High Court of Ireland Decisions

1. The Plaintiff (hereinafter "Chanelle") is a wholesale distribution company which was founded in 1980. The directors and principal shareholders are Michael Burke and his wife Mary Burke. Michael Burke is a qualified veterinary surgeon and practised as such for several years before setting up the Plaintiff company. He did so because he was unhappy with the distribution services offered by the multinational companies and he set out from the beginning to provide a twenty-four hour delivery service of the full range of animal health products. The business of Chanelle prospered and today enjoys a turnover of approximately £9 million out of a total animal health products market of some £55-£60 millions. Chanelle is part of a group of companies which includes Chanelle Pharmaceuticals Manufacturing Limited ("CPML") which was formed in 1985.

2. Amongst the suppliers serviced by Chanelle in the 1980s were Beechams Plc. and Smith Kline French Limited. These two companies merged in the early 1990s to become Smith Kline Beecham. In January 1995 the Pfizer Group of Companies, of which the Defendants are part, purchased the world wide animal health business of Smith Kline Beecham Plc., and as part of this take-over the first named Defendant took over the animal health business of Smith Kline Beecham in Ireland and commenced trading as Pfizer Animal Health. Chanelle continued to act as wholesaler and distributor of the Pfizer Animal Health range of products. In this way in 1996 Chanelle had acted as wholesaler or distributor for a decade and a half for the Beechams Plc. and the Smith Kline French range of products, now part of the range of Pfizer Animal Health.

3. The first named Defendant ("Pfizer") is the Irish subsidiary of a US Corporation, Pfizer International Inc. Pfizer carries on business from Pharmapark, Chapilizod, Dublin 20. Pfizer International Inc., is a multinational corporation with subsidiaries or branches in many countries throughout the world. It is a world leader in many veterinary and other pharmaceutical products.

4. Pfizer is the second largest supplier of animal health products in the Irish market the largest being Merck Sharp & Dohme which is a single product (Ivomec) company. The third largest supplier is Mallinckrodt. Pfizer has approximately twelve per cent of this market. Prior to the take-over of Smith Kline Beecham, Pfizer products in Ireland were distributed on an exclusive basis by Caffrey & Mallon in Limerick ("C&M"). Shortly after the take-over in January 1995 Chanelle was contacted by Pfizer who were considering appointing them as stockholding agents for the new company. After detailed negotiations this contract was given in November 1996 to Cathal May Roberts a wholesale distributor who also distributed the human range of Pfizer products.

5. The second named Defendant has been joined in these proceedings as the employer of one Mr. Johan Rabie and has offices at Louvain-la-Neuve, Belgium. As will be detailed hereafter, Mr. Rabie was involved to some degree in the events which lie at the heart of the dispute between the parties.

6. After the take-over of the Smith Kline Beecham range Pfizer continued the exclusive distributorship through C&M for Pfizer products, and Smith Kline Beecham's distribution system comprising sales to wholesalers and direct sales to customers also continued for several months. At the end of the year apart from appointing a national stockholder Pfizer decided to appoint a limited number of wholesale distributors. In this connection a meeting took place on the 29th November, 1995 between Chanelle and Pfizer. As a result of this and other meetings five wholesale distributors were appointed and the Pfizer products were available only through these five. The other four were Boileau and Boyd, Co-operative Animal Health, Agri Health and C&M . These five wholesale distributors were the largest wholesalers in the distribution market.

7. At the meeting on the 29th November, 1995 the parties reached agreement on the terms of the distributorship. The wholesaler's discount was 12.5%, a fee of £10,000 was payable for the monthly supply of information on Pfizer sales extracted from Chanelle's computer, credit was sixty days and the territory of the agreement was the Republic of Ireland. The discount represented a reduction from a previous 15% for Smith Kline Beecham products but this was offset by the increase in business resulting from the decision of Pfizer to distribute its goods exclusively through the five appointed wholesalers.

8. In parallel with these arrangements Pfizer also operated a rebate scheme which was available to those who purchased their range of goods through the five appointed wholesalers. Depending on quantity the maximum rebate was 10%. This rebate was paid monthly to end purchasers and was calculated on a rolling twelve month basis.

9. Mention should be made of a meeting between the parties which took place on the 12th February, 1996. There were complaints on either side. For their part Tom Mullaney (Managing Director) and Philip Howlan (Sales Manager) of Pfizer were, according to Michael Burke, very dissatisfied with Chanelle because they were giving away a 6% discount out of the 12½%. Chanelle for its part complained that other wholesale distributors were being given 15% discount on Dectomax which was the number one success product from Pfizer. Chanelle's response to the discount complaint was that they were followers, that others had done it before and they agreed that in one case a personal friend in Galway of Michael Burke had been given this discount. Pfizer's response to Chanelle's complaint was that they denied that other wholesale distributors were being given 15% and insisted that Chanelle's discount remain at 12½%.

10. A second matter of some significance is the fact that Valbazen which is a very successful white wormer supplied by Pfizer had come off patent some time before this. Chanelle's sister company CPML was engaged in developing a generic competitor to Valbazen namely Albex which was launched at a trade show in Athenry on the 25th May, 1996. In a telephone conversation on the 22nd May, Michael Burke notified Tom Mullaney of Pfizer that this launch was about to take place. In the course of that telephone call he asked Tom Mullaney whether he had any complaints about Chanelle and whether Pfizer was happy with them. He was told "You are doing an excellent job: no complaints". It was following this that Michael Burke told Tom Mullaney that Chanelle were bringing out a generic Valbazen and he got the impression that Tom Mullaney was a bit taken aback. He had made the telephone call in the first place because during the previous year Tom Mullaney had asked him on three or four occasions whether Chanelle were bringing out a generic Valbazen. He felt he owed it as a matter of courtesy to inform Tom Mullaney before the actual launch.

11. Following this Tom Mullaney tried to arrange a meeting between himself and Michael Burke between the date of the telephone call and either the 7th or 10th of June. He wouldn't give an agenda for the meeting. Because of conflicting commitments neither date proved possible and the meeting was never held.

12. On the 18th June, 1996 Michael Burke was in London. That morning Tom Mullaney delivered to Chanelle at its premises in Loughrea a letter dated 17th June indicating that it had been decided to implement new distribution arrangements for their animal health products from the 24th June, 1996. From that date the main wholesale distributors set out in the letter excluded the Plaintiff. The letter referred to the two failed meetings as business review meetings.

13. This decision to "de-list" Chanelle came without warning and immediately Michael Burke attempted to make contact with Tom Mullaney. He wanted to know why his company had been "de-listed". He failed initially to make contact and subsequently did so but did not succeed in getting a clear reason. Subsequently it has been suggested that Chanelle was not wholly committed to the Pfizer animal health product range, that it had made wrongful use of private or confidential information belonging to Pfizer or, alternatively, that it was in a position to make use of such information and there was a suggestion that it had not been prompt in paying monies it owed to Pfizer.

14. It is clear from the conduct of the defence, however, that the main stand of the Defendants is that the agreement between the parties is governed by common law and common law alone. They claim to have given adequate notice and indeed Chanelle makes no case based on notice. The relevance of the Defendants' reasons, stated or otherwise, for "de-listing" Chanelle is the bearing they may have on the complex issues arising under Irish Competition Law and Articles 85 and 86 of the Treaty of Rome.

15. Having failed to procure a satisfactory reason for their "de-listing", Chanelle instituted these proceedings and brought an application for an Interlocutory Injunction in July of last year. This was resolved by the Defendants' undertaking to continue to supply Chanelle on the same basis as theretofore until the determination of this action.

16. Chanelle says that the Pfizer range of animal health products is an essential part of its business. It is essential that Chanelle has access to this range which comprises approximately eighty different animal health products many of which are unique to the Defendants and many of which are market leaders in their own right. The Defendants are part of a world wide animal health producer and Chanelle claims that its wholesale business will not be viable without access to the products of the Defendants' range on the same terms as the other four appointed wholesalers.

17. Many of the Defendants' products are in a special if not unique position in the market according to chanelle. For example, Valbazen is the number one white drench product in Ireland for the treatment of endoparasites in cattle and sheep. It is the only white drench product effective against all types of stomach round worms and lung worms as well as adult fluke and fluke eggs. This arises because of its unique combination of chemicals. Furthermore, Valbazen is the number one world-wide Pfizer product. About 75% of chanelle's business is with vets. These tend to prefer to purchase all their supplies from one wholesaler. Chanelle also says that there is a preference in the market-place for brand leaders. All of his means, according to chanelle, that its wholesale distributor business will be damaged and possibly lost if it cannot have access to the Pfizer range of animal health products on the same terms as the other four appointed wholesalers.

18. The pleadings articulate a number of different claims and grounds of relief. These can be conveniently summarised as follows:-

19. In either event Chanelle says that the object of the agreements was anti-competitive. Furthermore, Chanelle says that the effect of such de-listing is to restrict or distort competition in a way which offends against Section 4(1) of the Act of 1991. Chanelle says not only that competition in the trade of distributing Pfizer animal health products in Ireland has been restricted or distorted but that, so also, has the market for the end user.

