the announcements ( 159 ) cannot as such be regarded as an indication of prior communication between producers.

303.

On the other hand, announcements ‘brought out’ on the same day by several producers would attract attention, particularly if they constituted the first announcements for a given quarter. In that regard, the Court will find that, according to the first experts, the announcements were not made on the same day.

304.

For my part, I have examined those quarters in which, according to the decision, the simultaneity of announcements was particularly striking. However, there is a preliminary remark I wish to make.

305.

The decision found that the members of KEA had committed a distinct infringement and, as I have already pointed out, the Commission cannot, on the same facts, have established that they committed an infringement by engaging in general concertation on prices. In any event, any simultaneity of the dates on which announcements were made solely as between members of KEA ( 160 ) cannot be regarded as evidence of concertation against producers outside that group. On the other hand, the simultaneity of those dates with that of other North American producers not members of KEA may in certain circumstances constitute a ‘cause for concern’.

306.

I now turn to an examination of the quarters in question, with reference to Table 6 annexed to the decision.

307.

Let me begin with the first quarter.

308.

The dates of announcements by the Swedish producers are particularly scant and neither the date of Finncell's announcements nor its prices are set out in the table.

309.

On the other hand, the dates on which certain North American producers made their announcements are very close. MacMillan seems to have been the first to announce its prices on 9 December, followed on 12 December by Bowater and ITT Rayonier, which was a member of KEA at the time, on 13 December by KEA itself and by Weldwood, a Canadian producer, on 15 December by another two KEA producers and on 16 December by Crown Zellerbach, also a KEA producer.

310.

It should be noted, however, that Weldwood's announcement was made four days after the first announcement by the only other Canadian undertaking referred to in the table, namely MacMillan.

311.

On the other hand, the simultaneous announcements made on 12 December by Bowater and ITT Rayonier, apparently the first United States undertakings ( 161 ) to have simultaneously ‘brought out’ their prices, may give pause for consideration. However, no conclusion can be drawn in that respect. The date of the announcement which the table attributes to Bowater was formally challenged by the latter. Bowater claims that it disclosed its prices on 5 January 1978 and the Commission has acknowledged that ‘the files do not contradict’ that claim. Furthermore, the price announced by Bowater is USD 10 or 20 lower than that of the other United States producers and Bowater was not found by the Commission to have been involved in any way in concertation on announced prices for 1978.

312.

I now turn to the first quarter of 1979.

313.

Apparently, the first undertaking to announce its prices was the Swedish producer Södra, on 17 November. Three days later, on 20 November, the table shows that six United States undertakings announced their prices simultaneously. Who would not be struck at first sight by the simultaneity of the dates on which five KEA producers and Bowater, a producer outside that group, made their announcements? It should, however, be observed that, once again, the date attributed to Bowater in Table 6 cannot be proved by the Commission. Bowater claims to have stated its prices on 15 December 1978 and the Commission considers, let us remember, that the file does not contradict Bowater's claim in that regard. Finally, scarcely any guidance can be obtained, in my view, from the announcements of the only two Canadian producers, which are recorded as having been made within a day of each other, and it is noteworthy that no date is given for Finncell's announcement.

314.

That brings me to the second quarter of 1979.

This table also reveals the proximity of the dates on which the Swedish producers made their announcements. No date is given for the Finncell's announcement, and the table made merely states that it was made in Marc for There is a specific date for only one North American producer, at least 13 days after the first Swedish announcement indicated on the table. For Bowater and Chesapeake, a United States member of KEA, the date is given merely as the start of March. What conclusion can be drawn from that succession of announcements, other than the proximity of the Swedish announcements?

315.

Finally, I turn to the third quarter of 1979.

This table has at least as many gaps as that for the previous quarter: a specific date is given for only one North American producer, and none at all for Finncell. As the table is so incomplete as regards the dates of announcements, it is impossible, once again, to establish the chronological sequence of announcements for the quarter in question, particularly in the case of the Canadian and United States producers.

316.

A detailed examination of the tables in question is a somewhat tedious exercise which I have been obliged to carry out on account of the Commission's contentions as to the striking simultaneity of the announcements made for the periods in question. Apart from the marked proximity of the dates of those announcements, I have not found any obvious evidence of an inexplicable similarity. ( 162 ) Have I perhaps overlooked a particularly ‘eloquent’ set of identical dates? I cannot rule out such a possibility, notwithstanding the care I have sought to exercise in analysing the periods in question. If that had been the case, it would nevertheless have been for the Commission specifically to draw attention to situations of that kind. Besides, those tables are evidently incomplete and the Commission has been unable to establish the dates of announcements — I have in mind Bowater's — which would at times have served to reveal the most marked similarities. In any event, let me emphasize, it is at first sight hard to understand why the Commission has persisted in arguing that announced prices could not accurately and rapidly come to the knowledge of competitors without attempting to explain how it justified that position in the light of certain evidence to the contrary which it has apparently disregarded, without even discussing it.

317.

In my view, therefore, the Commission's assessment of the system of announcements and of the details of that system has by no means been established beyond all doubt in the light of the documents before the Court. In any event, the defendant has failed to provide a satisfactory answer to the applicants' arguments to the effect that it underestimated certain features of the pulp market.

318.

The grounds of the decision seem to apply without distinction to concertation on transaction prices (Article 1(2)) and to concertation on announced prices (Article 1(1)), the only kind which I consider it necessary to examine here. The Court requested the defendant to explain, in particular, whether the analysis in paragraphs 90 to 105 of the decision also applies to parallelism of announced prices. The Commission replied that the analysis was also relevant to its conclusions regarding concertation on announced prices.

319.

Let us recall, therefore, the substance of the Commission's analysis in its entirety, which is aimed at excluding the possibility that parallelism can be explained without concertation. According to the decision, in the light of the conditions prevailing on the pulp market and of a proper economic analysis, the producers' conduct from 1975 to 1981 cannot be explained as independently chosen parallel conduct in a narrow oligopolistic situation.

320.

