Natal Wholesale Chemists (Pty) Ltd v/s A

The judgment

The court's own file

COMPETITION TRIBUNAL REPUBLIC OF SOUTH AFRICA Case No. 98/IR/Dec00 In the matter between: Natal Wholesale Chemists (Pty) Ltd Applicant and Astra Pharmaceuticals (Pty) Ltd First Respondent Merck Pharmaceutical Manufacturing (Pty) Ltd Second Respondent Pharmaceutical Healthcare Distributors (Pty) Ltd Third Respondent _______________________________________________________________________ DECISION ON APPLICATION FOR INTERIM RELIEF _______________________________________________________________________ DECISION This application for interim relief is denied. The reasons for our decisions follow. INTRODUCTION 1. This application for interim relief is brought by Natal Wholesale Chemists (Pty) Ltd (NWC) (‘the claimant’) against Astra Pharmaceuticals (Pty) Ltd (AZ), Merck Pharmaceutical Manufacturers (Pty) Ltd (Merck) and Pharmaceutical Healthcare Distributors (Pty) Ltd (PHD) (respectively referred to as ‘the first respondent’, ‘the ‘second respondent’ and ‘the third respondent’ and collectively as ‘the respondents’). 2. The claimant alleges that the first and second respondents have each entered into an agreement with the third respondent in terms of which the latter is designated as the exclusive provider of distribution services for the products of the two manufacturers, the first and second respondents. The claimant alleges that the effect of these agreements is to prevent it from participating in the distribution of the products of the first and second respondents thereby lessening competition in the market for the distribution of these products. They allege that these agreements violate Section 5(1) of the Competition Act which proscribe vertical 2 agreements that have the effect of lessening competition and that do not generate countervailing pro-competitive gains. The claimant also alleges that the first and second respondents discriminate in respect of prices and terms and conditions of sale as between the third respondent and other wholesalers, including the claimant and, as such, are in violation of Section 9 of the Act that proscribes discrimination by dominant firms. 3. The claimant has asked the Tribunal to find that the distribution of pharmaceutical products through an exclusive distribution system constitutes a prohibited practice in terms of the Act, alternatively, that the first and second respondents discriminate in respect of prices, terms and conditions of sale as between the third respondent and other wholesalers, including the claimant, constitutes a prohibited practice in terms of the Act, and that the Tribunal: · interdicts and restrains the first and second respondents from distributing their products exclusively through the third respondent; · interdicts and restrains the respondents from inducing and/or allowing any other manufacturers or importers to use or participate in the exclusive distribution firm that the third respondent has with the first and second respondents; · interdicts and restrains the respondents from forming any new agency distribution firm to distribute the products of the first and second respondent on an exclusive and/or discriminatory basis · orders the first and second respondents to continue supplying their products to the applicant on the most favourable terms and conditions available to any wholesaler and/or distributor, including the third respondent · orders the respondents to delete any reference in the agreements between the first and second respondents and the third respondent which enshrines the exclusivity of their agreements. BACKGROUND 4. Pharmaceutical products, including the ‘ethical’ or patented products manufactured by, inter alia, the two respondents in this matter, have traditionally been distributed to the retail trade, the pharmacies, through the medium of wholesalers, including t he claimant - the wholesalers purchase product from the manufacturers and on-sell this to the retailers. The wholesalers cover their costs and earn their profits in the difference between the price at which they purchase the product from the manufacturers and the price at which they on-sell to the retailers. This price differential has taken the form of a discount granted by the manufacturers off their list price. The wholesalers have traditionally received a discount of 17,5% off the list price, a significant part of which has been passed onto the retailers as the wholesalers vie for market share. 5. The most prominent of these wholesalers, including the claimant in this matter, are ‘full-line wholesalers’, that is, they stock the full-range of pharmaceutical products. Certain of the larger purchasers of pharmaceutical products – notably 3 the state but also other bulk purchasers – have traditionally purchased directly from the manufacturers. 6. In recent years pharmaceutical manufacturers have attempted to change the way in which they distribute their products. A raft of pharmaceutical manufacturers – predominantly, though not exclusively, the large multinational ‘majors’ – have designated exclusive distributors of their products. Three such mechanisms of distribution have been established. In two of these arrangements – hitherto referred to as the IHD and Kinesis arrangements – two distribution companies each jointly owned and controlled by separate groupings of manufacturers have been designated as the exclusive distributors of the products of their shareholder/manufacturers. In the instant case the first and second respondents have designated an independently controlled company, namely the third respondent, as their distributor, or, in the terminology employed by the respondents, as their exclusive provider of logistical services. 1 7. As will be elaborated below, the independent wholesalers allege that, in addition to placing them under severe commercial pressure, these various exclusive distribution arrangements have, collectively and separately, diminished intra- brand competition in the market or markets for pharmaceutical products, that they have not promoted and may have diminished already low levels of inter-brand competition in these markets, that they have raised barriers to new entry in the market(s) for pharmaceutical products, and that they provide the institutional basis for collusion between the various manufacturer groupings who are either the joint owners of their distribution companies (as in the case of the IHD and Kinesis groupings) or, in the instant case, who have entered into distribution contracts with a single agent. 