20. At the conclusion of the Plaintiff's case Counsel for the Defendants applied to have all claims dismissed on the grounds that there was no case. I ruled on this application in my judgment delivered on the 5th June, 1997. In the result I held that a prima facie case had been made out in relation to a breach of Section 4 of the Act of 1991 arising out of an agreement or concerted practice between Pfizer and the remaining distributors and also an agreement or concerted practice between Pfizer and the second named Defendant. I held that no case had been made out in relation to breach of a dominant position by Pfizer in any of the markets identified with the exception of a market for the product Synulox. I held, further, that a prima facie case had been made out to establish appreciable inter-State effect. The case proceeded thereafter in accordance with this Ruling.

21. Before dealing with the issues that have arisen, I propose to explain a little more about the background to the various claims.

22. Initially (at the stage of the Interlocutory Application) Mr. Burke made the case that de-listing chanelle would mean that it would not be able to carry the Pfizer range of products, that the customers would automatically switch to other suppliers who are chanelle's competitors and since the Pfizer cash save rebate scheme would not be available to these purchasers, chanelle would be put at an enormous disadvantage with the possible loss of £2.2 million turnover in a twelve month period implying lost profits in the region of £400,000 to £500,000 per annum. In the course of these proceedings it was made clear by Counsel on behalf of the Defendants that the rebate would be available to Chanelle in the event that it purchased Pfizer products from one of the four appointed wholesale distributors just the same as it would be available in the case of any other purchaser. This contrasts with the plea at paragraph 22 of the amended Defence which places emphasis on the availability of these products by way of import from other EC countries but I note that in an earlier letter of 24th June, 1996 Pfizer's solicitors did indicate that their client's product would continue to remain available to Chanelle after "de-listing". Furthermore, the point was made that because Chanelle's purchases would be large it would qualify for the top rate of rebate at 10%. Michael Burke and Mr. Moore McDowell, an economist, who gave evidence on his behalf, both commented that these four wholesale distributors are in direct competition with chanelle and would be unwilling to supply goods on terms which would enable it to compete effectively in the wholesale distributor market. In the course of cross-examination Counsel on behalf of the Defendants put to Mr. Burke the following questions:-

23. In commenting on the implications of "de-listing" for Chanelle as a wholesaler Mr. McDowell noted that it was commercially necessary for Chanelle to be able to offer a very wide range of products to potential customers. The reason for this was a widespread preference on the part of end users for "one stop shopping" even when they take supplies from more than one wholesaler. This witness noted that Pfizer had a share of around 15% in the total sales of animal health products in Ireland and said that de-listing Chanelle implied depriving it of the ability to offer the products of one of the top three producers in markets covering nearly half of the total animal health product sales in Ireland. This would clearly have the effect of seriously undermining its competitiveness as an independent wholesaler.

24. Mr. McDowell's evidence and report was initially prepared on the assumption that Chanelle would not be eligible for volume rebates after de-listing. This has been stated to be incorrect by Counsel on behalf of the Defendant and Mr. McDowell accepted that many of his conclusions would have to be modified in light of this information. He did not, however, withdraw any of them.

25. Mr. Tom Tierney, group financial controller of Chanelle, specified six consequences of "de-listing" which would apply even if the rebate is made available to it as it is to any other purchaser of Pfizer products. These included that Chanelle would not get a wholesaler's margin ("the wholesaler margin is how you survive") they would not get "referred orders" (that is orders referred to a wholesaler from Pfizer representatives in the field) and there were a number of promotional devices which would not be available to Chanelle.

26. In commenting on the fact that Pfizer's rebate system is given to end users rather than to distributors Mr. McDowell identified an element of loyalty bonus which he said reduces substitutability between Pfizer's products and those of other producers. In the context of wholesaling, he said that the rebate system:

27. A separate aspect of Chanelle's claim is that there is an element of "disciplining the market" in Pfizer's decision to "de-list" Chanelle. This arises in the following way. Valbazen is an extremely successful product of the Defendants both world wide and on the Irish market. Chanelle's competing product Albex was launched and competes directly with Valbazen. The consequence of de-listing not only weakens Chanelle as a competitor but also sends a clear signal to other distributors that the introduction of generic competitors for Pfizer's brand leaders is likely to be a costly exercise. This creates a barrier to entry to the market for the supply of generics. In this context I note that Michael Burke in a telephone conversation with Johan Rabie on 19th June, 1996 suggested a formula where "we would not bring out any further competing generics".

28. In a further submission Chanelle alleges that Pfizer's distribution system is what is termed a "selective distribution system". In principle a supplier is free to determine how his goods are to be distributed. (See for example Butterworths Competition Law , Chapter 4, para. 568). He may appoint a sole agent, he may distribute himself, he may select any number of wholesale distributors. If, however, he selects distributors and places them under an obligation to sell his goods only to end users or other distributors who have been appointed by the supplier this amounts to an intervention in the market place which requires justification. In order to justify such an intervention the goods themselves must be suitable for selective distribution. Examples of categories of goods which have been held suitable for selective distribution are goods that are technically complex, goods the brand image of which is particularly important, and goods with an extremely short shelf life necessitating particularly careful distribution. If a selective distribution system is not to run counter to competition law the products must be appropriate for such treatment, the criteria for selection of wholesale distributors must be qualitative and the system must be operated in a non discriminatory way which means that any wholesale distributor who qualifies under the qualitative criteria is entitled to be appointed a distributor.

29. Chanelle argues that the Pfizer distribution system is in fact a selective distribution system because whilst the appointed distributors are not restricted from selling only to end users or other distributors who have been appointed by Pfizer, the effect of the rebate system with its "loyalty tie" characteristics produces the same result. Appointed distributors will not sell otherwise than to end users and fellow appointed distributors because the only way the end user will qualify for the Pfizer rebate is to purchase from an appointed distributor. It is further argued that even if Pfizer's distribution arrangements do not amount to a selective distribution system in the strict sense, it is nonetheless a restricted distribution system which is contrary to Section 4 and Article 85 because it is operated in a discriminatory manner with anti-competitive effect, namely, Chanelle has been de-listed notwithstanding that it qualifies under all the appropriate qualitative criteria. The significance of a distribution system being a selective distribution system properly so called, is that there is authority to say that it could constitute an agreement or concerted action for the purposes of competition law despite appearing at first sight to be no more than a unilateral action.

30. Quite apart from the foregoing, Chanelle had alleged that in a number of identified markets the Defendants are in a dominant position and that withdrawal of supply constitutes abuse which should be condemned. In my judgment of 5th June, 1997 I restricted this aspect of Chanelle's case to an alleged market in Pfizer's product Synulox.

31. The identified markets appeared in a letter from Chanelle's solicitors dated the 25th April, 1997. Four of these are product markets identified in the Animal and Plant Health Association Report (the "APHA report") which has been introduced in evidence. The APHA Report is a quarterly compendium of data in relation to the animal health market and divides the range of products into categories and sub-categories where a specific product is identified and sales figures given. Not all suppliers to the animal health market are members of APHA but the evidence shows that approximately 90% of the volumes in each market are included.

32. Thirdly, it is claimed that the de-listing of Chanelle will have an effect on the markets in other member States of the European Community. Chanelle points to the fact that the first named Defendant imports products from the United Kingdom, Belgium and France as well as the United States and also to the fact that its sister company exports the bulk of its manufactured products from this country. Chanelle's Counsel cites Bellamy and Child , Common Market Law of Competition , 4th ed., at paragraph 2-129, where the test laid down by the Court of Justice for assessing whether trade between Member States will be affected is as follows:-

33. I now turn to consider these claims and will deal first with the issue of alleged agreement or concerted practice.

34. In order to make a case under Section 4 or Article 85, it is clearly necessary to establish that there is an agreement or concerted practice. This topic is introduced in a general way in Bellamy and Child (op. cit.) at paragraph 2-015 as follows:-

35. Counsel for Chanelle has submitted that the decision to de-list his client was part of both an agreement and of a concerted practice. On day 19 of the hearing Mr. Tom Mullaney was questioned and gave answers as follows:-

36. On day 23 the same witness was cross-questioned and answered as follows:-

37. It is abundantly clear from the foregoing that there was no prior communication between Pfizer and the other four wholesalers. They were informed of the situation after the event. They had no part in reaching the decision. They were told what the future arrangements were going to be and they accepted it. In light of this evidence there can be no question of an agreement between Pfizer and the other four wholesalers in relation to the de-listing of Chanelle. I will consider later whether a concerted practice can be inferred from the continuing business relationship between these parties.

38. Before I do so, however, I must first consider whether there was an agreement in the ordinary sense of that term between the first Defendant and the second Defendant.

39. Mr. Michael Burke says that on hearing of the "de-listing" letter on the 18th June 1996 he telephoned the second Defendant, whom he had known a few years before when Mr. Rabie was working in South Africa but whom he knew at the time was working for a Pfizer company in Brussels, to see if there was anything that could be done for Chanelle which had just been dropped as a wholesaler in Ireland. Mr. Burke says that in that phone call Johan Rabie said that he had been involved in the fringes of the decision, that Chanelle had been dropped because they had a competing product with the Pfizer Valbazen product and that Pfizer were also afraid of what other generics Chanelle might be working on. Mr. Burke was staying at the Sheridan Hotel in London at the time and he had a little Sheridan Hotel pad with him on which he made notes of the telephone conversation and this pad has been produced in evidence and shows that such notes were in fact made. The words "laundered generics", "effected business", "involved on fringes" and "working on more generics" were written by Mr. Burke on the Sheridan Hotel pad.