The Commission stated that the European market is supplied by 50 or so producers and consists of several hundred customers; it emphasized that, among both suppliers and purchasers, there are a large number of substantial undertakings whose size and market position enable them to pursue an independent and active competitive policy. The large number of competitors, the many customers, and the very wide range of products give the undertakings an opportunity to improve their position at the expense of their rivals. However, they have continually foregone such opportunities, which are the mainspring of competition, and have systematically offered their products on the same terms. The Commission further explained how the system of price announcements had created artificial transparency on the market, which neutralized the fact that concertation had been made more difficult by the large number of competitors. ( 163 )

321.

The decision also noted that the simultaneity of the price announcements and the similarity of announced prices cannot be explained by the existence of a ‘price leader’ whose prices are adopted by competitors. In the period under consideration, the order of the undertakings announcing new prices changed from quarter to quarter and no individual undertaking had a sufficiently strong market position to establish itself as a market leader.

322.

Furthermore, the similarity of announced prices and actual prices cannot be explained as so-called ‘equilibrium prices’. The Commission acknowledges that effective competition tends to lead in any market situation to an ‘equilibrium price’. However, that price, which fluctuates with any change in market conditions, is the result of a testing of the market by trial and error which can lead to higher or lower prices and compel the undertaking which was the first to announce a higher price to abandon it. From that point of view, it is inexplicable, under normal competitive conditions, that in the step-by-step rise in prices, such as occurred in the case of northern softwood from the third quarter of 1979 to the second quarter of 1980 in four stages, and in the case of southern softwood from the second quarter of 1979 to the second quarter of 1980 in five stages, the first higher price demanded should in each case prove to be the new ‘equilibrium price’ and that no undertaking should have tested the market with a different price. It is equally inexplicable that under real competitive conditions on a rapidly changing market and during a period of two years (from the first quarter of 1975 to the fourth quarter of 1976) no undertaking except Finncell should have made the slightest attempt to test, by charging a different price, whether the price charged still corresponded to the ‘equilibrium price’.

323.

The decision also stated that, in view of the large number of producers and the differences in economic conditions which determine the pricing policy of individual producers, the uniformity of prices cannot be explained by mere coincidence of independent decisions. In view of the differences in the producers' positions, there is no reasonable economic explanation for the fact that the addressees of the decision behaved with regard to their pricing policy in an absolutely similar fashion. An analysis of the differences in question is set out in paragraph 90 et seq. of the decision. To summarize, the Commission has found that:

324.

According to the decision, the differentiation of prices for bleached sulphate wood pulp would nevertheless have been possible. The decision thus refers to the differences between the second quarter of 1977 and the second quarter of 1979 when, as regards the prices invoiced, the producers must have decided not to charge the unrealistic concerted prices. ( 165 ) The Commission also refers to the practice of two undertakings, Domtar and Bowater, in showing that during certain periods their transaction prices differed from those of their competitors.

325.

Furthermore, the Commission considers that the shifts in market shares between the producers between 1975 and 1981 do not support the conclusion that there was no concertation. Those shifts were much less marked at times when the producers charged uniform prices than at other times.

326.

Finally, it should be noted that according to the decision, the prices announced in 1976, 1977 and 1981 were kept at an artificial level which differed widely from that which would have been expected under normal competitive conditions. For example, it would have been inconceivable, given unrestricted competitive conditions, for the same price of USD 415 announced for northern softwood to have remained unchanged from the first quarter of 1975 to the third quarter of 1977. Thus, in the second and third quarters of 1977, announced prices sometimes stood at up to USD 100 above the selling price which could in fact be obtained on the market. Similarly, the rapid fall in actual selling prices in 1977 (25%) and in 1982 (20%) showed that, for a significant period, selling prices were higher than those which would have been obtained under normal competitive conditions. According to the decision, the fall in prices was triggered by market forces which compelled the producers to bring to an end concertation between them. The stagnation or slight fall in consumption in 1977 and in 1982 would not have been sufficient to trigger a fall in prices of that magnitude if the prices achieved in 1976 and in 1981 had corresponded to the competitive market equilibrium price. Furthermore, in 1975 the producers had absorbed a fall in consumption of more than 15% and a fall in imports of just under 30% without that leading to a fall in prices. The prolonged strike in Canada in 1975 and the Swedish Stock Incentive Scheme were not sufficient to compensate for the overall decline in demand. The Swedish suppliers capacity utilization ratios fell, despite the State aid programme, by 10% to 15%, that of Finnish suppliers by some 15% and that of United States suppliers by up to 20%.

327.

That is the Commission's argument, which led it to conclude that parallel conduct could not be explained without concertation.

328.

Against that analysis, the applicants have marshalled abundant arguments, which are very clearly set out in the Report for the Hearing. I refer to that document for details of the discussion before the Court between the parties in that regard. Let me say simply that the applicants challenge the Commission's conclusions on the ground that their conduct with regard to announced prices reflected independent decisions to align themselves with the prices announced by their competitors, which were known on account of the natural transparency of the market, and that their individual pricing policy was determined in a rational and independent manner in the light of their own economic circumstances which they frequently describe in great detail.

329.

I would observe at this juncture that most of the applicants claim, at least with regard to the periods in which they acknowledge that announced prices were the same, that this can be explained, in particular, by the homogeneity of the product in question and the transparency of the market. A producer has no interest in announcing prices higher than those of his competitors since he would then lose customers, as the Swedish and Finnish producers found in 1975. Nor, according to some of the applicants, does a producer have an interest in undercutting announced prices since such prices merely constitute a ceiling for the negotiation of actual transaction prices with individual customers. The Finnish applicants emphasize, for their part, that a fall in Finncell's prices would be followed by their competitors, whilst the quoting of prices higher than those of their competitors leads to a fall in sales.

330.

It should also be stated that, according to the applicants, certain groups of competitors acted as market leaders, although none of the applicants admits to belonging to a group which was strong enough to do so. The Finnish applicants claim that, having sustained losses in 1975, when they announced prices higher than those of their competitors, from the following year until the beginning of 1977 ( 166 ) they had no choice but to fall into line with the prices generally quoted by their competitors, emphasizing in that regard the strength of the North American producers whose policy with regard to prices they could not afford to disregard if they wished to avoid losing more customers. According to the members of KEA, the ‘KEA recommended price’ was generally adopted after the Swedish, Finnish and Canadian producers had announced their prices. IPS observes, for its part, that, for most of the quarters in question, the Scandinavian producers were the first to announce their prices. Bowater and St Anne claim always to have acted as ‘followers’ in relation to other producers, and were generally the last to announce their prices. St Anne states that when it announced the same prices as its competitors, it was merely aligning itself independently with the prices of KEA and Finncell, which — St Anne claims — acted as market leaders. Finally, the five producers from British Columbia also claim that they did no more than align themselves with the prices of other producers, always waiting for Scandinavian, KEA or individual United States producers to make their announcements. They also emphasize that the Scandinavian producers are the natural market leaders and that the decision contains firm evidence of the existence of an export association (KEA), a joint marketing agent (Finncell) and discussions on prices under the auspices of the Bristol Club. Those different groups, to which they never belonged, account for a substantial share of the Community market.