8. The various distribution agencies have been scrutinised by the competition authorities. In 1999 the Competition Board, the predecessor of the Competition Commission, found that a joint exclusive distribution agency for pharmaceutical 1 Note that the distribution mechanism under examination here represents a shift from the business model traditionally utilised by pharmaceutical manufacturers for the distribution of their product. In the traditional wholesaler model the distributors are wholesalers who are downstream purchasers of pharmaceutical products and who then on-sell this product further downstream. In the arrangement with which we are presently concerned the model is of an up-stream supplier of distribution services contracting with clients who require the distribution of their products. Whether or not distribution takes place through a sole downstream wholesaler or through a sole upstream supplier may not influence the assessment of whether or not a restrictive practice operates in what are both vertical agreements. But, were a restrictive practice to be found, the differences between the two business models would make the identification of an appropriate remedy considerably more difficult. Take the prayer that would have the Tribunal ordering ‘the first and second respondents to continue supplying their products to the applicant on the most favourable terms and conditions available to any wholesaler and/or distributor, including the third respondent’. This presupposes a calculation that equates the extent of remuneration represented by the discount extended to the downstream wholesaler with the remuneration represented by the fee rendered to the upstream supplier of services. This would be an extremely difficult calculation to make and suggests that, notwithstanding its protestations to the contrary, the claimant is effectively requesting us to impose the wholesaler model of distribution on the respondents. 4 products – in that instance, the IHD arrangement - constituted a vertical restrictive practice. 2 In August 2000 the Competition Tribunal granted interim relief to nine wholesalers against six pharmaceutical manufacturers and their joint exclusive distribution agency, Druggists Distributors, (the Kinesis arrangement) for contravening section 4(1)(a) of the Act. 3 Section 4(1)(a) proscribes horizontal agreements – agreements between competitors – that lessen competition without generating countervailing pro-competitive gains. 4 9. Natal Wholesale Chemists have now brought an application for interim relief against PHD and the manufacturers who utilise this distribution company, allegedly the third of the exclusive arrangements active in the distribution of ethical pharmaceutical products. The Parties The Claimant Natal Wholesale Chemists (Pty) Ltd 10. The claimant is a full-line wholesaler trading as Alpha Pharm-Durban. Alpha Pharm is a joint venture formed by the four co-operative wholesalers in South Africa. It has nine distribution centres that distribute products from manufacturers to doctors, hospitals and other health care suppliers. The Respondents 11. Both the first and second respondents, AstraZeneca Pharmaceuticals (Pty) Ltd and Merck (Pty) Ltd, are subsidiaries of multinational foreign-based pharmaceutical manufacturers. The third respondent, Pharmaceutical Health Distributors (Pty) Ltd is a logistics company. AstraZeneca (AZ) 12. As of the 25 November 2000 AZ, the first respondent, appointed PHD, the third respondent, as its distribution agent. PHD is a third party logistics provider operating on a fee-for-service basis. In terms of the agreement AZ outsourced its warehousing and distribution functions, as well as its order generation, credit control and debt management operations to PHD until 31 December 2002. AZ retains ownership of its stock until it is sold to a third party. 13. Before this arrangement came into affect AZ distributed its products to its direct purchasers, i.e. clinics, hospitals, dispensing doctors, mines, mail order retailers, the State and wholesalers through the agency of Railit Total Transportation 2 Competition Board Report No. 75 3 Case no: 68/IR/Jun00 4 In essence, the Tribunal panel in this matter found that the vertical agreement was a mechanism for affecting a horizontal agreement between the manufacturer/shareholders. This issue is dealt with below. 5 (RTT). Wholesalers then on-sold the stock they had purchased to their retail customers at prices determined by them. AZ sold its products to its direct customers at various discounts off list price, depending, inter alia, on the nature and volume of p roducts purchased by such customers. Wholesalers received the traditional uniform minimum discount of 17,5% off list price. 14. However, AZ decided to phase out its manufacturing function and to outsource its warehousing, distribution, order-generation, d ebt management and credit control functions. AZ avers that the decision to outsource these activities was taken because they did not form part of its core business, namely the sale and the marketing of its products, because its Alrode distribution facilities were outdated and required considerable investment and because there were significant economies of scale to be reaped from using a single agent/distributor. 15. According to AZ all its clients now have the choice of buying either directly from it, with PHD doing the physical distribution, or buying from the wholesalers but who now also receive the product that they purchase via the third respondent’s network of distribution services. Merck 16. The second respondent, Merck, has been dealing with P HD, its exclusive distribution agent, since 27 March 2000. Although the service fees paid to PHD differ it has exactly the same logistics services arrangement with PHD as the one between AZ and PHD. Merck decided to appoint PHD as its agent as part of its strategy to outsource non-core business activities to enable it to concentrate on the manufacture, marketing and sales of pharmaceutical products. The arrangement expires in March 2002. 17. According to Merck it has had an established working relationship with RTT who provided a transport and delivery service to it prior to the arrangement with PHD. PHD has outsourced the transport of Merck’s products to RTT. 18. According to Merck 60% of its sales continue to be made to the traditional wholesalers including the applicant, who, as in the arrangement between the first and third respondents, are now also required to utilise the distribution and other services of the third respondent. PHD 19. The third respondent has been granted the exclusive rights t o act as the logistics service provider to the first and second respondents. It may expand its services to other pharmaceutical companies – indeed, avowedly because of the positive impact of scale economies on the cost of the distribution services, all three respondents commit themselves in their various agreements to encourage others to 6 use the services of the third respondent. At present PHD also distributes products of Sekunjalo (Pty) Ltd, a manufacturer of generic pharmaceutical products. 20. PHD provides the services of warehousing, distribution, debt collecting, batch tracking, order processing, picking, packing, credit control and debt management to its principals, the first and second respondents. These services are provided through PHD’s association with the following companies: · Kite Logistics (Pty) Ltd (Kite), which performs the physical transport of the pharmaceutical products to pharmacies and doctors. · Order Pharm (Pty) Ltd (Order Pharm) processes the orders received from customers such as wholesalers and pharmacies. · Railit Total Transportation (Pty) Ltd (RTT) performs the physical distribution and transport of pharmaceutical products to the Government and the wholesalers. · Recall (Pty) Ltd performs the debt management sector of the service provided by PHD. 21. PHD and Recall are wholly owned subsidiaries of Fuel Logistics Holding Company Limited (“Fuel Logistics”). Fuel Logistics and International Health Distributors (IHD) each own 50% of the shares of Kite. IHD, referred to above, is a joint e xclusive distribution agency controlled by several other pharmaceutical manufacturers. INTERIM RELIEF 22. The applicant applied for interim relief under section 59 of the Competition Act of 1998 on 1 December 2000. The Act has since been amended by the Competition Second Amendment Act, No. 39 of 2000 with effect from 1 February 2001. Section 59 was replaced by Section 49C. 23. The Tribunal has been asked to consider whether the amended Act should apply etrospectively in this application. 