40. Mr. Rabie agrees that he spoke on the telephone to Mr. Burke but refers to several telephone calls as follows: The first, towards the end of April or beginning of May, was when Mr. Mullaney rang him being aware of some arrangement between Smith Kline Beecham (now Pfizer) in South Africa and Chanelle and indicating that he was reviewing the distribution agreement between Chanelle and Pfizer in Ireland and wondering whether any decision made in this regard would have a material effect on the business in South Africa. Mr. Rabie indicated that he was not really able to answer but he would contact Mr. Mike Hagamman in South Africa who was General Manager for Pfizer. He did so and was told that the relationship in Ireland would not have a material effect on their relationship with Chanelle. Mr. Rabie phoned Mr. Mullaney with this information. Subsequently, some weeks later towards the third or last week of June 1996, he got a telephone call from Mr. Burke. This was one of two phone calls on that day from Mr. Burke. In the first phone call

41. Mr. Rabie says that he was able to say he had some idea of what was going on but that he was certainly not involved in any of the decisions and never had been. He told Mr. Burke that the policy decisions were taken locally. He was asked to try and contact Mr. Mullaney. He agreed to do that and did so. Mr. Burke rang him a second time and on this occasion he confirmed that his involvement in the decision was limited to the phone call that

42. Mr. Mullaney made to him in the first place asking about the nature of the agreement between Chanelle and Pfizer in South Africa. In the second phone call Mr. Rabie said he informed Mr. Burke that Mr. Mullaney had told him that Chanelle had planned to or had introduced a generic Albendazole and that Chanelle were targeting the Valbazen customers with this new compound. He said that the second phone call after he spoke to Mr. Mullaney was made by Mr. Burke phoning from London Heathrow just prior to him leaving for a flight back to Ireland.

43. He also stated in evidence that his responsibilities in his job in Brussels did not involve decisions in relation to wholesale distributors.

44. The clash of evidence between these two witnesses relates, in essence, to whether Mr. Rabie was able to communicate the reasons for the de-listing decision to

45. Mr. Burke on the first telephone call (as Mr. Burke says) or on the second telephone call after Mr. Rabie had communicated with Mr. Mullaney in Ireland in the meantime and been given the relevant information (as Mr. Rabie says). I consider that Mr. Burke's recollection is likely to be the more accurate. He says that the explanations were given by Mr. Rabie on the first phone call made by himself from the Sheridan Hotel. He made notes of the essential points on the Sheridan pad which was produced in evidence. If Mr. Rabie's account is correct, the reasons were given by him to Mr. Burke in the second phone call when Mr. Burke was at Heathrow Airport on his way back to Ireland. If that happened it is unlikely that the notes of the phone call would have been made on the Sheridan pad. I consider that Mr. Burke's recollection is more likely.

46. I accept, however, that Mr. Mullaney did telephone Johan Rabie to enquire whether the proposed de-listing of Chanelle might affect the South African company and was given this information. I accept further that Mr. Mullaney was merely making a general enquiry and was not involving the second-named Defendant or indeed Mr. Johan Rabie personally in any significant way in a decision-making process. I think it likely that if

47. Mr. Mullaney wished to have an input into this decision-making process from the second-named Defendant, he would have contacted Mr. Rabie's superior, Mr. Jo Jensen. I further accept that the purpose of Mr. Mullaney's phone call to Mr. Rabie was simply to ascertain the viewpoint of the South African company and because he was aware that Mr. Burke had had links with that company through Mr. Rabie. This was the evidence of Mr. Rabie. Mr. Mullaney also gave evidence that he did not speak "higher up the tree". Subsequently the Brussels company was informed when the new system was put in place. This was after the event.

48. In my view the second Defendant was not involved in the decision to de-list Chanelle and was not asked to endorse or approve it. Mr. Rabie was consulted in advance on a point of information by Mr. Mullaney and the second-named Defendant company was informed after the event of the new distribution system. In my view there was no agreement between the first and second-named Defendants in relation to the de-listing of Chanelle.

49. Chanelle contends, further, that Pfizer's distribution arrangements, which exclude Chanelle, amount to a concerted practice within the meaning of Section 4 and Article 85. Reference is made to Bellamy and Child (op. cit. ), Chapter 2 Part III generally, and in particular at 2-022 as follows:-

50. Counsel for Chanelle argues that the apparently unilateral decision of Pfizer to "de-list" Chanelle when seen in its true context, namely in the context of a selective distribution system or indeed a restricted system of distribution as referred to in the above quotation, should be regarded as a concerted practice within the meaning of Section 4 and Article 85. Specifically it is contended that there is at least a tacit understanding between the supplier and the remaining four distributors to exclude Chanelle from the distribution network and that this tacit understanding "may be inferred from all the circumstances" , to use the phrase of Bellamy and Child at paragraph 2-016, which circumstances include the continued operation without protest by these four distributors of Pfizer's distribution arrangements which exclude Chanelle. Particular reliance is placed by Chanelle's Counsel, in this context, on the rebate scheme and its alleged anti-competitive features which are analogous to resale price maintenance control by Pfizer of their products on the market, an engagement by Pfizer in a form of loyalty rebate and a restriction of competition at the wholesale level. There is authority, says Counsel for Chanelle, to show that in all the circumstances the continued operation of this distribution system should be seen as a concerted practice carried on by Pfizer and the four remaining distributors. In this context Counsel refers to a number of authorities and because this argument is of such importance to Chanelle, I propose to deal with the most important of these, briefly, in turn.

51. In assessing existing case law the Court of Justice noted that it was clear that "it is sufficient that the undertakings in question should have expressed their joint intention to conduct themselves on the market in a specific way" for there to be an agreement within the meaning of Article 85(1). In dealing with the definition of the concept of a concerted practice the Court observed (at p. 1805):-

52. It was found in that case that the applicant had participated in meetings concerning the fixing of price and sales volume targets during which information was exchanged between competitors as specified above. The applicant had pursued the aim of eliminating in advance uncertainty about the future conduct of its competitors. It also took account of information during the course of the meetings already referred to.

53. This case establishes that evidence of parallel behaviour may amount to strong evidence of a concerted practice.

54. In joined cases 25 and 26/84 Ford-Werke AG -v- Commission (No. 4) [1985] ECR 2725, the supply company discontinued sales of right hand drive cars to its German dealers in order to protect its UK dealers because the German cars were cheaper. The distribution agreements constituted a selective distribution system which specifically left certain future decisions to be dealt with by the manufacturer. One such future decision was the decision to protect the UK dealers' prices by refusing to supply cheaper right hand cars to the German market. The Commission held that this decision was part and parcel of the agreement with the suppliers although it appeared at first sight to be unilateral. This aspect of the case turned upon the fact that the agreements specifically left this matter to be decided in the future by the manufacturer.

55. I would comment in passing that this case might be authority to show acceptance by the four distributors in the instant case of the terms of Pfizer's invoices which indicate that the rebate will be calculated by reference only to goods purchased from these distributors. There is nothing in the invoices, however, to indicate that Chanelle is to be de-listed or excluded.

56. The decision in AEG -v- commission [1983] ECR 3151, was intimately bound up with the Court's perception that the distribution system involved constituted a selective distribution network. The case concerned a complaint by certain distributors that this system was operated in a way which discriminated against them. In dealing with the argument by AEG that any decisions were unilateral the Court said (at pp. 3194-3195):-

57. Such a system was permissible

58. If the system was applied in practice in a manner incompatible with these principles then such a practice was unlawful, and in those circumstances

59. I regard this case as authority in the context of a selective distribution network because it is clear that the judgment of the Court is intimately bound up with the principles relating to such networks. Chanelle has submitted that the distribution system in the present case amounts to a selective distribution system and I will consider that argument at a later point in this judgment.

60. In addition specific reference is made to four cases referred to in a footnote to paragraph 2-022 of Bellamy and Child already cited. Two of these four AEG and Metro No. 2 cases have already been dealt with. A third case referred to in this footnote is Holloran -v- Thwaits [1989] 2 CMLR 917. This is the judgment of Peter Gibson J. in the Chancery Division of the English High Court on behalf of the tenants of two tied licensees of public houses who had been served with notice to quit by the brewery which was offering a new agreement which it claimed was in conformity with EEC law. The plaintiffs claimed that it was not and sought an interlocutory order prohibiting the brewery from implementing the notice to quit. As part of this submission Counsel for the plaintiffs relied on the AEG case already referred to in order to overcome an argument that the brewery's decision was unilateral. The plaintiffs' argument on this aspect was accepted at the interlocutory stage and injunctions were granted.

61. The final authority I am asked to consider in this context is the case of Tipp-Ex [1989] 4 CMLR 425. This case arose out of distribution arrangements between Tipp-Ex and its exclusive distributors in several EU countries. These arrangements were aimed at preventing parallel imports or exports within the Common Market. The authorised dealer undertook not to sell the contract goods to customers who to his knowledge intended to re-sell them in areas outside the contract territory. Enquiries by customers from such countries were to be forwarded to the supplier. One of these dealers, ISA France, appeared to have breached these policies and re-sold products in the Federal Republic of Germany. It even told its customers where and at what prices it bought them. Tipp-Ex terminated the relationship. It sent several telexes to other exclusive distributors impressing upon them the policy of not supplying parallel importers. ISA France was succeeded by Beiersdorf. ISA France procured some Tipp-Ex products from another distributor, BDF Tesa. Beiersdorf wrote to BDF Tesa complaining about this and under pressure of repeated representations from Tipp-Ex expressed its readiness to support Tipp-Ex in the latter's policy of preventing parallel imports by ISA France. The Commission found that the authorised dealers adopted Tipp-Ex's ideas regarding the mutual protection of territories and that these ideas had become an integral part of their agreements. It accepted that the continuation of contractual relations between Tipp-Ex and the exclusive distributors was proof that these undertakings were prepared to accede to Tipp-Ex's wishes. At least one of its distributors explicitly agreed to endorse Tipp-Ex's policies and it is clear that the Commission was satisfied that Tipp-Ex took active measures to prevent parallel imports or exports and also exerted pressure on its exclusive distributors in order to obtain their agreement and support. The Commission's findings that the authorised dealers adopted Tipp-Ex's ideas connotes, in the context, something more, in my view, than mere passive acquiescence.