331.

In support of their view, some of the applicants submitted, either during the administrative procedure or in the proceedings before the Court, opinions drawn up by distinguished economists, Professors Budd, Hart, Jacquemin, Phlips and Von Weiszäcker.

332.

I would simply observe here that some of those opinions focused more specifically on analysing the behaviour of individual addressees — such as Professor Budd's report concerning Bowater and the reports compiled by Professors Hart and Von-Weiszäcker on the members of KEA — in reaching the conclusion that such behaviour did not in any way point to their participation in concertation with other producers. In their analysis, Professors Jacquemin and Phlips considered, for their part, that the structure of wood pulp prices could be explained without reference to the possibility of concertation, having regard to the characteristics of that market.

333.

In its rejoinder the Commission carried out a critical examination of those different opinions, firmly rejecting their conclusions. On completion of the written procedure, therefore, the Court was confronted with a substantial body of economic argument, referring at times to theoretical models which, whilst doubtless familiar to an economist, are nevertheless, in my view at any rate, of manifest complexity (I am thinking, for example, of the Cournot-Nash equilibrium which, according to Professor Phlips, was a feature of the pulp market). The Court was therefore obliged to order an expert's report whose compilation was entrusted, with the parties' consent, to Mr Cockram and Mr Fishwick.

334.

In substance, Mr Cockram and Mr Fishwick came to the conclusion that the uniformity of prices could be explained by the natural operation of the wood pulp market. Indeed, such uniformity of prices was in some respects inconsistent with the operation of a cartel. They pointed out, however, that this did not per se prove the absence of concertation.

335.

The Commission, both in its observations on the expert's report — annexed to which is an opinion by Professor Neumann on which the defendant relies — and at the hearing, challenged the expert's report and requested the Court not to accept its conclusions.

336.

It is now time to embark on a discussion of this issue, which I shall attempt to organize around those aspects of the experts' report at which the Commission's major criticisms are directed.

337.

The Commission raised the objection first of all that the expert's report was based on a misleading oversimplification of the reasoning in the decision. In asserting that the Commission's conclusion was ‘based mainly, but not entirely, on the indirect evidence of apparently similar prices’ the experts disregarded the fact that the decision was based on a ‘dynamic’ approach to price trends. Thus, it was the manner in which identical prices were announced and, in particular, the simultaneity or near-simultaneity of the announcements, the total lack of an alternative announced price and the fact that the announced prices continued to be applied by the applicants in transactions over at least the following three months, notwithstanding differences in the economic conditions determining the pricing policies of individual producers which, together with the uniformity of the prices themselves, was regarded as evidence of the concerted practice. By contrast, the models used by the experts were essentially static, taking no account of the manner in which the ‘equilibrium prices’ were achieved on the pulp market.

338.

I have three observations to make in that regard.

339.

If the Court refers to the task entrusted to the experts, it will find that the second question put to them was as follows: ‘the expert is requested to give his opinion, in the light of those factual characteristics, on whether, and for what reasons, the natural operation of the wood pulp market should lead to a differential price structure, or whether, and for what reasons, it should lead to a uniform price structure’. It must be noted that the question did not lay emphasis specifically on the details of the changes in prices underlined by the Commission. When the Court communicated to the defendant the draft definition of the task of the experts, which was precisely couched in those terms, ( 167 ) the defendant replied by letter of 31 August 1990 that it was satisfied with it.

340.

Secondly, the Court will find that some of the aspects which the experts did not deal with, wrongly according to the Commission, in fact relate to the details of the system of price announcements, and in particular the simultaneity of those announcements. I have already analysed the defendant's views in that regard.

341.

Finally, at the hearing I asked the experts to explain whether they considered that the expression ‘equilibrium price’ was applicable to announced prices or whether it applied to transaction prices alone. That is the view taken by several of the applicants who maintain that only transaction prices are capable of reflecting the balance between supply and demand, whilst the announced prices merely act as a ceiling. Mr Fishwick stated that the announced price acted as a ceiling price, whilst the transaction price would be the ‘equilibrium’ price. I have already pointed out in that regard that the Commission, in contending that the applicants' argument that announced prices constituted a ceiling is manifesdy unfounded, had stated that, in one quarter out of the 28 covered by the decision, certain United States producers had charged a price higher than that which they had announced. In that regard, I pointed out that the defendant had drawn certain conclusions which, in my view, were highly debatable, from a situation which, according to its own findings, was exceptional and that the pulp purchasers themselves regarded announced prices as buffers.

342.

The Commission further considers that the experts referred in some places to the ‘perfect competition’ model, and in others to the ‘pure oligopoly’ model. In its view, neither of those models applies to the pulp market and it is, in any event, inconsistent to use them simultaneously.

343.

As regards, first of all, the ‘perfect competition’ model, it should be noted that the expert's report contains a number of express references to that model and that the defendant has, moreover, drawn attention to those characteristics of the pulp market which depart from it, emphasizing in addition that, as the relationship between undertakings' average costs and observed prices on the pulp market also demonstrates, the standard model is quite inapplicable in this case. ( 168 )

344.

At the hearing of the experts, Mr Fishwick informed the Court that he had used the concept of a ‘perfect market’ and did not suggest at any point in the report that the structure of the wood pulp market approximated the conditions of ‘perfect competition’, which requires an atomistic supply structure. The definition of a ‘perfect market’ is to be found on page 5 of the report: it is a market where customers are perfectly informed about prices and qualities of goods available and are willing and freely able to substitute one product for another. In a market of that kind, identical products would sell at the same price. Demand for the product of any one undertaking would be perfectly elastic and would not be affected by the structure of supply. From that point of view, in the short term pulp purchasers depended on a small number of suppliers for each type of pulp and would pay similar prices. The references to perfect competition on pages 7 and 8 of the report were a response in theoretical terms to the observations in paragraphs 84, 90 and 93 of the decision in which the Commission implied that supply by a large number of undertakings with different costs would lead to price differences. ( 169 ) The experts thus sought to demonstrate that, even under conditions of ‘atomistic’ supply or perfect competition, prices would be uniform irrespective of differences in costs.