24. The parties have dealt with this issue in great detail in their heads of argument. Both parties refer to Section 23(5) of the Competition Second Amendment Act which provides that: Any proceedings that were pending before the Competition Commission, Competition Tribunal or Competition Appeal Court before the date of commencement of this Act must be proceeded with in terms of the principal Act as amended, except to the extent that a regulation under section 21(4) or 27(2) of the principal Act as amended, or a rule of the Competition Appeal Court, provides otherwise. 7 25. The claimant argues that by virtue of Section 23(5), the Competition Second Amendment Act retrospectively applies to pending legal proceedings, including the applicant’s section 59 application. This means that the Section 59 application filed prior to the commencement date of the Competition Second Amendment Act must now be proceeded with in terms of the new section 49C, which replaces it. 26. The respondents on the other hand argue that section 23(5) only applies to procedural amendments and not to amendments affecting parties’ substantive rights and obligations. They submit that the changes to section 59 as reflected in section 49C of the Competition Second Amendment Act are matters of substantive law, hence, the substantive legislation applicable to these proceedings is that set out in section 59 of the original Act. 27. Section 59(1) of the Act provided that the Tribunal may grant interim relief if: (a) there is evidence that a prohibited practice has occurred; (b) an interim order is necessary to i. prevent serious, irreparable damage to that person; or ii. to prevent the purposes of this Act being frustrated; (c) the respondent has been given a reasonable opportunity to be heard, having regard to the urgency of the proceedings; and (d) the balance of convenience favours a granting of the order. 28. To obtain interim relief a claimant had to satisfy each of the elements from (a) to (d). According to Section 68 of the Act the standard of proof that had to be met was “on a balance of probabilities”. 29. Section 49C(2)(b) provides that the Competition Tribunal may grant an interim order if it is reasonable and just to do so, having regard to the following factors: (i) The evidence relating to the alleged prohibited practice; (ii) The need to prevent serious or irreparable damage to the applicant; and (iii) The balance of convenience. 30. The amendments bring about three important changes to interim relief proceedings under the Act. 31. Firstly, Section 49C(2)(c) provides that the standard of proof in interim relief proceedings under the Act is the same as in a High Court common law application for an interim interdict. The standard of proof for an interim interdict at common law was laid down in the case of Webster v Mitchell 5 where it was held that: “the right to be set up by an applicant for a temporary interdict need not be shown by a balance of probabilities. If it is ‘prima facie established though open to some doubt’ that is enough …” 5 1948 (1) SA 1186 (W) 8 32. An applicant under Section 49C(2)(b) therefore has only to establish his case on a prima facie basis; this is a departure from the approach in the old Section 59(1) where as we have seen a claimant had to prove its case on a balance of probabilities. 33. Secondly, under Section 59 a claimant had to show that the interim relief order was necessary to prevent serious irreparable harm to itself or to prevent the purposes of the Act being frustrated. The amendments have done away with the alternative requirement (the necessity to prevent the purposes of the Act being frustrated); Section 49C(2) requires evidence that the order is necessary to prevent serious or irreparable harm. 34. Thirdly, in terms of Section 49C(2), the Tribunal no longer has to consider whether each of the requirements has been established in isolation, but rather looks at all the factors listed in Section 49(2)C as a whole to see whether a case for interim relief has been established. This feature of Section 49C(2) distinguishes it from the old Section 59 where interim relief could only be granted where each of the listed requirements had been satisfied. Section 49C(2) follows the approach at common law as applied by Appellate Division in the case of Eriksen Motors (Welkom) Ltd v Protea Motors, Warrenton 1973 (3) 685 (A) . The court held that in deciding whether to exercise its discretion to grant interim relief the court should not look at the prerequisites 6 in isolation but should consider all of them in conjunction with each other. The court went to state that these prerequisites “… are not individually decisive, but are interrelated, for example, the stronger the applicant’s prospects for success the less the need to rely on prejudice to himself. Conversely, the more the element of “some doubt”, the greater the need for the other factors to favour him.” 7 35. It has not, however, been necessary for us to decide which section should be applied to this matter. Even on the lower burden of proof required under the amended Act, the claimant has not succeeded in proving the existence of a restrictive practice. Accordingly, we have not had to consider the other elements necessary for sustaining a claim for interim relief – the question of irreparable harm and the balance of convenience. Accordingly, the legal dispute regarding the retrospectivity or otherwise of the amended Act has no bearing on the outcome of this matter. 6 The prerequisites for interim relief at common are: a prima facie right; a well-grounded apprehension of harm if the order is not granted and the ultimate relief is granted; a balance of convenience in favour of granting the order and the absence of any other satisfactory remedy (see Setlogelo v Setlogelo 1914 AD 221 at 227). 7 At 691 E-G. 9 THE ALLEGED RESTRICTIVE PRACTICES Section 9 – Price Discrimination by Dominant Firms 36. Although not formally withdrawn, neither the papers filed by the claimant, nor its written heads of argument, nor its oral arguments persist in the claim – contained in its notice of motion – that all or any of the respondents are in violation of Section 9 of the Act. Accordingly, this claim is dismissed without further comment. 8 Section 5 – Restrictive Vertical Practices 37. Section 5(1) provides: An agreement between parties in a vertical relationship is prohibited if it has the effect of substantially preventing or lessening competition in a market, unless a party to the agreement can prove any technological, efficiency or other pro-competitive gain resulting from that agreement outweighs that effect. 