62. Counsel for the defendants, having submitted his own analysis of a number of the foregoing authorities submitted that in order to find a concerted practice there must be evidence of some form of positive co-operation in reaching it or of acceptance of it thereafter as a term of trade. The distributors in the instant case were not consulted in advance of Pfizer's decision to de-list Chanelle, they were informed of it over the telephone after the event and there was no discussion with any of them on the topic. They accepted the decision and it was clear that the continued business relationship would necessarily involve four wholesalers in the future.

63. In my view it is clear from the evidence that there was nothing like positive co-operation between Pfizer and the four distributors before the de-listing of Chanelle. It is equally clear that they were informed after the event. Their only involvement was to accept the fait accompli and react to the market as they found it. In this they were merely adapting themselves intelligently to the conduct of their competitors as defined in Hercules. There was no question of them being put under any kind of pressure to assist Pfizer in continuing the de-listing of Chanelle: there was no question of an invoice or other documentation which could conceivably be regarded as part of the ongoing terms of agreement which specifically supported the exclusion of Chanelle as a wholesale distributor. The facts of the present case are very far from those in Hercules for example where there were several meetings concerning price fixing and sales volume targets. There was no question of the remaining distributors being "asked" to co-operate with Pfizer in continuing the de-listing of Chanelle in a way that might be compared to the request by Konica UK to its UK wholesalers to co-operate with it by not exporting film directly or indirectly into Europe. ICI -v- Commission was a case involving parallel behaviour which was accepted as strong evidence of concerted practice. In the present case there was one act and one act alone: namely the act of Pfizer to de-list Chanelle. In Ford it was held that a contractual clause to leave over future decisions to the supplier meant that those future decisions, apparently unilateral, were in fact part of the agreements. The invoices in Sandoz were evidence of terms written on the invoices: the invoices in the present case have nothing to say about de-listing. AEG -v- Commission is a selective distribution network case and for the reasons already given the position is not really advanced by Metro 2 , Holleran or Demo Studio Schmidt . As I have already stated, in my view the Commission in Tipp-Ex had evidence of and accepted something more than mere passive tacit acquiescence on the part of the distributors.

64. In my opinion the acceptance by the four remaining distributors of Pfizer's de-listing of Chanelle and their continuation as distributors thereafter was not and is not a concerted practice for the purposes of Section 4 or Article 85. The evidence is quite clear that so far from being involved before the event they were merely informed after it. There is no evidence that they actively support or aid this policy of Pfizer in a manner analogous to the behaviour of those involved in the cases relied upon or even that Pfizer ever sought or desired such support. Pfizer's decision to de-list Chanelle was, in my opinion, a unilateral act and I hold, therefore, that it did not constitute nor was it part of any concerted practice as contended by Chanelle.

65. In case I am incorrect, in the conclusion I have come to in relation to the claimed agreements or concerted practices and in deference to the comprehensive submissions of Counsel, I think I should set out as briefly as I can, my views in relation to the other main issues in the case.

66. Chief among these is the question whether any agreement (or concerted practice) to de-list Chanelle has as its "object or effect the prevention, restriction or distortion of competition" in trade contrary to Section 4 of the Act of 1991 or Article 85 of the Treaty of Rome. I turn first to consider the anti-competitive object issue. Chanelle argues that the object of such an agreement is anti-competitive. It is clear from Société Technique Minière -v- Maschinenbau Ulm GmbH [1966] ECR 235 and Delimitis -v- Henninger Brau AG [1991] 1 ECR 935 that the correct approach is that the Court would first consider the object of the agreement before considering its effect. If the object is clearly and unambiguously anti-competitive then, subject to the de minimis rule, the agreement will be condemned and it will not be necessary to consider its effect.

67. Chanelle suggests that the object of any agreement between the first named Defendant and the four remaining distributors (or between the first and second named Defendants) is either to exclude Chanelle as an appointed distributor because it has misused allegedly private information for the purpose of promoting ("piggy backing") its own product, Albex, in competition with Pfizer's brand leader Valbazen. Alternatively the object of the agreement is to de-list Chanelle because it "targeted" Valbazen customers (known to Chanelle in its capacity of wholesalers/distributors for Pfizer's Valbazen) for the purpose of persuading them to purchase its own product, Albex, instead.

68. In ascertaining the object of an agreement the Court looks at the objective circumstances rather than the subjective intentions (declared or otherwise) of the parties to the agreement (see Bellamy and Child : op. cit. paragraph 2-097). "Horizontal" agreements (that is between undertakings at the same level of distribution) are more likely to be anti-competitive than "vertical" agreements (that is agreements between parties at different levels of the distribution chain). Bellamy and Child at paragraph 2-100 write:-

69. In dealing with this issue in Masterfoods Limited -v- HB Ice Cream Limited [1993] ILRM 145, at page 205, Keane J. noted the following:-

70. In the present case Pfizer has emphasised that Chanelle is perfectly free to purchase and re-sell its goods from other listed wholesale distributors. It will qualify for the rebate and it is likely to purchase in such volumes that it will qualify for the rebate at the highest (10%) level.

71. I am unable to hold that the "object" of any agreement involving the de-listing of Chanelle is anti-competitive per se as claimed by Chanelle because I cannot see that such an agreement is of a kind which restricts competition "of its nature". There has been a lively difference of opinion between the economic experts and others in this case as to whether competition (as distinct from the competitiveness of Chanelle) will be affected at all by this decision. Regardless of outcome, I consider that the fact that there can be responsible debate at all, demonstrates that the object of the agreement could not be restrictive of competition of its nature. I am compelled to hold, therefore, that any agreement involving the de-listing of Chanelle does not have as its object the prevention, restriction or distortion of competition.

72. Accordingly, I must now turn to consider whether the effect of such an agreement has been shown to be anti-competitive.

73. Before summarising the evidence submitted by the economists in the case, I intend to set out one or two principles of law.

74. Firstly, I accept the following statement by Costello P. in Donovan & Ors -v- ESB [1994] 2 IR 305 at page 322, namely:-

75. Secondly, I accept that the applicable standard of proof is proof on the balance of probabilities as stated by Keane J. in Master Foods Limited -v- HB Ice-cream Limited [1993] ILRM 145, at page 183 where he said:-

76. Thirdly, I intend to adopt the approach set out by Keane J. in the same judgment (at page 185) as follows:-

77. I will now attempt to summarise the economic evidence given by the expert witnesses on either side of the case.

78. This evidence was given on behalf of Chanelle by Mr Moore McDowell.

79. Mr McDowell is a statutory lecturer in economics at University College Dublin where he obtained his primary degrees. He obtained a post-graduate degree in the University of Oxford. He has been teaching in Dublin since 1969 and has held visiting lectureships in the United States. Since the introduction of the Competition Act in 1991 he has been extensively engaged in consultancy and to some extent research in the area of industrial economics and has been involved in several Competition Law cases.

80. His report recites that if Chanelle were "de-listed" it would not be feasible for it to match Pfizer's discount to its wholesalers without selling Pfizer products at a loss. The total value of sales of animal health products in Ireland is between £55-£60 million. He describes the "vertical market structure" as comprising three levels, the "up-stream" level being the producers of both proprietary and generic products. The intermediate stage comprises a group of independent wholesalers including Chanelle at least four of which have links with manufacturers. Prior to 1995 the producers of Pfizer and Smith Kline Beecham products had direct sales to end users. In December, 1995 Pfizer decided to offer all their products through wholesalers and in appointing five wholesalers included, apparently, all possible national distributors of animal health products.

81. The final (retail) level comprises vets, chemists, local co-operatives, farmers and other end users.

82. The primary producers (comprising in the main multi-national corporations) largely import their products and promote them. They supply wholesalers at a discount on a recommended price (Chanelle carries approximately 1,500 lines in stock). The wholesalers compete for business on service (speed of delivery, range of product, advice) and price (discounts). The Pfizer rebate system is based on accumulated sales and therefore has an element of loyalty bonus. The wholesalers gather information and pass it on to Pfizer for calculating the rebate. It is critical that only products purchased by end users from wholesalers will be calculated for the purpose of assessing the end users' rebate. Turning to the market for these products some 75% of sales was in the hands of the eleven largest producers. The largest two each enjoyed 15% and Pfizer at number three had 12% of sales. Mr McDowell said that it was hard to sustain an argument that Pfizer had sufficient market power to be treated as dominant, if one takes the animal health market as a single market.

83. Looked at from the point of view of the consumer (demand side) it became clear that there were a large number of markets because the demand was for the treatment of different conditions, each of which could constitute a market on its own. This approach depends crucially on the technical evidence of experts in veterinary medicine to establish accurately the market boundaries.