345.

For my part, I can only take note of those explanations and add that the expert's report does indeed state that ‘uniform prices would prevail in a perfect market irrespective of the structure of supply to that market. Provided that buyers remain aware of the homogeneous nature of the product and are free to transfer between suppliers, it does not matter whether there are very few suppliers or very many — identical products must sell at the same price’. ( 170 ) I find, however, that there is at least one other passage in the expert's report which refers to competition on the pulp market between a large number of competitors. ( 171 )

346.

The Commission's second criticism of the expert's report touches, in my view, one of the key points of the report in that the experts concluded that the pulp market reflected short-term oligopolistic situations.

347.

It is necessary to refer to the expert's analysis in that regard. They took the view that although price-elasticity of demand for pulp in aggregate may be low, price-elasticity in response to a price change by an individual supplier would be very high, if other manufacturers of interchangeable products keep their prices unchanged. Each manufacturer is acutely aware of the cost of his pulp in relation to that used by his competitors, since pulp represents a major element of cost. However, such elasticity is reduced owing to the effect of two factors: the products are not altogether interchangeable and the existence of stable relationships between producers and purchasers.

348.

In that regard, according to the experts, hardwood pulps are generally mutually interchangeable, whereas only 75% of northern softwood pulps and 95% of southern softwood pulps are. Southern softwoods may, for certain types of paper, be substituted for northern softwoods but, since they are of inferior quality and slow down the papermaking process, such substitution is possible only in weak market conditions. In order to make substitution permanent, specific equipment must be installed, which means that southern softwood must sell at a lower price.

349.

Paper is produced from a mixture of different types of pulp. Each paper manufacturer needs a particular kind of mixture and identifies suppliers who are reliable in terms of delivery and quality. Thus, even though the number of suppliers and purchasers may seem large, for any one category of pulp within which products are interchangeable there may be only a few suppliers and a small number of purchasers accounting for a substantial proportion of total purchases. The experts have described that situation as an ‘oligopoly-oligopsony’, that is to say few sellers to few purchasers.

350.

In view of the interdependence between pulp producers and papermakers, long-term relationships have been established. Those relationships were mutually advantageous to the producers and their customers, providing security of demand and supply. They have been sustained by the cyclical nature of the market: customer loyalty in a weak market balancing supplier priority in a more buoyant market. Advance notification of prices to purchasers for each quarter forms an integral part of that arrangement.

351.

According to the experts, that analysis has implications for price-elasticity, which they describe as the most complex aspect of the case. In the short term, commercial relationships between purchasers and suppliers are likely to have reinforced the oligopoly-oligopsony for individual qualities of pulp. That being so, pricing strategies must take account of the interdependence of producers. Thus, any producer reducing his prices would expect competitors to follow suit. His customers would probably demand parallel price reductions as an alternative to switching orders. In the short term, no producer would wish to initiate a price war, if the foreseeable result was a reduction in prices in a market where aggregate demand is considered inelastic. Similarly, no producer would raise prices unless he was confident that his competitors would follow suit, and they would only do so if they did not have the capacity to increase sales at the expense of the producer who was the first to raise his prices.

352.

In the short term, prices would thus tend to be static in a weak market, because each supplier of a particular type of pulp would realize that any price cut would be followed by other suppliers of that product. In short-term oligopoly conditions, prices would tend to be slow to fall in conditions of excess supply and that inflexibility of prices downwards would be reinforced by market transparency, every supplier realizing that any change in price would quickly be matched by competitors. Under tighter conditions in a strong market, transparency would facilitate price increases because the reactions of competitors would soon be revealed. Concertation on prices would not be necessary to produce that result.

353.

Still according to the experts, however, short-term oligopoly was limited in its effects by longer-term substitutability.

354.

To begin with, long-term commercial relationships depend largely on price uniformity. Ability to raise prices depends not only on the willingness to follow on the part of other suppliers of the same category of pulp, but also on parallel price increases for other pulps which can be substituted by means of changes in the mixture or in the production process. Except in the short term, the number of potential suppliers far exceeded the number likely to be supplying one grade of pulp to a particular papermaker. Secondly, in the late 1970's and early 1980's there was potential competition from the south of the United States and from tropical climates, particularly from eucalyptus of Brazilian origin. The use of recycled paper provided another possible brake on price increases. Those longer-term considerations limited the upward movement of prices which would otherwise have occurred during periods of buoyant demand under oligopoly conditions.

355.

The Commission disputed the principle on which the experts' analysis was based. In its view, it was impossible to isolate relations between certain market participants in order to determine whether those relations constituted an oligopoly; it was necessary to consider the market as a whole. At the hearing, Mr Fishwick stated in that regard that, where relationships exist between sellers and purchasers, the theory has to be adapted in some respects because of the situation which he regards as abnormal but that, in any event, it did not affect the validity of the principles concerned.

356.

I do not believe I am flinching from my responsibilities in considering myself unable to express a view on the theoretical pertinence of an analysis which distinguishes a series of ‘small oligopolies’ in the short term within a given market. At the very least, is it possible to establish whether relations between sellers and purchasers took the form indicated by the experts? The Commission emphasizes in that regard that some of the applicants — and certain economists whose opinions they have produced — have themselves laid emphasis on the large number of competitors and the distribution of sales on the market.

357.

At the hearing, Mr Fishwick stated that each purchaser bought a given grade of pulp from two or three sellers. He also pointed out, referring by way of illustration to the situation of the Canadian producers for the first quarter of 1977, that although each producer sold to a large number of customers, a few customers accounted for most of the sales.

358.