38. The claimant alleges that the respondents have, by entering into agreements whereby the third respondent is vested with the exclusive right to distribute the products of the first and second respondent, lessened competition and, absent countervailing technological, efficiency or other pro-competitive gains, are accordingly in violation of the Act. 39. Anti-trust scholarship and jurisprudence conventionally adopts a sceptical attitude to claims of anti-trust harm arising from all species of vertical agreement. In particular it is widely recognised that the diminution of intra-brand competition consequent upon exclusive distribution arrangements is frequently compensated for by pro-competitive benefits that enhance the ability of the producer to compete against its competitors, that is, by the strengthening of inter-brand 8 Alleged discrimination perpetrated by the first and second respondent as between the claimant and the third respondent does of course remain at the heart of the claimant’s allegation that the respondents are engaged in vertical agreements that provide for exclusive distribution. As will be elaborated below, the claimant does not allege that it is unable to gain physical access to product manufactured by the first and second respondent – it is, indeed, common cause, that it, as well as other wholesalers, continue to purchase Merck and AZ product in significant volumes. The claimant’s case effectively rests on the allegation that the distribution fee whereby the third respondent is remunerated constitutes a greater reward for the services provided than the (reduced) discount available to the claimant and its fellow wholesalers. This differentiation or discrimination in the effective earnings for performing distribution – whether fee-based or discount-based - is, the claimants allege, the mechanism whereby the exclusive distributor is interposed and is the substance of the exclusivity between the third respondent and the manufacturers who utilise PHD as their distribution agent. It is, argues the claimant, the basis for its lack of competitiveness vis a vis the third respondent. 10 competition. 9 This general approach, which we follow, is recognised by the claimants in the present matter. 40. We stress that this does not mean that we propose following the influential scholarship that argues for treating vertical agreements as legal per se. 10 It simply serves to underline the requirement, even under the less rigorous evidentiary burden that attaches to an application for interim relief under the amended Act, to provide concrete evidence in support o f a claim that purports to identify anti- competitive consequences flowing from a vertical agreement. As will be elaborated below, we have concluded that the applicants in this matter have failed to complement hypotheses and speculation with the necessary supportive evidence. 41. The claimant identifies anti-competitive consequences of the exclusive distributorship under four headings: the impact on intra-brand competition 11 , on inter-brand competition 12 , on entry barriers and on the platform provided for horizontal collusion. However, before examining each of these we must briefly address four arguments traversed in this application. These concern, firstly, the significance of the continuing involvement of the wholesalers in the distribution of pharmaceutical products including those produced by the first and second respondents; secondly, the pertinence, for the purposes of adjudicating this matter, of an agency arrangement as opposed to alternative modes of vertical agreement; thirdly, the relevance, in a djudicating this matter, of the existence of the IHD and Kinesis arrangements; fourthly, the identification of the relevant market. Some preliminary issues Is distribution exclusive? 42. Both distribution agreements read as follows: AZ/Merck wishes to appoint PHD as AZ’s/Merck’s sole and exclusive agent for the physical distribution of its products including warehousing, order processing, picking and packing and debt collection in the Territory. 13 43. The agreements entered into between the first and s econd respondents and the third respondent clearly state that the latter will be the exclusive provider of distribution services to the two manufacturers. There is no gainsaying a strong 9 Antitrust Law; Phillip Areeda, Volume VIII, para 1611, page 149. Also see Continental TV Inc. v GTE Sylvania Inc. 433 US 36, 55 (1977) 10 R. Bork, The Rule of Reason and the Per Se Concept: Price Fixing and Market Division II, 75 Yale L.J. 373 (1966) 11 That is, competition between different sellers of the same brand. 12 That is, competition between different brands of substitutable products. 13 Clause 2.3 of the Heads of Agreement between the first and third respondent and between the second and third respondent. 11 exclusive element – certain activities that were previously performed by the wholesalers in the process of on-selling to their customers are now the exclusive preserve of a designated distribution company. 44. And yet, this is no ordinary exclusive distribution arrangement. The arrangement has been structured, in particular the system of bulk discounts has been structured, so as to accord the wholesalers a continuing role in the distribution chain. Indeed the respondents have made much of the continuing role played by the wholesalers, including the claimant, in the distribution of ethical pharmaceutical products including those produced by the first and second respondent. 45. On the evidence presented the value of the claimant’s purchases from the first and second respondents has declined only marginally since the advent of the exclusive distribution arrangements. The respondents readily acknowledge that the wholesalers are obliged to physically source their stock from PHD rather than from the manufacturers themselves. Nevertheless they insist that the wholesalers are free to purchase their stock from the manufacturers at prices and on terms and conditions determined by the manufacturers who will effect physical distribution as well as payment and credit arrangements through the agency of the third respondent. But, from there on, the respondents emphasise, the wholesalers may continue to on-sell product to the retailers as in the past. 46. The claimant effectively retorts: ‘This is as well as may be, but our customers in the retail trade are equally free to purchase directly from the manufacturers through the agency of the third respondent thus eliminating that which previously distinguished us, the wholesalers, as the intermediary link in the chain. We have lost our privileged access to the manufacturers and, more important, we have lost the discount that enabled us to cover our costs and earn our profits.’ In short, the claimant alleges that while in form it may continue to distribute pharmaceutical product, in substance the requirement to source the product from PHD has eliminated the competitiveness of the intermediaries in the erstwhile chain of distribution. 47. The respondents argue that far from depressing competition the entry of the third respondent effectively provides the pharmaceutical retailers with an additional source of the products of the first and second respondent – the retailers may elect to purchase directly from the manufacturers through the agency of the third respondent; or they may continue to source product through the wholesale mechanism. 