84. He referred to the APHA Report and noted that it divided the overall market into eighteen main product groupings with approximately eighty sub-groupings.

85. He concluded that on the demand side there were a very large number of markets the exact number and boundaries of which would have to depend, for their definition, on expert evidence. As an economist he did not have sufficient data to add to the establishment of boundaries by reference to veterinary evidence.

86. On the supply side it was important to determine the degree of ease of entry to the markets (whether they were contestable). An unbiased observer would probably not regard the animal health products market as being closed to entry, especially in light of a significantly large number of suppliers with general ability to use reverse engineering to produce competing generic products. This applies less so where a producer (Pfizer) has products supported by patents or branding which have a large share in a market.

87. Dealing with the market structure in terms of product markets and taking information from the APHA fourth quarter Report for 1996, Pfizer had 13.8% of the overall market for the year 1996 and 16.5% for the fourth quarter of that year. These are not percentages usually associated with substantial market power. The market power of Pfizer may, however, be under-estimated by treating the animal health product market as a single aggregate market.

88. In the absence of actual hard data, Mr McDowell calculated that in twelve of the thirty six APHA product sectors in which Pfizer had products (out of a total of some eighty APHA product sectors), Pfizer was one of the top three which between them had 60% or more of the market.

89. Turning to Chanelle (as manufacturer) the figures (from the same source) show that Chanelle is a fringe producer with between 2-3% of the product market. By reference to individual markets Chanelle had products in twenty-five of these with shares ranging up to 40% but the higher shares were in small value market segments.

90. Comparing Chanelle and Pfizer it appeared that sales of Chanelle products seemed to have little impact on Pfizer sales (for example in two markets where Chanelle significantly increased its market share, so did Pfizer). Again, Chanelle's Albex competes with Pfizer's Valbazen: Chanelle's share rose by 40% on a low base in 1996 whilst Pfizer's Valbazen rose by 7% on an already high base in a nearly stationary market.

91. Turning to the intermediate (wholesale/distributor) level of the market, there appeared to be five firms at this level including Chanelle. Economists would call this stage oligopolistically competitive. (The percentage shares of the five were 31%, 15% (Chanelle), 12%, 7% and 6% with 29% accounted for by direct sales and others). The value of Chanelle's 15% was approximately £7.9 million in the year ending April 1996 and of this, its own products (i.e. the products of its sister company CPML) accounted for 22% of turnover. Chanelle increased its market share from 9.5% to 14.1% in the four years to 1996. This increase was accounted for by an increase in the classic wholesale business rather than from selling its own products.

92. Mr McDowell next offered an economic analysis of the foregoing.

93. There was no information on any agreements between the major (mainly multi-national) firms. They could be taken to compete at the level of research and development but given large "sunk costs" it would not be in their interests to compete on price.

94. A concentrated wholesale stage is more consistent with the objectives of avoiding price competition in contrast to an atomistic structure which would make monitoring of pricing by suppliers more difficult.

95. Dealing with the effect of the rebate scheme, Mr McDowell noted that information on sales was passed to Pfizer from its wholesalers. Whilst the retailers were not obliged to enter this arrangement, they were obliged to do so if they were to get the benefit of the rebate offered by Pfizer, based on achieved sales to them by approved wholesalers. He said:-

96. He referred to evidence which he understood to amount to complaints by Pfizer of deep discounting by Chanelle which could be seen as a restriction on intra brand competition at wholesale level thereby, he said, reducing inter brand competition on price. The fact that the price at the end user stage is sensitive to scale of purchase in the context of the rebate system has the effect, he said, of reducing the incentive to engage in intra brand price competition by wholesalers. There would, moreover, appear to be no sound commercial reason for Pfizer to limit the number of wholesalers. These features suggest that there may be "strategic considerations" for Pfizer's distribution arrangements. Two such considerations would be:-

97. If any tacit agreement on price competition exists, it would be difficult to police given the discounting by wholesalers, and the loyalty rebates and bonus offers of goods from Pfizer. There is evidence suggesting pressure on wholesalers not to engage in discounting, or at least deep discounting. The entire pricing structure operated by Pfizer begs questions. If the object is to increase sales, it is difficult to see why Pfizer does not offer deep discounts to wholesalers based on volume sales rather than to the retail trade.

98. A second possible "strategic consideration" would be to discourage generic competition. The refusal of Pfizer to permit Chanelle to market its products in the face of a competitive threat from Chanelle's generic product challenge to a Pfizer brand leader can be seen not only as weakening Chanelle as a competitor, but also as sending a clear signal to other distributors that introduction of generic competitors for Pfizer's brand leaders is likely to be a costly exercise. Thus, a barrier to entry to the market for supply of generics is created.

99. Mr. McDowell next deals with the implications of Pfizer's refusal to supply Chanelle as a wholesaler. Chanelle argues that it needs to offer the entire range of animal health products including Pfizer's products because vets have a preference for "one stop shopping". Pfizer has approximately 15% of all total sales of animal health products in Ireland. It has products in markets which between them account for over £26 million from a total estimate sales value of £55 million. This is over 47% of the total. De-listing Chanelle implies depriving it of the ability to offer the products of one of the top three producers in markets which cover nearly half the total animal health product sales in Ireland. This would clearly have the effect of seriously undermining its competitiveness as an independent wholesaler.

100. The impact on Chanelle's competitive position can be quantified to some extent in that Pfizer products appear in orders which constitute 24% of Chanelle's sales.

101. For Chanelle as a manufacturer the impact is also serious because it is vital for a small scale generic producer to have access to the market. Thus, if Chanelle were forced to close down its wholesaling arm by Pfizer's action, this would be a threat to the commercial viability of Chanelle's manufacturing sister company, which would be deprived of ready and friendly access to the market.

102. Mr. McDowell then turned to the implications (of Pfizer's de-listing Chanelle) for competition in general.

103. Dealing with competition at the distributor level, Mr. McDowell could not say whether de-listing would cause it to withdraw from this market. However, it was generally accepted that as the number of players in a market fall, "the consequence of a rising recognition of inter-dependence is a blunting of competition" whatever form that competition takes. Furthermore, this effect increases dramatically when the number of "players" falls below five. Collusion would be made much more likely by any factor which restricts entry into the industry. A reduction in the number of competitors, especially if it means eliminating an aggressive discounter, would be consistent with the objective of avoiding price competition with other suppliers.

104. Dealing with the impact on the manufacturing market, Mr. McDowell said an independent wholesaler was likely to offer less favourable terms to a small producer than it would to a relatively risk-free producer of substantial brand image or patented products. The de-listing of Chanelle is immediately anti-competitive, in that such an action raises entry costs into product markets by existing manufacturers insofar as they have distribution operations. Such a distributor clearly faces an entry cost into the relevant market.

105. Mr. McDowell would not accept Pfizer's argument that Chanelle's Albex posed a successful challenge to its Valbazen product. Information in the APHA Report was to the contrary. He did not agree that the sales information in Chanelle's computer was private Pfizer information, because it could have been obtained from market research and indeed from sales staff employed by Pfizer. In relation to Pfizer's perceived objection to Chanelle as the manufacturer of competing products, he noted that other wholesalers continued without objection from Pfizer to have links with manufacturers of similar competing products. In addition he offered his view that a wholesaler does not necessarily have an incentive to substitute its own product for that of a "bought in" product.

106. In cross-examination Mr. McDowell accepted that his report and evidence was prepared on the basis that the end user rebate would not be available to Chanelle once it was de-listed. If the rebate was available to Chanelle, this would have a mitigating effect on his conclusions. He also accepted that it would be significant if Chanelle remained viable as a wholesaler even if it did not carry Pfizer products. He accepted that if the APHA Report only accounted for 85% of the overall market, it would be necessary to go outside the Report to assess the true market share and market power of Pfizer, especially if there were significant competitors not accounted for in the APHA Report. The calculation of Pfizer's market share was a "guesstimate". They were "best estimates with very imperfect data. A guesstimate, an inference I would draw from the data. One which could be contradicted by better evidence".

107. He accepted that at the wholesale level there appeared to be a highly competitive market and clarified that his evidence that Pfizer were in some way trying to control price, was that Pfizer's intervention was "consistent with not being anxious to see price competition between (wholesalers)". With reference to his evidence that a reduction from five wholesalers to four wholesalers would increase the likelihood of collusion, he clarified that he would have to have much more evidence before he could give a considered opinion that such was highly probable to happen. His evidence was that there was an increase in probability. The heart of the matter was that Chanelle was being placed at a competitive disadvantage by virtue of the decision to de-list it, which arises from Chanelle's introduction of a competing product. The new system (a de-listed Chanelle entitled to an end user rebate) would not necessarily in all cases produce as big a profit for Chanelle as being a distributor. That went to the heart of the matter, in that Chanelle was being treated in different terms to other distributors and he thought that, under the law, in terms of price, you must apply similar conditions amongst similar trading partners.

108. Expert economic evidence was given on behalf of Pfizer by Dr. William Bishop who is Chairman of a London firm of consultant economists specialising in the application of economics to Competition Law. He is professor of the economics of Competition Law at the College of Europe, Bruges, has an academic background in economics and considerable practical experience of market definition for competition policy purposes. He has been asked to provide advice to public bodies, notably the European Commission, inter alia, in connection with its drafting of a Notice on Market Definition issued in May 1997.