For my part, I have already observed that the Commission did not apparently dispute that the number of major customers was relatively small. However, although the pattern of relations between sellers and purchasers as found by the experts may have corresponded to the pattern that existed on the pulp market, I believe that it can hardly be regarded as constant in view of the documents before the Court. Besides, the applicants have scarcely drawn attention to the phenomena found by the experts. As the Commission points out, some of them have on the contrary laid emphasis on the large number of sellers and purchasers. It is true that the experts' analysis applies to short-term relationships between sellers and purchasers as regards pulp of a particular grade. In that connection, however, the report emphasizes that ‘the supplier's perception of “price-elasticity” is of greater significance than any objective measurement, because it is the perception which would have determined pricing policy’. In other words, on the assumption that for pulp of a particular grade there are few purchasers and few sellers, at least in the short term, and that this situation has given rise to interdependence between the producers concerned, it is puzzling to observe that not all the producers have emphasized that phenomenon which, however, according to the experts, led to a policy of fixing prices. With regard to interdependence, however, attention must be drawn to the arguments of Finncell in which it expressly emphasized that ‘any substantial departure by a large supplier from the prices generally quoted has the effect either of inducing his competitors to align themselves with his price or of preventing him from gaining customers' acceptance of his new price’. ( 172 ) The Commission apparently acknowledged that, in view of Finncell's market share, it was plausible that competitors reacted to the fall in Finncell's prices. But that was not the case, according to the defendant, for the individual United States and Canadian producers who, in view of their much smaller market share, would have been far less likely to be followed if they had reduced their prices.

359.

Let me point out, moreover, that at the hearing the experts were asked to explain how they reconciled their view that, in times of rising prices, parallelism without concertation is a result of market transparency with the principle, referred to in their report, that in an oligopolistic market prices are slow to rise.

360.

Mr Fishwick stated in that regard that the increases in 1979 and 1980 were followed because stock levels were low and utilization of production capacity was high. This meant that producers could increase prices in the knowledge that they would not lose market shares because their competitors did not have spare capacity and would therefore follow suit. He also pointed out that Table 6 annexed to the decision did not provide evidence of price ‘leadership’ for the period from 1977 to 1980. Different undertakings were the first to announce price increases in different quarters, which showed that there was no collusion.

361.

According to the Commission, that analysis was based on the assumption that producers were aware of their competitors' capacity utilization ratios. From that point of view, Mr Fishwick pointed out that he did not think that producers knew each other's capacity utilization ratios but that the market was extremely transparent and yielded a remarkable amount of information concerning capacity utilization and the level of stocks in the industry as a whole.

362.

With regard to the ostensible absence of ‘price leadership’ in times of rising prices, the Commission raised the question whether it mattered which undertaking announced a price first, if there was prior agreement on what the announced price would be.

363.

I wish to make two observations with regard to that aspect of the matter. First of all, it should be noted that the expert's report itself has relatively little to say with regard to parallelism of price increases. Mr Fishwick's explanation is, certainly, not altogether implausible in that respect but I must say that a more detailed examination in the written report itself would have been welcome. However, contrary to the attitude which the Commission seemed to take during the oral procedure, the experts expressly referred in their report to the publication of stocks and utilization ratios in stating that transparency may have played an important part in this case by facilitating price increases when the market was strong.

364.

It should also be noted that Mr Fishwick stated, in reply to a question from the Commission, that consultation between producers before price announcements would have been an effective way of exploiting the increase in demand and maximizing the price that could be obtained on the market. He pointed out that such incentives to concert were not peculiar to the pulp market but were true of any market in which aggregate demand was inelastic.

365.

The experts pointed out that on an oligopolistic market, the smaller the number of competitors, the greater the degree of interdependence and the greater the excess of price over marginal costs. According to the defendant, however, the experts failed to mention that it is possible to examine the amount of the price excess for any given degree of concertation in an oligopolistic industry to determine whether it is merely of the size that would be expected without collusion. A higher excess is cogent evidence of collusion.

366.

In that regard, the Commission relied on Professor Neumann's opinion. He considered that if all 39 producers listed in Table 5 annexed to the decision had been acting independently, a Cournot-Nash equilibrium would have emerged where the ratio of ‘price minus marginal cost over price’ would have been equal to the Herfindahl Index of concentration divided by the price elasticity of demand. That is a standard result of oligopoly theory. Applying that formula to the data given in the decision, Professor Neumann considers that the price excess over marginal costs in the event of non-collusion should have been around 5 to 10%, whilst it amounted to around 40 to 60% from 1974 to 1976 and around 36% in 1981. Whilst he acknowledges that those rough and ready calculations are subject to a wide margin of error, there can be no doubt, in his view, that, given that order of magnitude, collusive arrangements led to price-cost margins being substantially higher than they would have been under truly competitive conditions where producers acted independently.

367.

At the hearing, Mr Fishwick challenged that analysis. He considered that it was based on the assumption that all the producers were in the oligopoly. In the short term, there was a greater degree of restriction on the supply available, with the result that the Herfindahl Index should have been calculated differently. The ‘time dimension’ should be introduced in the market definition and, in the short term, profit margins were greater than would be the case if all the producers were included in the oligopoly.

368.

In my view, if, as the Commission contends, economic theory provides a mechanism for establishing cogently whether the level of prices is collusive, it is unfortunate that it was not referred to either in the decision or in the Statement of Objections and that the Commission invoked it only in its observations on the expert's report. The decision does not specify the formula for computation on which the defendant relies in discussing the conclusions reached by Mr Fishwick and Mr Cockram. But, as the Commission pointed out at the hearing, paragraphs 112 and 113 of the decision do refer to the artificial level of prices at certain periods.

369.

In that regard, the experts considered that the fall in prices in 1977 was the result of the massive excess supply resulting from the termination of the Swedish stock subsidy scheme in June 1977 at a time of stagnating demand. It should be borne in mind at this juncture that the Commission considered in paragraph 113 of the decision that the very substantial fall in transaction prices in 1977 (25%) showed that the previous level of prices had been artificial since such a fall in prices could not have been triggered by the stagnation in consumption or the slight decline in imports during that year. I must observe that, so far as concerns the fall in prices in 1977, the decision makes absolutely no mention of the possible consequences of the ending of the Swedish stock incentive scheme, although it does refer to the huge increase in stocks in Sweden caused by that scheme. ( 173 ) However, one does not have to be a top flight economist to imagine the impact on the level of prices of the disposal of stocks accumulated under the Swedish State aid scheme. All I need say is that in failing to refer expressly to the ending of that scheme, the contested decision seems to me to contain an analysis which is to say the least incomplete in that respect. Moreover, the difference of USD 100 found by the decision to have existed in the second and third quarters of 1977 between announced prices and actual prices is indeed substantial. But, assuming that announced prices constitute ceiling prices, does the fact that they apparently lost all credibility at the time when the Swedish scheme was on the point of being discontinued or had just ended carry decisive weight?