48. In particular, the respondents point to the discount structure as the continuing basis for wholesaler participation in distribution. Or, expressed conversely, the discount structure continues to provide an incentive to most retail pharmacies to continue sourcing product from the wholesalers. Large purchasers – and this certainly includes the wholesalers – continue to receive a discount on their purchases. The size of the discount varies from 11-13%, down from the 17,5% 12 previously granted to the wholesalers. Purchasers of single units, on the other hand, are not, for the most part, entitled to receive any discount on their purchases while purchasers of two units are entitled to a discount that is generally lower than that available to the large bulk purchasers. The respondents point out – and the claimants concur – that the bulk of purchases of the retail pharmacies are for single units thus allowing the wholesalers to retain a margin from this trade albeit one that has been reduced from the traditional 17,5% discount previously available to the wholesale trade. The wholesalers will retain the custom of those retailers who are obliged to purchase single units by their ability to offer the latter a lower price than that available through the third respondent. Moreover by offering facilities not provided by the third respondent but nevertheless required by small retailers – notably multiple deliveries – the claimant may use non-price services to retain its competitive edge in this market niche. 49. This evidence is uncontested and it suggests that the competition authorities are being asked to regulate a commercial dispute, one that goes to determining the size of the discount (or the actual level of the price paid by the wholesalers), rather than to the impact of the new system on competition. Although it is common cause that the wholesalers’ returns are being squeezed by the entry of the new exclusive intermediary, they are not contractually eliminated from the chain of distribution – the manufacturers or the third respondent have not entered into agreements with the retailers that purport to prevent the latter from sourcing their product from the wholesalers although, clearly, those wholesalers who are unable to sustain the cut in margins (through, for example, reducing their costs), may be forced to exit the trade altogether. 14 50. In short, we do not hold that the mere fact that wholesalers are able to retain a distribution function means that exclusivity does not operate. The third respondent has, as its agreements with the manufacturers explicitly state, the exclusive right to perform certain distribution and related services for its principles, the first and second respondent. These exclusive rights effectively interpose the third respondent in the chain of distribution between the manufacturer and their various customers, including the claimant and other wholesalers. This interposition unquestionably eats into the wholesalers’ margins. But while commercial harm is, particularly in vertical agreements, a frequent accompanist of anti-trust harm, a successful prosecution under the Act requires an actual showing of anti-trust harm. 14 We should note here that there is, on the face of it, no particular reason why, when faced with a decline in the profitability of distributing ethical pharmaceutical product, the wholesalers should not enter other fields of distribution. Why, in essence, they too should not become specialist providers of distribution or logistical services to a range of manufacturers rather than specialist pharmaceutical distributors only utilising a traditional wholesaler model. We will return to this theme below. 13 Agency v. Ownership 51. The respondents make much of the fact that the relationship of the first and second respondents to the third respondent is, in contradistinction to the relationships of manufacturer and distributor in the IHD 15 and Kinesis 16 arrangements, that of principal and agent. They clearly seek refuge in an interpretation of the decision of the panel in the Pharmaceutical Wholesalers/Glaxo Wellcome interim relief application (Tribunal Case no. 68/IR/Jun00) 17 that suggested that while exclusivity did not necessarily offend against the Act, the joint ownership by competing manufacturers of their exclusive distributor implied horizontal collusion and, as such, constituted the basis for the granting of interim relief. For its part, the claimant, clearly drawing on the same reasoning, seeks to show that the agency arrangement is a mere sham disguising actual control of the third respondent by the first two respondents. 52. We are unable, on the evidence presented, to find that this arrangement is anything other than that reflected in the formal agreements – in other words, on the evidence, this is an arrangement governed by a number of agency agreements. However, we do not believe that this conclusion disposes of the restrictive practices claim, any more than we believe that ownership or control of the distributor necessarily establishes the existence of a r estrictive practice. An agency relationship between manufacturer and distributor may or may not embody a vertical restrictive practice just as an ownership arrangement may or may not embody a vertical restrictive practice. The latter is simply a more ‘complete’ mode of vertical integration. Our reading of the panel’s decision in the IHD matter is that in the finding against the joint ownership by the manufacturer/shareholders of the distributor, IHD, it was the joint, rather than the ownership, aspect of the arrangement that offended against the Act. In other words the vertical arrangement was found to be a mechanism for consolidating a horizontal arrangement. We deal with this issue more fully in our consideration of the alleged restrictive practice. 53. Nor, should we add, is the peculiarly limited exclusivity – that is, an exclusivity that nevertheless allows, even encourages, those who are excluded to nevertheless maintain a role in distribution - a direct function of agency rather than control. It appears that both the ‘exclusive-agent’ and ‘exclusive-owner’ modalities allow the wholesalers a continuing role in distribution of pharmaceutical product to the retailers. As we will demonstrate, while both modalities embody aspects of exclusivity, the i mpact of these arrangements on competition, particularly intra- 15 Competition Board Report No. 75 16 Competition Tribunal Case No. 68/IR/June00 17 This effectively concerned the ‘Kinesis arrangement’. 14 brand competition, is undeniably ameliorated by the partial nature of the exclusivity. Are the IHD and Kinesis arrangements relevant in our consideration of this matter? 