109. I will attempt now to set out the salient points of Dr. Bishop's evidence.

110. Like Mr. McDowell, he also relied on expert veterinary advice particularly in relation to market definition.

111. He said that competition (commercial rivalry) was seen by economists as favouring the efficient use of resources in the interests of improving the welfare of individuals. Economists see efficiency as concerned with the optimal allocation of production resources and with maximising output. It is also concerned with innovation and technological advance which can only be assessed over time. Economists differ to some extent as to how best economic efficiency can be achieved.

112. A text book model of "perfect competition" is rarely appropriate in practice: real competition is a dynamic process characterised by rivalry. A dilemma for competitive policy is that the outcome of this process may result in a firm or group emerging with a stronger position in the market: if the goal of achieving efficiency is to be achieved, however, care must be taken before intervening in such a case. It is unsound to interpret harm to a competitor as equivalent to harm to competition.

113. There is no simple solution to this dilemma because there will always be tension between the process of competition and its outcome. A guiding principle ought to be that intervention is only warranted where it is likely to result in an improvement over the probable outcome in the absence of intervention. The underlying aims of any policy must be borne in mind: these include the enhancement of economic efficiency in the ultimate long-term interest of the consumers. The primary concern should be to maintain an appropriate market environment over the long term and a "snap shot" at any one time could be misleading.

114. In relation to dominance, this is characterised broadly as a persistently high position in the market place which permits of substantially greater freedom of commercial action than in the presence of close competition. Concern arises when a firm exploits or seeks to reinforce a dominant position. Dominance is only a problem if potential competitors face barriers that they cannot overcome by themselves within a reasonable time.

115. A second area of concern is with "anti-competitive agreements". These should be tested from an economic point of view, against their likely impact on the objectives of economic policy which is concerned ultimately with the welfare of the consumer.

116. In the context of Section 4 or Article 85, the criterion given by law (the impact of an agreement on competition) can give rise to conflicts. This is because of the somewhat ill-defined nature of the concept of "competition". Given that an agreement will inevitably impose conditions limiting freedom of action, a narrow view of competition policy will almost inevitably see any agreement as "anti-competitive" - leading indeed to outcomes contrary to the aims of the legislation itself. The same agreements could, on a wider view, be seen as enhancing competition if they resulted in economic activity which would otherwise not occur. Hence the need, from an economist's point of view, to take a broad view.

117. A relevant market can be defined for a given product as a group of products all of which consumers would consider to be reasonable close substitutes for one another. It is necessary that all consumers be considered because different individuals will have different views on acceptable substitution: it is the aggregate demand which constrains the supplier to the market. Therefore the essence of market definition is to identify a group of products which is:

118. This concept is similar to those underlying definitions provided by the European and US authorities.

119. In the present case, the geographic area of any relevant market is the Republic of Ireland.

120. In applying these notions to the animal health products market, it is important that any serious attempt to define a market in practice must make reference to some empirical evidence of consumer demand. Dr. Bishop relies on the APHA Report and on information from Tom Mullaney and David Pepper (a veterinary surgeon). In light of this, it was immediately apparent that there was not a single market for animal health products in Ireland, but that there was a diverse collection of products in different markets.

121. The key test in market definition is the substitutability of products. The demand in this case is what economists would call a "derived" demand, namely, a demand for the result which the product produces rather than for the product itself. This is often determined by the view of the veterinary surgeon.

122. He was advised (by Mr. Mullaney and Mr. Pepper) that there were two primary features of this demand at retail level, namely, (a) a condition requiring action, and (b) the species of animal, together with a possible third, (c) the mode of administration. The delineation of product market at wholesale level is likely, in the absence of special features, to follow the delineation of market at end user level. In the markets in the present case both branded and generic products appeared: differentials in price between these did not imply that they were in distinct relevant markets.

123. In dealing with the allegation that Pfizer's distribution arrangements were "anti-competitive", Dr. Bishop perceived a close connection between market power and anti-competitive arrangements: the objectionable outcome of abuse of a dominant position under Article 86 is the same objectionable outcome achieved through "co-ordination" by more than one undertaking under Article 85. Before considering anti-competition it is necessary to be clear about the markets affected. If an undertaking or group of undertakings has little or no market power, it is unlikely that it can restrict competition to any appreciable extent by entering into a "vertical" agreement.

124. In dealing with Chanelle's claim under Section 4/Article 85, insofar as it alleges that the reduction in the number of wholesalers of Pfizer products is likely to affect competition at the wholesale distribution level, there are two determinants of this question, namely, (a) the likely intensity of intra brand competition between the remaining four Pfizer wholesalers/distributors, and (b) the extent to which non-Pfizer products provide a competitive constraint on the price of the four Pfizer wholesalers.

125. In relation to the first of these, there is no evidence to suggest collusive tendencies between Pfizer's wholesalers: on the contrary the evidence suggests keen competition on price between wholesalers: furthermore, to the extent that there are competitors for Pfizer products, this question loses much of its significance. Dr. Bishop sees no evidence suggesting Chanelle would be unable to remain competitive as a wholesaler of non-Pfizer products and there is a possibility that it would continue wholesaling Pfizer products, albeit on a smaller margin. To argue that de-listing Chanelle would appreciably impair competition in wholesale distribution would, therefore, first require demonstration that at lease some Pfizer products have no close competitors in their particular markets.

126. In relation to the effect on "up-stream" (manufacturing) competition,

127. Dr. Bishop noted that all wholesalers carry a variety of generic products, some in direct competition with Pfizer products. Accordingly, Pfizer's own evidence does not entail an implicit general requirement of its wholesalers to refrain from competing generics. It does not follow that competition per se is jeopardised even if Pfizer's dealing with Chanelle is motivated by a perception of a specific threat from Chanelle's Albex.

128. Turning to Mr. McDowell's report, Dr. Bishop observed that it fails to deal adequately with the specific features of the products or markets at issue in this case. It deals largely in general principles with which Dr. Bishop has no quarrel, but these do not much advance the Plaintiff's case.

129. In relation to market definition, Mr. McDowell offers no conclusions of his own, although Dr. Bishop agrees with his general approach. It is not justifiable, however, to equate the APHA categories with markets and Mr. McDowell's market share analysis based on the APHA categories is therefore undermined in a fundamental way.

130. Mr. McDowell offers no explicit conclusion on dominance or even on market power. Dr. Bishop notes that Mr. McDowell makes no reference to the product markets identified in the Plaintiff's solicitor letter of the 25th April, 1997. In regard to the treatment by Mr. McDowell of market shares and contestability, this treatment does not permit of conclusions about the specific markets at issue because in the case of market share the markets are not defined and the APHA Report is incomplete and in the case of contestability, Mr. McDowell's treatment is only in general terms, for example " in so far as the markets for AHB can be so characterised (i.e., by the importance of branding and/or intellectual property rights) supply side substitutability is weak, and the markets are not in general highly contestable" (p.15); and " to the extent that products are protected by patents .... and/or by substantial expenditure on marketing and branding supply side competition is weakened" (p. 16) (emphasis added to both quotes). The factual premise necessary for the conclusion to apply to the specific markets in question is not established in either case."

131. In relation to the claim that Pfizer's decision to de-list Chanelle is "anti-competitive", Dr. Bishop notes that there was no firm conclusion on market definition and that therefore, it is difficult to assess any alleged effect on competition. Specifically, in relation to the claim that Chanelle is not able to compete on an equal footing, Dr. Bishop says it is not at all clear that Chanelle would cease to wholesale Pfizer products, there being examples of other wholesalers not appointed by Pfizer who do this. It has not been demonstrated that Pfizer is immune from competition in any market and therefore the four remaining wholesalers would continue to face competition from Chanelle wholesaling other products. The empirical evidence shows that "one stop shopping" is not reflected in the actions of individuals.

132. Even if, contrary to the foregoing, the wholesalers were reduced from five to four, it does not follow that there would be a reduction in intensity of competition at wholesaler level. Mr. McDowell's evidence in this connection is the outcome of a particular theoretical model which can not be taken as a serious prediction of what will happen. There is evidence of keen price competition between wholesalers and even evidence that Chanelle's rivals seek to undercut Chanelle on Pfizer products. Dr. Bishop notes that in December, 1995 the number of wholesalers had been reduced to five. This indicates an implicit acknowledgement that the criterion for judging whether a particular distribution arrangement distorts competition is not whether there is a potential buyer who is treated differently from an other, but rather whether the likely net effect of the arrangement is to reduce the intensity of competition.

133. Dealing with the alleged anti-competitive effect at manufacturing level,

134. Dr. Bishop suggests that this claim does not conform to the facts. There are a lot of Pfizer distributors who carry a large range of generics which are in competition with Pfizer products and therefore Pfizer's objection is not an objection to generic competition per se . Even if the object or effect of the de-listing was to make circumstances more difficult for the manufacturers of products which compete with Pfizer products, before concluding that this was anti-competitive one would have to consider that Pfizer is not shown to be dominant and that it was therefore difficult for Pfizer to have an effect on the wholesale market. Furthermore, Mr. McDowell had argued that it was in Pfizer's interests to reduce the intensity of "up-stream" price competition by controlling down stream pricing, and in particular by reducing the intensity of competition on Pfizer products between wholesalers. This may be true in a general sense, but it is a long way from showing that Pfizer has actually behaved in this way, much less that any such actions have been successful. On the contrary, the evidence shows lively price competition between wholesalers. The ultimate test must be whether this allegedly anti-competitive agreement will have an overall chilling effect on the intensity of competition. The extent to which Pfizer distributors have in the past been able to carry generic products without eliciting such a reaction from Pfizer is relevant to answering this question.