370.

In their report, the experts stated that the fall in prices for 1977 had brought prices down towards the variable costs per unit of the most efficient producers. According to the Commission, that finding showed that prices were previously higher than the producers' marginal costs. It emphasized in that regard that Table 1 annexed to the decision shows that prices were far higher than the producers' total variable average costs, whilst the capacity utilization ratios of a number of producers were not very high.

371.

From that point of view, Mr Fishwick stated that profits were no doubt greater than they would have been under conditions of perfect competition, but considered that the excess could be explained by the oligopolistic nature of the market. He pointed out that prices for 1975 and 1976 had remained the same as that determined in 1974 by excess of demand over supply. Oligopoly theory does not serve to account for the level of prices but merely explains their tendency towards rigidity. In that regard, according to the experts, special factors, to which I now turn, contributed to the persistence of that situation, which was particularly marked before the fall which occurred in 1977.

372.

It should be noted, in the first place, that an announced price for northern softwood pulps, zone 1, of USD 415 was maintained by all the producers for most of the period in question, and by certain producers for the whole of that period. The experts considered that there had been a number of specific reasons why the increase in the level of stocks and the fall in capacity utilization had not led to a decrease in prices.

373.

The experts focused on 1976, in particular, emphasizing that this was the year in which world production of paper recovered and economic forecasts were considered good. They also pointed out that, in view of the rate of inflation, the real value of list prices fell, even though their nominal value remained unchanged.

374.

So far as the situation of the producers was concerned, the experts analysed it by reference to national groups.

375.

With regard to the Swedish producers, the experts found that they qualified for a government stock incentive scheme. Financing was based on the value of stocks and therefore any fall in prices led to a decrease in the aid granted.

376.

With regard to the United States producers, some of whom had, however, been building stocks, the experts emphasized the strength of demand on the United States market in 1976 and the high capacity utilization ratios. Apparent consumption of paper on the United States market during that period was at record levels, showing an increase of 32.8% on 1975.

377.

The buoyancy of the United States market was also beneficial to the Canadian producers, whose utilization ratios were reasonably high. In connection with the Canadian producers, Mr Fishwick also emphasized at the hearing that they benefited at the time from negative real interest rates. The cost of stock-building in those circumstances was therefore particularly attractive. Accordingly the experts consider that the Canadian producers had no interest in lowering prices having regard to the spare production capacity of the Swedish and Finnish producers, who would undoubtedly have followed suit. In that connection, when the Commission emphasized at the hearing that Table 3 annexed to the decision refers to average utilization capacity for individual producers, Mr Fishwick stated that he had not been aware of this. According to the Commission, the table in question shows that there was at least one Canadian producer with a capacity utilization ratio of 63% for 1976, which was incompatible with prices remaining at the same level. Mr Fishwick stated that if the table were read in that manner, it would certainly raise a question as to why a producer operating at 63% of capacity did not reduce his prices. But he emphasized that it would have been extremely dangerous even for such an undertaking to initiate a price war in view of the risk that the Swedish and Finnish producers, who also had substantial spare capacity at the time, might have followed suit.

378.

The experts further point out that the Finnish producers had substantial spare production capacity although the growth in their stocks was unaccompanied by offsetting subsidies as was the case in Sweden. The experts considered that, in view of the conditions prevailing in the four countries in question, the traditional theory of competition among a very large number of undertakings suggested that in spite of longer-term strategic considerations, Finnish undertakings would have been the first to cut prices. That theory was confirmed, in their view, by Table 6 annexed to the decision, which showed that Finncell began to cut announced prices for hardwood in the first quarter of 1977, followed by a similar reduction in the second quarter for softwood.

379.

The Commission has vigorously challenged that analysis of the Finnish undertakings' behaviour. It contends, in that regard, that over the same period Finncell discussed prices with other European producers within the framework of the Bristol Club, which is hardly consistent with a traditional theory of competition. It has also pointed out that all the producers began to reduce their transaction prices at the end of 1976 and the beginning of 1977, quoting, moreover, extracts from documents in which Finncell claims to have adopted a policy of following its competitors.

380.

The Commission's criticisms in that respect run counter at first sight to the indication in Table 6 of a fall in the prices unilaterally announced by Finncell during the first two quarters of 1977. Furthermore, Professor Neumann, whose opinion was submitted by the Commission, has pointed to that apparent fall in Finncell's prices, which was followed by a fall in the prices of two Swedish undertakings. His explanation of that fact is that it was the result of Finland and Sweden entering a recession, which, he claims, bears out the view that peaceful cooperation breaks down in times of economic recession.

381.

A careful examination of the documents before the Court, which I carried out earlier in this regard, shows that the data in Table 6 are inaccurate so far as concerns Finncell's behaviour at the beginning of 1977. As I said earlier, the Commission demonstrated in its defence, on the basis of telex messages, that the prices announced by Finncell remained unchanged for the first three quarters of 1977 contrary to what is set out in Table 6.

382.

There are two observations I wish to make in that regard.

383.

First of all, although I am aware of the considerable administrative burden entailed by this case, I find it regrettable that the decision contains inaccuracies which in the circumstances are far from negligible. I find it equally unsatisfactory that the defendant, in its observations on the expert's report, failed to remind the Court that the data in Table 6 concerning Finncell's prices for 1977 were incorrect.

384.

Secondly, the fact that two conflicting explanations could be given for data that is incorrect justifies a cautious approach to what may be deduced with any certainty from what is shown by economic arguments. Allow me simply to point out, however, that the experts appointed by the Court have shown, it seems to me, a degree of caution in explaining the conduct allegedly engaged in by Finncell at the beginning of 1977.

385.