54. The respondents deny the relevance of the IHD and Kinesis arrangements in the adjudication of this matter. We, however, concur with the claimants that this would constitute an inappropriately ‘blinkered’ approach. Our brief extends beyond examining the legality of a particular agreement. Or, expressed differently, assessing the impact of a particular agreement on competition, may and usually does, require broad consideration of the state of competition in the market as a whole, including the impact of the existence of a network of broadly similar distribution arrangements, the more so if there is evidence that establishes linkages between these various distribution arrangements. There is judicial support for this view. 18 This is elaborated below. The Relevant Market 55. The respondents argue that the claimant has failed to identify the market relevant to its claim. It argues that on this omission alone the application falls to be dismissed. 56. In fact the claimant has, at various stages of its pleadings, asserted t he relevance of a number of markets. In particular, the claimant asserts the relevance of both the market for the distribution of pharmaceutical products as well as the market for ‘all pharmaceutical products’. It is not clear whether this latter refers t o ‘all pharmaceutical products’ collectively or whether it refers to a range of markets in separate therapeutic categories. The fact that the claimant has also identified five markets – defined by therapeutic categories – in which the first and second respondent are ‘dominant’ suggests that it is this definition of pharmaceutical markets that is contended for. 57. We do not share the respondent’s view that a formal market definition is a necessary precursor to an enquiry into an alleged restrictive practice. We concur with the claimant that the purpose of defining a relevant market is to identify the exercise of market power defined in the Act as ‘the power of a firm to control prices, to exclude competition or to behave to an appreciable extent independently of its competitors, customers or suppliers’ and that market definition is only a tool for estimating market power, not a scientific test. 58. In FTC v Indiana Federation of Dentists, 476 US 447, 1986, the Court states : “Since the purpose of the inquiries into market definition and market power is to determine whether an arrangement has the potential for genuine adverse effects on competition, proof of actual detrimental effects, such as a reduction of output, 18 See, for example, the decision of the European Court of Justice in Delimitis v Henninger Brau (1991) E.C.R. 1-935, par 19 – 26 15 can obviate the need for an inquiry into market power, which is but a surrogate for detrimental effects… We conclude that the finding of actual, sustained adverse effects on competition in those areas where IFD dentists predominated, viewed in light of the reality that markets for dental services tend to be relatively localised, is legally sufficient to support a finding that the challenged restraint was unreasonable even in the absence of elaborate market analysis.” 59. Antitrust scholars Thomas Krattenmaker and Steven Salop suggest in their a rticle Anticompetitive exclusion: Raising Rivals’ Cost to achieve Power over Price 19 that a two-step analysis to estimate the probability of anticompetitive effects should be followed: “First one should ask whether the conduct of the challenged firm unavoidably and significantly increases the cost of its competitors. If so, one then should ask whether raising rivals’ costs enables the excluding firm to exercise monopoly power – that is to raise the price above the competitive level.” If the exercise of market power, as defined, is identified – if, for example, the firm is able to raise appreciably the price of its product without occasioning a significant reduction in demand – then a market relevant for the purposes of the enquiry will have been identified. 60. When examining the exclusive vertical agreements, rather than attempting to define the relevant market in the abstract, we will ask ourselves whether the exclusionary right will give one or both parties to the arrangement the power to raise prices i n the market. Competition will be harmed only if, as a result, prices can be raised above the competitive level. 20 Has there been a substantial lessening or prevention of competition? Intra-brand Competition 61. The claimant asserts that intra-brand competition has been eliminated by the exclusive distribution arrangement. This is, indeed, usually true per definition – where previously the same brand was available from a number of sellers, exclusivity in distribution implies that there will now be only a single source for the branded product. Standard anti-trust treatment of the elimination, through exclusive distribution arrangements, of intra-brand competition is to balance this diminution of intra-brand competition against the pro-competitive impact of the same arrangement on inter-brand competition. 62. However, in the instant case it is not clear that either of the predicted effects operate – that is it is neither immediately apparent that intra-brand competition 19 The Yale Law Journal Vol. 96: page 209, 1986 20 In merger analysis the identification of the relevant market is a necessary prior step precisely because merger regulation is directed at forestalling the prospect of a market structure conducive to the future exercise of market power. A restrictive practices investigation, on the other hand, is concerned with behaviour, with identifying an exercise of market power – in this type of analysis the act of establishing an exercise of market power is equivalent to the identification of the relevant market. 16 has been comprehensively eliminated, nor that inter-brand competition has been promoted. 63. Where intra-brand competition is concerned it appears that with respect to a large proportion of the purchasers of the first and second respondent’s products the range of alternative distribution mechanisms is, at most, only partially limited. The bulk purchasers – the hospitals, mines, Direct Medicines (a mail order retailer), etc - have always purchased directly (and exclusively) from the manufacturer and this will continue, albeit now through the mechanism of the third respondent. Where the small retail pharmacies are concerned, those who purchase single units of product will have a continuing price incentive to source their product from the wholesalers – the wholesalers will continue to pass on part of their (reduced) discount to the retail pharmacies, a discount that is not available in respect of direct purchases of single units of product from the manufacturer through the third respondent. In other words continuing discrimination as between the wholesalers ( qua bulk purchasers) and the retailers ( qua single unit purchasers) in the prices (discounts) charged (extended) by the first and second respondent enable the third respondent and other wholesalers both to remain active in the chain of distribution. Certain of the retail pharmacies who purchase in volumes sufficiently large to qualify for the discount available to the wholesalers will presumably cease sourcing product from the wholesalers. 64. However, the claimant, in attempting to discharge its onus to identify a lessening of competition or anti-trust harm, avers that the reduction in the discount available to the wholesaler will inevitably manifest itself in an increased price to the retailer and, from there, to the end consumer. 