135. In relation to dominance, Dr. Bishop defined this as a persistently high position in the market place which permits substantially greater freedom of commercial activity than in the presence of close competition. Concern arises when a firm exploits or seeks to reinforce a dominant position. Dominance is only a problem if potential competitors face barriers that they cannot overcome by themselves within a reasonable time.

136. Market power can be seen when a firm is in a position to maintain prices above the competitive level without losing profit. The competitive level is the level which allows it to cover the costs of production and earn a return on capital just sufficient to attract funds to run the business. This concept is closely allied to the concept of dominance.

137. It is more likely that a firm (or group of firms) with a high market share would be able to distort competition than in the case of a low market share. In Dr. Bishop's experience, it is unusual that a firm with less than 40% of market share would be able to exercise any appreciable degree of market power. Market share is only one of several possible indicators.

138. Applying these concepts to the allegation that de-listing Chanelle constitutes an abuse of a dominant position in the market for Synulox, Dr. Bishop is advised (by

139. Mr. Pepper) that there are four different Synulox sub-product markets depending on species of animal and mode of administration. According to Mr. Pepper, these are not substitutable one for another. Furthermore, each of these Synulox products has substitutes outside the Synulox range. These four markets are:-

140. Synulox 500 mg. Bolus: the market is cattle oral antibiotics: the APHA class is oral preparations I/N SCO cattle and Pfizer's market share is 26.4%.

141. Secondly Synulox lactating cow intramammary suspension: the market is lactation cow intramammaries: the APHA class is intramammary lactating cow and the Pfizer share is 13.2% to 13.8% (estimated).

142. Thirdly, Synulox palatable tablets: the market is small animal oral antibiotics. The APHA class is oral preparations I/N SCO small animal and Pfizer's market share is 34.5%.

143. Fourthly, Synulox ready-to-use injection: the market is injectable antibiotics: the APHA class is injectable preparations (whole category-7 classes) and Pfizer's market share is 13%.

144. Pfizer's market share alone in three of these is sufficient to conclude with a high degree of confidence that Pfizer cannot be dominant. In the third case (Synulox palatable tablets), Mr. Pepper does not believe that Pfizer is dominant in the market for small animal oral antibiotics. He notes that the Synulox palatable tablet share fell in each of the years 1995 and 1996: this was against a background of falling real prices and is therefore strong evidence of increasingly keen competition.

145. In cross-examination Dr. Bishop accepted that there are three separate levels of market in the present case, with a specific market at wholesale level. It is possible to enter an agreement with other undertakings to distort competition at the wholesale level but one would need to monopolise the wholesale level which is difficult to do because of the ease of entry into wholesaling generally. Any perceived "rip-off" would produce a reaction by way of the rapid setting up of alternative wholesalers. A reduction of wholesalers in number from five to four is not likely to reduce competition. In fact, even if the number were reduced to one, this is not likely to have an effect on competition because of the ease of entry into the wholesale market.

146. Dr. Bishop accepted that if Chanelle offered to withdraw Albex this might affect competition, if over time no one else could be found who would take it up. He also accepted that it is probable that over time the amount of Pfizer products that Chanelle will purchase will decrease because of the non-availability of the rebate to Chanelle's customers.

147. From the foregoing, it can be seen that both economic experts accepted that the definition of the relevant market in the present case is crucially dependant upon the evidence of the vets. It is also clear that a crucial difference between them is that

148. Mr. McDowell regards the placing of Chanelle at a competitive disadvantage as the heart of the matter, whereas Dr. Bishop makes a key distinction between harm to a competitor and harm to competition. To a significant degree this distinction also underlies the approach to the interpretation of Section 4(1)(d) of the Act of 1991 (and the equivalent sub-paragraph in Article 85), between respective Counsel in this case, and accordingly, I will now turn to consider the legal submissions in this regard.

149. Counsel for Chanelle argues that Section 4(1)(d) is to be interpreted as standing alone as an alternative to the generality of Section 4(1). He says that if the present case involves an agreement or a concerted practice which applies dissimilar conditions to equivalent transactions in the way identified in this sub-paragraph then, without more, it is prohibited and void under sub-section (1). He says the generality of sub-section (1) which refers to agreements and concerted practices having as their object or effect the prevention, restriction or distortion of competition in trade is another category which is an alternative category to agreements which apply dissimilar conditions etc. under sub-paragraph (d). The effect of this interpretation is that agreements which apply dissimilar conditions to equivalent transactions can be condemned as contrary to sub-section (1) even if it is not established that they have as their object or effect the prevention, restriction or distortion of competition. Counsel for Chanelle goes on to say that it is clear on the evidence that Chanelle has been placed at a competitive disadvantage by being de-listed by Pfizer and that this is clearly an instance of the application by Pfizer of dissimilar conditions to equivalent transactions contrary to Section 4(1)(d) and that therefore, without more, the agreement or concerted practice (alleged) is prohibited and void.

150. Counsel for Pfizer says, on the contrary, that sub-paragraph (d) must be read together with sub-section (1), that sub-paragraph (d) is an instance of the generality of agreements or concerted practices having as their object or effect the prevention, restriction or distortion of competition in trade and that to isolate sub-paragraph (d) and treat it as identifying a comprehensive and independent test which is quite separate from the generality of sub-section (1) would be to ignore the purpose of the Act and would produce results never intended by the legislature.

151. I have carefully considered these submissions because they represent not only a difference between Counsel but to a significant extent a difference between the economic experts. Mr. Moore McDowell emphasised that the heart of the matter lay in the fact that Chanelle was at a competitive disadvantage after de-listing and he specifically referred to equivalent transactions in this context. Dr. William Bishop, on the other hand, emphasised throughout his evidence that a Court must look at competition and the effect on competition of any allegedly improper agreement rather than its effect on an individual competitor.

152. In reaching a conclusion on these submissions I have had regard to the following factors:-

153. If the interpretation advanced by Counsel for Chanelle is correct this would, to my mind, prejudice the generality of Section 4(1), because it would focus the exclusive attention of the Court onto the fate of an individual trading party (or parties) who are placed at a competitive disadvantage, without any reference to the issue as to whether competition was prevented, restricted or distorted. To my mind, this interpretation could itself lead to anti-competitive results in the way identified in the evidence of Dr. William Bishop and I accept that the Court must look at competition rather than the competitor when testing any allegedly anti-competitive agreement or concerted practice. I cannot accept, therefore, that the evidence that Chanelle is placed at a disadvantage as a competitor establishes, without more, that its de-listing is in breach of Section 4 (1). In my view, to establish a breach of the sub-section it is necessary to show on the balance of probabilities that competition has been or will be prevented, restricted or distorted.

154. In light of my understanding of the correct interpretation of the relevant legislative provisions, I am not satisfied that the evidence establishes on the balance of probabilities, that any agreement or concerted practice has or will have the effect of preventing, restricting or distorting competition contrary to Section 4 of the Act of 1991 or (assuming an inter-State impact) Article 85 of the Treaty of Rome.

155. Chanelle argues that Pfizer's distribution system is what has come to be termed a "Selective Distribution System" in competition law. These systems are treated as a special category and because they are to some extent regarded as sui generis I think I should set out, briefly, my views on this submission.

156. Clearly the distribution system operated by Pfizer is "selective" in the sense that Pfizer has selected (now four) appointed wholesaler distributors to service the entire country. At the end of 1995, the present five member distribution system was put in place to replace an earlier distribution system involving many more wholesaler distributors and also some "direct" sales to end users. The system, therefore, is clearly selective or restricted. The concept of a "selective distribution system" as a term of art in competition law, however, involves not only the selection of a limited number of wholesale distributors but also the imposition by the supplier upon such distributors of a requirement that they would re-sell only to end users or other appointed wholesale distributors. The concept is defined in Bellamy and Child (op. cit. paragraph 7-073) as:-

157. It is this latter element, denying access to the suppliers' goods to non-appointed wholesaler distributors, which characterises selective distribution systems in competition law and these are justified only when they conform to specific criteria dealing with such issues as whether the goods or products are appropriate for selective distribution, whether the distributors are selected on a "qualitative" (as distinct from quantitative) basis, and whether the system is operated in a non-discriminatory way, with the result that any distributor wishing to be supplied and who qualifies must be appointed.

158. It is clear that Chanelle will be entitled to purchase Pfizer's goods from any of the other four appointed wholesale distributors. Furthermore, they will be entitled to the rebate and are likely to qualify for the rebate at the highest level. Nonetheless, Chanelle argues that the distribution system in the present case is a selective distribution system because the effect of the rebate system will be to preclude Chanelle from having access to the Pfizer range of goods. Dr. William Bishop accepted that there will be a diminution in the amounts of goods that Chanelle would purchase after de-listing, possibly to zero in the context of the non-availability to Chanelle's retail clients of the Pfizer rebate. Furthermore, it is stated, that restrictions are imposed by statutory regulations on the sale of animal remedies. With regard to the latter point, I cannot see that a regime of general statutory intervention in the market place by way of licensing persons and categories of persons who may or may not deal in particular animal health products, adds anything more by way of end user curtailment in the context of a restricted distribution system than it does in the case of an open distribution system. I accept the Defendants' submission that the only restriction on wholesalers in relation to onward sales which can be taken into account for the purpose of ascertaining whether the distribution system is a selective one or not, must be a restriction imposed by the supplier in relation to its specific goods.