The experts did indeed state that the theory of competition suggests that it was confirmed by the data in Table 6. But at the same time they noted that Finncell was ‘in a weak position to sustain a long-run price war’ and that the decision to cut prices was made ‘in spite of longer-term strategic considerations’. The experts acknowledged, moreover, that analysis of why prices suddenly fall after periods of stability is one of the least satisfactory elements in the economic model of pricing behaviour under oligopoly. In that regard, they raised the question why Finncell decided to cut prices given the oligopolistic nature of the market in the short term. They then went on to list the reasons which seem to have led the Finnish producers to take that decision.

386.

Although I thus find that the experts did not state that Finncell's unilateral reduction was unavoidable at the beginning of 1977, it must be observed that their conclusions are in any event ‘unusable’ in connection with that particular point since, contrary to the information incorrectly given in the decision, that reduction did not take place.

387.

I shall confine myself to consideration of four of the five aspects of price uniformity contemplated by the experts. ( 174 )

388.

In this context, the experts considered that for any category of pulp within which there is complete elasticity, there may be taken to exist a near-perfect market for the following reasons:

389.

According to the experts, the principle of identical prices for identical products in a perfect market is not negated by differences in unit costs. Equality of prices results from the transferability of demand, which is independent of costs of supply. Under conditions of perfect competition, producers would supply additional units until opportunity costs were equal to the market price. On a competitive oligopolistic market, which fits some of the short-term market situations in the pulp sector, prices exceeded ‘opportunity costs’ but were equal for all suppliers.

390.

Let me point out first that, in reply to a specific question at the hearing, Mr Cockram stated that the phrase ‘well-developed informal network of information within the industry’ related to contacts between purchasers of paper pulp and not between the sellers themselves.

391.

The Commission's criticisms of that part of the report essentially reiterate certain objections which I have already considered (failure to examine the manner in which prices were announced, uniformity and simultaneity of announcements, reference to the perfect competition model, theory of short-term oligopoly, inadequate analysis of the level of prices). I therefore refer to my earlier observations on those points.

392.

Furthermore, the defendant considers that the experts failed to examine in their report whether the high degree of market transparency came about naturally or artificially. The Commission points out that the experts stated that transparency had worked uniformly in favour of sellers — by discouraging price cuts and facilitating price increases during a market upswing — and considers that the producers had every incentive to have recourse to artificial devices and that there was ample direct evidence of unlawful exchange of information between producers.

393.

I have already examined the Commission's analysis of the artificial nature of the transparency which is said to result from the system of price announcements. I shall return in due course to the question of the use in the contested decision of direct evidence of exchanges of information.

394.

The experts stated that, in view of the fact that pulp is not homogeneous, prices in a perfect market would be expected to vary according to quality. They pointed out that in 1975 and 1976, southern hardwood pulp was sold at a lower price than northern hardwood pulp.

395.

With regard to softwood, however, the prices announced for northern softwood and southern softwood — the latter being of lower quality — were the same in 1975 and 1976. A differential only became apparent in the first quarter of 1977.

396.

According to the experts, substitution of southern softwood for northern softwood (NBSK) requires a slowing-down of the production process or investment in new equipment, both of which imply the need for a price difference. The fact that there was no such difference in 1975 and 1976 may reflect the relatively recent introduction of southern softwood pulp into the Community, so that the need for a differential had not yet become established. The high level of demand in the United States and almost 100% capacity utilization of southern United States producers were probably other factors involved.

397.

The Commission has emphasized in that regard that if the use of southern pulp had the impact on the papermaking process referred to by the experts, the added cost should have been reflected in the prices. The defendant's objection would not seem to be wholly unfounded. It would have been helpful if during the oral procedure the experts had been able to explain the reasons which led them to suggest that explanation which, in their report moreover, is put forward as a matter of conjecture.

398.

The experts pointed out, on the basis of the invoices forwarded by the producers from British Columbia, that discounts offered on small orders were apparently greater than those offered on large orders. They explained that result by the willingness of regular customers to pay announced prices or prices close to the average prices in order to ensure supplies and by the fact that fidelity rebates would not be shown on individual invoices and might not be recorded for that reason. The Commission emphasized in that connection that those two explanations are mutually exclusive.

399.

I shall not dwell on this point since I consider that the Court will in any event have to annul the findings of concertation on transaction prices set out in the decision even if the Commission had not also called in question the existence of secret discounts as evidence of concertation on announced prices. In tracing the historical background to the period prior to 1975 on the basis of their interviews with pulp purchasers, the experts stated that papermakers welcomed at the time the existence of a uniform list price because their main concern was to ensure that they paid no more for pulp than their competitors. There was a widespread belief ( 175 ) that firm pulp prices helped to stabilize prices for paper and that pulp price reductions led to demands for lower paper prices. Purchasers were therefore in favour of high announced prices, which were known to their own customers, but were quick to exploit market weaknesses by negotiating confidential rebates that did not necessarily appear on invoices. Some were based on annual tonnages purchased and could not be calculated until the end of the year. It was even possible for a papermaker to receive a cheque, without any paperwork whatsoever, but understood to be for fidelity and long-term business. That situation is said to continue today.

400.

Professor Neumann considered that secret discounts were evidence of collusion whose existence, moreover, was established by the experts for periods prior to 1975 ( 176 ) since in the absence of collusion secrecy would not be necessary.

401.

From that point of view, the Commission stated during the written procedure that the grant of rebates by a separate credit note, as was allegedly done, would make sense only if the producers wished to give their competitors the impression that transaction prices were the same as announced prices and wished to conceal any deviation from the pricing policy of those competitors. In so far as each purchaser compares the net prices he has to pay to a potential supplier, genuine rebates cannot be concealed from the customer even if they are given some other designation. However, they can be concealed from competing producers who are unable to check whether a discount is justified to offset poorer quality or transport costs or whether it is a competitive discount.

402.