65. There is however no evidence provided to support this latter assertion and although the theory may appear internally consistent we cannot make our finding on the basis of theoretical or hypothetical speculation alone. As with many overly speculative arguments, there is, of course, an equally plausible alternative hypothesis that suggests the opposite conclusion: the action of the manufacturers, though manifestly self-interested insofar as it designed to enable the manufacturers to increase their own margins through absorbing part of the wholesale margin, may in turn also compel the wholesalers to search for means of reducing their own costs in order to maintain their competitive edge thus maintaining or even reducing the prices at which they on-sell product. Needless to point out, this latter version comports with the very stuff of competition – a price squeeze occasioned by the entry of a new competitor forces other distributors to seek out new sources of efficiency that, in turn, enable them reduce their charge. On this version then, not only has an element of intra-brand competition maintained, it may even have been strengthened in consequence of the squeeze on one of the participants in intra-brand competition. We cannot conclusively confirm either version because the claimant has not discharged its onus to provide any evidence in support of its contention that prices down the distribution chain have increased in consequence of the new distribution arrangement. 17 Inter-brand Competition 66. As already noted, any diminution of intra-brand competition occasioned by exclusive distribution is frequently compensated by the boost provided by this genus of distribution arrangements to inter-brand competition. By the same token, where inter-brand competition in the markets in question is already muted, a diminution of intra-brand competition will loom larger in the concerns of the anti-trust authorities. 21 67. The claimants argue that the ‘must-have’ nature of pharmaceutical products acts as a considerable dampener on the extent of inter-brand competition in pharmaceutical markets. The demand for ethical pharmaceutical products in particular is price inelastic because the choice of brand purchased is determined not by a price sensitive final consumer but rather b y the pen of the prescribing doctor. The latter’s choice is influenced, at best, by pure therapeutic considerations, more likely by inertia and habit, and, even, it is suggested, by the vast promotional resources devoted by the pharmaceutical companies to winning the endorsement of the doctor’s all powerful pen. 68. The claimant has provided powerful scholarly and judicial support for this view. 22 However, again, there is little concrete evidence provided – neither from the markets from which this opinion emanates nor, certainly, from the South African market. The lack of evidence is particularly damaging to the claimant’s case when it is acknowledged that, opinion and appeals to common sense notwithstanding, the demand side of market for pharmaceutical products is not static and relatively recent developments may have conspired to diminish the doctor’s authority. The use of formularies and in general the weight of the powerful medical aid funds combined with incremental developments in generic substitution may all have contributed to weakening the authority of the prescribing doctor. This is not to say that these developments will move purchasing authority in the direction of the end consumer. Large retailer outlets and the medical aid industry may rather be the growing power on the demand side. This may support the claimant’s contention that a key objective of the new distribution system is the removal of the wholesaler in order to secure the access of pharmaceutical manufacturers to this new source of power on the demand side. However, one way or another, it may be reasonably hypothesised that these developments will, and possibly already do, impact on the price elasticity of demand for pharmaceutical products. Again, we are, in the absence of supportive evidence, forced to indulge in speculation. We stress that making a case for the purposes of interim relief does not require that the claimant puts up elaborate 21 Simon Bishop and Mike Walker - Economics of E.C. Competition Law: Concepts, Application and Measurement. (Sweet and Maxwell, 1999) para 4.31, page 91 22 In re Brand Name Prescription Drugs, 1999-2 Trade Cases P 72,576 Judge Posner, drawing on the ‘must have’ nature of pharmaceutical products, states “..It would not be surprising, therefore, if every manufacturer of brand name prescription drugs had some market power”. 18 econometric data. But in the absence of evidence of any sort, the respondent is entitled to prevail. 69. The claimant has argued that the factors that frequently result in a strengthening of inter-brand competition as a result of a vertical agreement are absent in this case. The standard argument holds that exclusive distribution arrangements provide incentives for the manufacturer to invest in the distribution system and for the distributor to provide a high quality dedicated service. This will serve to strengthen the market position of those products that benefit from this advanced support forcing their competitors to emulate them or risk losing market share. In this instance, the claimant argues, both respondents benefit from any competitive gains that accrue from the exclusive distribution system – neither receives a competitive boost vis a vis the other and inter-brand competition is unaffected. This argument appears to accurately represent the relative positions of the first and second respondent who naturally do both benefit from the services of the third respondent. However, bear in m ind that the claimant has only identified five therapeutic categories in which the first and second respondent both hold significant market shares. On the other hand the claimant’s argument regarding the impact of the vertical agreement on inter-brand competition says nothing about the impact of the agreement on competition between the first and second respondent, on the one hand, and, on the other hand, its competitors outside of the PHD distribution network – competition with these producers may have intensified as a result of the vertical distribution arrangement. Exclusionary Effects 70. The claimant insists that the mushrooming of exclusive distribution systems in the pharmaceutical trade be viewed against the threat posed by generic substitution and parallel importation to the dominance of ethical pharmaceutical products. More precisely, the object of the exercise, argues the claimant, is the destruction of independent pharmaceutical distribution capacity in favour of a distribution system controlled, or susceptible to control, by the pharmaceutical companies. The claimants also emphasise the importance of scale economies in distribution and point out that a new full line wholesaler has not entered the industry for many years. When parallel importation and generic substitution constitute a serious threat to the established pharmaceutical manufacturers, the would-be new entrants – the importers and the producers of generic substitutes – will, argues the claimant, find themselves excluded from the distribution networks which will be controlled by their competitors. Because of the importance of scale economies it will be extremely difficult for new distribution capacity to be established. It is here that the claimants insist that we view the distribution agency currently under the spotlight in the context of the IHD and Kinesis initiatives, that is, in the context, they argue, of a concerted effort by all the major pharmaceutical companies to tie up distribution facilities. 