159. I further accept the submission that the essence of a selective distribution system is the imposition of a restriction on the members of the distribution system in relation to onward sales. There is thus, a restriction on the numbers of potential competitors at the wholesale level which can be seen as anti-competitive in the absence of justification. But there is no restriction on the number of competitors at the wholesale level as a result of the distribution system operated by Pfizer. This is demonstrated by the fact that wholesale distributors who are not appointed continue to purchase from the appointed wholesale distributors and also by the evidence that Chanelle would be free so to do after de-listing. To say that Chanelle might opt not to purchase Pfizer goods, or that they would opt to reduce the quantity of Pfizer goods over time, is very different in my opinion, from asserting that Pfizer prohibits its nominated wholesale distributors from supplying Pfizer goods to Chanelle. Only in the latter case would the distribution system be a selective distribution system properly so called.

160. I note, in addition, that Chanelle has not been able to refer to a case analogous to the present one where either of the Courts or the Commission of the E.E.C. has been prepared to identify a selective distribution system properly so called where the appointed distributors are free to supply the relevant goods to any wholesaler who wishes to have them. Perhaps the closest that Chanelle comes to producing such authority is the case of Krups [1980] 3 CMLR 274. This was a Commission decision on an application by Krups for "negative clearance" in relation to a distribution system. It is noteworthy that the system received negative clearance, which is tantamount to a declaration by the Commission that there were no grounds for action in relation to it under Article 85(1) of the Treaty. In Krups the Commission noted that as a general rule Krups admits all dealers who can perform the functions specified in the dealership agreement and that the agreement imposes no restrictions on the sale of the relevant goods by Krups or its approved dealers. It also noted that the services, discounts and advantages available to appointed dealers were available to outside dealers. The system was not a closed system, in the sense of allowing access to the relevant goods exclusively to member dealers. Furthermore, the standard dealership agreement contained no obligation constituting an appreciable restriction of competition.

161. Chanelle seeks, notwithstanding the foregoing, to argue in effect that Krups would not have received negative clearance if the situation had been that the excluded distributors were not treated in the same way as those included. Chanelle has been placed at a competitive disadvantage and it is argued on the authority of Krups that once this can be established the system should be condemned.

162. I think this is reading far too much into Krups and I do not read this case as authority for the proposition that the supplier is obliged to supply all applicant wholesale distributors in exactly the same way.

163. In my judgment of the 5th June, 1997, I concluded with some doubt that a prima facie case had been made out to establish a separate market for Synulox in this country. In deciding that such a case had been made out I emphasised that I was particularly conscious of the distinction between a prima facie case on the one hand and on the other a case that will probably succeed. Counsel for the Defendants referred to this ruling as leaving Chanelle's case on this aspect as hanging by a thread. I would not quarrel with this description.

164. Since that ruling the Court has heard evidence, inter alia, from Mr. David Pepper a veterinary surgeon with twenty years experience in a mixed practice, and an ex-president of the Council of the British Veterinary Association, who has since 1991 been a consultant to the Veterinary Defence Society, which is a professional indemnity insurance company which insures many Irish as well as U.K. vets. In the latter capacity during the past six years, Mr. Pepper has made connections on a weekly if not a daily basis with many vets who are in practice all over Ireland. He has also advised a number of the bigger animal health product multinationals. He says that Synulox is an extremely effective broad spectrum antibiotic produced in a variety of formulations with a unique constitution, a unique mode of action but, crucially, its efficacy and what it does are not unique. It is a valuable part of the veterinary surgeon's armoury, but is not unique in terms of what it does. As a clinícian his view is that there was not any situation in which there would not be some other drug that could be used. This witness saw the different Synulox formulations as fitting into four separate markets, namely,

165. This evidence together with the analysis of Dr. Bishop on this point to which I will return in a moment, must be balanced against the evidence adduced by Chanelle already summarised in my earlier judgment. This was to the following effect:

166. Michael Burke said that Synulox is a product that a vet must have with him in order to carry out his work and he referred to its uniqueness and to its broad range of activity. He described it as the "Rolls Royce of antibiotics". In certain cases it is the only antibiotic that will work for a particular type of disease. A number of vets referred to Synulox as a drug of choice or equivalent. One of these, Mr. O'Keeffe, was not, however, prepared to say that there were cases which may not be treatable without Synulox. Another, Mr. Brendan Gardiner, stressed on a number of occasions that there was no substitute for Synulox. He said that there was no other product on the market that contained both Amoxycicillin and Clavulinic Acid. He referred to the results of State laboratory testing which on occasion would indicate that the only drug to treat a particular difficult situation was, indeed, Synulox. Another witness, veterinary surgeon Pat Murphy, gave evidence that there were a number of occasions where Synulox has proved to be the only antibiotic that is effective. It is particularly effective with calves. There are cases where the animal will die if it is not given Synulox and he thought that a vet would be negligent if he did not carry it when faced with such case. It was particularly effective in the case of gangrenous Mastitis.

167. Mr. Pepper in his evidence commented on the limitations of laboratory testing which in the first place did not necessarily reflect conditions in the field and secondly, there were only a limited number of antibiotics that one could physically accommodate on one test plate, leaving open the possibility that an indication that Synulox was the only antibiotic suitable for dealing with a particular bug did not communicate the entire picture.

168. This evidence must be seen, in my view, against the economic evidence of

169. Dr. Bishop who pointed out it does not necessarily follow, even if one accepted that Synulox could be the only antibiotic to save the life of an animal in a certain situation, that there must be a separate market for Synulox to deal with that situation. Whether a separate product market exists depends on whether or not it would be profitable for the supplier of Synulox to exploit that market. If the only way Pfizer could profitably do so would be to charge a higher price, then the supplier would sacrifice sales to other competitive uses and if that would lead to overall less profit than by competing generally, then there is no market for the specific situation. It follows that even if there were instances where the life of an animal might be lost in the absence of Synulox, this did not mean that this alone proved that there was a separate market for Synulox to deal with that circumstance.

170. Bearing in mind, as was pointed out by Keane J. in Masterfoods Limited -v- HB Ice Cream Limited that the onus rests on Chanelle to establish on the balance of probabilities that a market exists, I have come to the conclusion that the evidence does not establish that a separate market exists for Synulox either in general or in any one of the sub-categories. I think some vets may certainly require Synulox in their armoury as a "must" but I think that the balance of evidence shows that there are substitutes for Synulox in each of its applications. Even if I am wrong about this and if there are circumstances where Synulox and Synulox alone will save the life of an animal, I do not think it follows that a market exists for Synulox to deal with this particular circumstance because Chanelle has not shown that its producer would probably be prepared to raise the price of Synulox to capture this specialist market, thereby incurring a loss of sales of Synulox in other markets at the cheaper price.

171. The question of dominance (and of course abuse of dominance) does not, therefore, arise: however, I would say that the evidence of Dr. Bishop does establish to my mind that the Synulox product market shares he describes are not such, as are likely to establish dominance.

172. As part of its defence Pfizer alleged that information on sales of Valbazen was misused by Chanelle for the purpose of "targeting" Valbazen buyers or "piggy backing" Chanelle's new Albex product on the back of Valbazen sales. In the course of the hearing I asked Counsel for Pfizer what was the relevance of this allegation in the context of a competition case? Either the alleged Pfizer agreement/concerted practice was in breach of Section 4/Article 85 or it was not. A similar comment could be made in relation to the abuse of a dominant position point. Counsel for Pfizer agreed that this point had no direct bearing on the competition issue. In these circumstances it is not necessary that I rehearse the evidence in detail in relation to this point. I think I should record, however, that I have carefully considered the evidence adduced by both parties in relation to this issue. In particular I would accept Michael Burke as a reliable witness. He gave his answers readily, without delay and, in my opinion, in a forthright and candid manner. He accepted responsibility where that was appropriate. On this particular issue where he insisted that he did not use allegedly private Pfizer information for the purpose of launching Albex, he was supported by Mr. Tom Tierney who stated that he and he alone had the expertise to extract the relevant information from Chanelle's computer system. He also stated that a check which he conducted in the course of the trial indicated that the information had not been extracted up to that point. Furthermore, Pfizer's suspicions that Chanelle was misusing information in relation to Valbazen was stimulated by a memorandum of information prepared by Laura Ruane at Michael Burke's request and dated the 21st March, 1996. This memorandum contained details of numbers of sales of Valbazen in its different formulations across three years from May, 1994 to March, 1996. Laura Ruane was marketing manager in Chanelle and reported directly to Michael Burke. She prepared the Valbazen memo at Michael Burke's request. Her evidence was that it contained more information than he had requested and when she produced the memo to Michael Burke she felt embarrassed because she had the distinct impression from his reaction that her efforts in producing the unasked for information had been a waste of time particularly, because Michael Burke had written down the pack sizes he wanted anyway, which as a matter of fact did not correspond to the Valbazen pack sizes.

173. I specifically refer to this matter, not because it makes any difference to the result which is brought about by what I consider to be coercive economic analysis and the relevant principles of law and interpretation of the legislation, but because I do not wish to leave room for any inference that I did not accept the evidence of Michael Burke, Tom Tierney and Laura Ruane on this topic. I did accept such evidence but as I have indicated I do not think it makes any difference to the outcome.

174. In the result the Plaintiff's claim must be dismissed.