There can be no doubt, in my view, that the concealment of rebates may, in certain circumstances, be regarded as ‘cheating’ on concertation. Is that the only possible explanation? The applicants were at pains to challenge that conclusion, even though their own arguments are not always persuasive. That is the case, for example, with regard to the producers' concern to prevent their competitors from discovering a rebate in an attempt to regain the customer. A customer can after all always disclose to the supplier's competitors the actual price he has paid, whether or not the discount appears on the invoice. However, I would attach more weight to the explanation that the concealment of a rebate may reflect the intention of the seller or the customer — or both — to conceal the rebate from the latter's competitors, since the seller has no wish to grant such a discount again and the purchaser wishes to retain his competitive advantage. However, that explanation presupposes that purchasers show their invoices to one another, which clearly raises other questions as to the reasons for a practice of that kind. I would simply point out that from a document produced during the proceedings the — no doubt tenuous — inference may be drawn that the practice of separate credit notes does not necessarily reflect the will of the sellers themselves. ( 177 )

403.

Let me say, moreover, that, in support of its view that the practices in question afford evidence of concertation, the Commission considers it unsurprising that most of the applicants refuse to allow other producers to inspect their invoices and related documents. In my view, that is a bold interpretation, with regard to which the producers from British Columbia have very humorously pointed out that had the producers permitted such inspections, the Commission would no doubt have viewed this as evidence that they had been in the habit of exchanging information.

404.

According to the experts, in paragraphs 136 to 140 of its decision, the Commission considered that the producers had foregone the opportunity to exploit differences in market conditions in different Member States. In their view, that analysis is incorrect. In the first place, if there was no barrier to resale, international trade (arbitrage) would eliminate any price differences between the different States. Secondly, the absence of collusion is evidenced precisely by the absence of price differences between Member States. An international structure of different prices would have been possible only if it had been collectively implemented. In the absence of an effective cartel, embracing all existing and potential suppliers, and of barriers to resale, international price discrimination would be impossible. In the absence of collusion, prices would therefore be the same even in market segments with different demand conditions.

405.

In its written observations, the Commission considered that the experts had essentially addressed the issue of the effect on competition between Member States, disregarding the question whether the use by the producers of the United States dollar was a result of the natural operation of the market. For my part, I have already considered the argument concerning the use of the dollar by all the producers. Let me point out, moreover, that paragraph 99 of the decision, which contains an economic analysis of price parallelism, refers to the differences in demand conditions between the European States in support of the view that uniformity of prices cannot be explained without concertation. In that passage the Commission seems at first sight to be suggesting that, in the absence of concertation, prices should have varied from one Member State to another.

406.

In my view, however, it is impossible to elicit any decisive indication of the absence of concertation from the experts' conclusions in that regard. Admittedly, the fixing of prices which vary from one State to another may be regarded as constituting a highly organized form of cooperation between undertakings, which is designed to maximize profits on each ‘national segment’. From that point of view, the absence of differences in price between Member States may serve to demonstrate that there was no such cartel on the pulp market with an organized strategy in order to maximize profits. However, the absence of differences in price in no way proves that there was no concertation. Common sense suggests, surely, that a single price may be far easier to fix or to ‘apply’ by the participants in a concerted practice. Accordingly, the most that can be said, in my view, of that part of the report, is that the absence of price differences between the different countries could be explained in theory by the absence of any barrier to resale and does not therefore necessarily constitute evidence of concertation. Mr Fishwick has stated that he ‘understood’ that the clauses precluding export and resale were inoperative. In addition, certain applicants contended in the written procedure that because, in particular, of transpon and storage costs, arbitrage was relatively unattractive. Mr Fishwick disagreed with that statement on the ground that any differences in price would be cleared by arbitrage on the part of papermakers who purchased pulp for use in different parts of the Community and that, in any event, even the threat of arbitrage would be sufficient to prevent any differential pricing.

407.

Under this heading, the experts stated that although price uniformity could be explained as a natural consequence of the structure and trading conditions of the pulp market, that does not per se prove that there was no concertation. In a passage which I consider should be cited in its original version, Mr Cockram and Mr Fishwick expressed themselves in the following terms:

‘Transparency and interdependence provide conditions likely to induce some collusion, but there are several indications from our analysis of the wood pulp market between 1975 and 1981, that any collusion which occurred was limited in its effects’.

408.

According to the experts, the following factors were involved:

409.

Let me point out that, at the hearing, the experts were in fact asked whether they considered that the uniformity of prices was to be explained by concertation or by the natural operation of the market. Mr Fishwick reminded the Court — quite rightly, indeed — that it had asked the experts not whether concertation existed but whether the uniformity of prices could be explained by natural forces in the market, and added that the latter was the more probable explanation. In his view, the aforesaid factors would have thwarted any attempt to concert effectively. He pointed out, however, that those same factors may have been consistent with ineffective collusion.

410.

I consider that the existence of competitors from outside the group of addressees of the decision and the changes in market shares call for two series of observations in that the decision seems to me manifestly ambivalent on those points.

411.

It should be borne in mind, first of all, that, as is clear from the decision itself, its addressees accounted for approximately 60% of total consumption of bleached sulphate pulp in the Community. Certain applicants, in particular the members of KEA, regarded as implausible the Commission's view that concertation took place between the producers covered by the procedure. They contended that the defendant was unable to explain how concertation could have worked since producers accounting for 40% of production could have constituted an alternative for papermakers if concertation had existed between the undertakings covered by the decision.

412.

The Commission's explanations in that regard are far from clear. In its defence, it stated that the producers in question were Outsiders' who usually sold only small quantities of pulp and were not regarded by papermakers as an alternative source of supply to the addressees of the decision. Historically, it says, the pulp market in the Community was influenced by the addressees of the decision and not by the ‘outsiders’ who could regularly be expected to be followers.

413.

I would note at the outset that it is a priori surprising that producers accounting for 40% of consumption should, taken together, be described as ‘outsiders’, with no further qualification other than the bald statement that they sell small quantities.

414.

In any event, the Court will find it impossible on the basis of the information submitted by the Commission to determine which undertakings supply the 40% share of Community consumption which is not accounted for by the addressees of the decision. In reply to a number of questions from the Court, the Commission explained, in the first place, that most of the producers who were addressees the Statement of Objections but were not the subject of a finding of infringement sold only small quantities of bleached sulphate pulp or did not manufacture it, and secondly that the producers who were never involved in the procedure and whose invoices it had checked dealt entirely, or almost entirely, in a product other than bleached sulphate pulp. Therefore, the Commission did not deem it necessary to pursue enquiries into their pricing policy. In the light of those explanations, all I would say is that the Commission's analysis of the market is quite inadequate so far as concerns the composition of supply on the pulp market.