19 71. This hypothesis warrants closer investigation but until that happens it will remain mere conjecture. The respondents point out that, far from denying access by new entrants to their respective distribution networks, the impact of scale upon the costs of distribution make it imperative that they attract additional capacity through the various networks. Moreover, it is by no means clear that dedicated pharmaceutical distributors are exclusively capable of providing distribution and other services to the pharmaceutical trade. Certainly, in a diverse range of industries, the use of specialist distributors or logistics providers is on the rise and there is little reason why these should not successfully distribute the product of new suppliers of pharmaceutical product. Again evidence w ould take us out of the realm of speculation and enable us to make an informed decision. Until then we must conclude that the claimant has failed to convince us that the vertical agreements under examination, even when read in the context of the IHD and Kinesis arrangement, will raise barriers to entry on the part of competitors to the established participants in the pharmaceutical products market. There are no legal impediments in the agency agreements that preclude the third respondent from distributing the product of other pharmaceutical companies. Horizontal Collusion 72. Finally, the claimants argue that the vertical agreement is simply the site for consummating a horizontal relationship between two competitors, the first and second respondent. 23 73. The mere fact that competitors are utilising the same distribution agency lends the allegation an immediate degree of credibility. That the relationship is concerned with a close-to-market function like distribution further legitimises a close examination of its actual content. On the other hand, it must be acknowledged that it is one area in which the fact that we are dealing with a number of avowedly independent agency agreements as opposed to a relationship of the IHD or Kinesis variety in which competing manufacturers exercise collective control over the distributor is, on the face of it, significant – in the instant case there is, after all, no board of directors on which the competitors meet and possibly collude. 74. However, none of these conflicting indications constitutes a sufficiently strong basis upon which to rest a finding. The fact that competitors utilise the same supplier – in this case a supplier of distribution and other logistical services – cannot be condemned in the absence of further evidence. And the fact that this is an agent contracting with independently controlled principals does not, on its 23 In U.S Healthcare, Inc. v. Healthsource, Inc., 589 61 USLW 2595, 1993- 1 Trade Cases P 70,142 in which related competitors brought action against Healthsource, its founder a health maintenance organization, alleging that an exclusive dealing clause in its service agreements with physicians violated antitrust laws. In this case the plaintiff tried to characterize a vertical agreement as horizontal by saying that the challenged exclusivity clause amounted to an implicit horizontal agreement among participating doctors. The court of appeals refused to characterize the challenged arrangement in this manner but stated that “formally vertical arrangements used to disguise horizontal ones are not unknown”, however, it found that the plaintiff had supplied “no evidence of such a masquerade in this case.” 20 own, allay all suspicion. What is required is evidence of actual collusion or, at least, an indication that the utilisation of the shared facilities generates outputs that facilitate collusion between the principals. 75. The claimant has pointed to the fact that the parties have utilised the same standard terms and conditions of sale; that the sales made by the first and second respondents are recorded by the third respondent on the same tax invoice; that the respective heads of agreement between the first and second respondent and the third respondent are identical; that both agreements commit the first and second respondent to encourage other pharmaceutical manufacturers to utilise the services of the third respondent; that, in general, the respondents have laid considerable store by the information generated through the new distribution system and that shared information is a critical ingredient in the maintenance of a collusive horizontal agreement. 76. In addition the claimant has pointed to evidence indicating co-operation between this distribution network and the IHD arrangement. The standard terms and conditions utilised by PHD are identical to those employed by IHD; Kite Logistics, which performs the physical transport of the pharmaceutical products to pharmacies and doctors, is jointly controlled by PHD and IHD; a previous CEO of IHD served for a time as CEO of Fuel Logistics, the company that controls PHD. 77. We have considered this evidence at some length. Again we conclude that, in the face of the respondents’ denials and explanations, the evidence is not sufficiently strong to sustain the allegation that the distribution arrangement has been put in place to facilitate collusion between the first and second respondent. In fairness to the claimant, the sort of evidence required to sustain this allegation – even on the lower burden of proof required for interim relief under the amended Act – probably necessitates a more elaborate investigation than is possible in interim proceedings. We should also note that, by the claimant’s own data, the incentive for the first and second respondent to collude is weak – they only compete to any significant extent in five therapeutic categories and while the downside from collusion for those patients who ‘must have’ the drugs in these therapeutic categories is considerable, the upside from collusion for the first and second respondent is relatively slight, too slight, on the face of it, to risk detection. Has there been a substantial lessening or prevention of competition? – Conclusion 78. We conclude that the claimant has not adduced sufficient evidence to prove, even on the lower standard of proof specified in the amended Act, that the vertical agreements constitute a restrictive practice, that is, a practice that gives rise to a substantial lessening or prevention of competition. 79. Given that the claimant has failed to establish the existence of a restrictive practice, the requirement to examine whether the agreement gives rise to any technological, efficiency or other pro-competitive gains falls away, as does the 21 necessity for examining the other requirements necessary for supporting a claim for interim relief. Accordingly, the application for interim relief is denied. COSTS 80. The complainant is ordered to pay the respondents’ costs in the application on a party and party scale, including the costs of two legal representatives. ___________________ 12 March 2001 David Lewis Date Concurring: Urmilla Bhoola and Norman Manoim