England and Wales High Court (Administrative Court) Decisions

QUEEN And The SECRETARY OF STATE FOR HEALTH Ex parte BRITISH ASSOCIATION OF EUROPEAN PHARMACEUTICAL DISTRIBUTORS And DOWELHURST LIMITED ASSOCIATION OF BRITISH PHARMACEUTICAL INDUSTRY [2001] EWHC Admin 183 (14th March, 2001)

Case No: CO/4132/1999

Neutral Citation Number: [2001] EWHC ADMIN 183

IN THE HIGH COURT OF JUSTICE

QUEENS BENCH DIVISION

ADMINISTRATIVE COURT

Civil Justice Centre

Chester

Wednesday 14^th March 2001

B e f o r e :

THE HONOURABLE MR JUSTICE THOMAS

- - - - - - - - - - - - - - - - - - - - -

MR JUSTICE THOMAS:

I: INTRODUCTION

1. There is before the court an application by the Association of Pharmaceutical Importers (API) (now known as the British Association of European Pharmaceutical Distributors) and one of their members, Dowelhurst Limited, against the Secretary of State for Health (the Department) seeking relief in respect of certain provisions of Pharmaceutical Price Regulation Scheme made in July 1999 (the PPRS) which they contend are contrary to provisions of community law, Articles 28EC (formerly Article 30) and 81EC (formerly Article 85).

2. The PPRS is a voluntary scheme agreed between the Department and the Association of the British Pharmaceutical Industry (ABPI) to achieve a price reduction and regulate the profits made from pharmaceutical products purchased by the National Health Service. It will be necessary to set out the details of the PPRS in due course, but it is more convenient first of all to explain in outline the operation of the market in pharmaceuticals, as it appeared from the evidence.

3. The API is the representative body for parallel importers of pharmaceuticals into the UK; out of the 16 companies licensed to bring parallel imports of pharmaceuticals into the UK 14 are members of the API and one is in the process of joining; Dowelhurst is one of the largest.

The market in pharmaceutical products

4. On the evidence before the court, there was little dispute about the operation of the market in pharmaceutical products in the United Kingdom.

5. The dominant purchaser in the market is the Department which through the National Health Service (NHS) spends approximately £7bn on medicines each year. It does not have, however, the power of a dominant or almost monopoly purchaser to force down prices through the ordinary operation of a market. Although the Department pays for the purchase made by the NHS and can be described as "a monopoly payer" (a description put forward by the Department), the decision to initiate the purchase of individual products is made by individual medical practitioners in response to the needs and demands of the ultimate consumer, the patient. Doctors who respond to the therapeutic needs of patients are only affected to a limited extent by price; although the Department has made extensive efforts to make available information about pricing to medical practitioners and introduced schemes such as Primary Care Groups and established the National Institute for Clinical Excellence, the evidence is clear that, because therapeutic need must be the major determinant in prescribing a pharmaceutical, there is more limited sensitivity to price. Thus the operation of the relationship between buyer and seller in this market is very different to an ordinary market.

6. The second characteristic of the market is the nature of the product. There are three distinct categories:

i) Branded products protected by patents where there is no therapeutic interchangeability:

At any one time there are on the market products protected by a patent where there is no therapeutic alternative.

ii) Branded products protected by a patent where there is a therapeutic interchangeability -

There are also on the market at any one time many instances where two or more products objectively can perform the same therapeutic function; they are referred to as branded products within the same therapeutic class. These products therefore compete against each other. However the competition is limited by the fact that such products may not be interchangeable in respect of a particular patient. The different chemical or molecular composition of pharmaceutical products within the same therapeutic class may affect individual patients in different ways, in particular as regards their side effects. Moreover medical practitioners, if persuaded of the particular value of a branded product, may be reluctant to use another branded product within the same therapeutic class, because of the risks of different side effects.

iii) Branded products not protected by patents where there is competition from generics

When a pharmaceutical product loses its patent protection or its supplementary protection certificate, other manufacturers will generally (save where the market is very small) produce identical copies of the product. These are then marketed in direct competition with the branded product which no longer has patent protection. These products, known as "generics", are sold at a price generally far below that which the product enjoyed when it was protected by a patent; the manufacturer of a generic will have no research costs and very low development costs. The holder of the patent must therefore drop the price dramatically at about the time it loses the protection of the patent, if it wishes to maintain a sufficient volume of sales; it is not usually necessary for the manufacturer to drop the price to the level at which the generic is sold, as the value inherent in the brand will generally enable the branded product to command a premium. Where there are generic alternatives, there is strong and effective competition.

7. The market also has a high degree of opacity; each manufacturers' pricing policy is commercially confidential. Many of the manufacturers of branded products supplied to the NHS operate on a transnational basis and manufacture products in the UK and overseas; a significant proportion of products supplied by these companies are therefore imported.

Government intervention in the market

8. The characteristics of the market in other developed countries, apart from the United States where there is much less state funded health care, are similar. Because of these characteristics, governments intervene in the market to control the price of prescription medicines. The governments of the EU use one or more of the three following methods:

i) In 9 out of the 15 EU Member States, the government controls the price of prescription medicines directly. It sets the price either by negotiation or by reference to prices charged in other countries.

ii) Most of the other EU member governments intervene in the markets by setting a price for a group of products which the government considers interchangeable; this is known as reference pricing and the price set becomes the maximum price which will be reimbursed by the government.

iii) The government of the UK intervenes primarily by controlling the profits of pharmaceutical manufacturers. This policy is implemented by the Department. It is described by the Department as "relatively light touch regulation" as it seeks to influence the market less than rigid price fixing. It will be necessary to describe this in more detail.

9. The market has been the subject of EU action. For example, the Transparency Directive, 1989/105/EEC, on transparency in governmental drugs pricing regimes (the recital to which emphasises the inadequacy of competition in the market) specifies a number of requirements intended to ensure that action taken by Member States to control pharmaceutical prices does not infringe provisions of Community law relating to the importation of goods; they are required to publish certain information.

The parallel trade

10. Although parallel trade is a feature of several markets, it has become important in the market in the EU in pharmaceuticals because of the wide differential in the price paid or reimbursed for the same branded product between some states and other states. This differential has arisen as a result of the different policies to which I have referred. It is, therefore, profitable for traders to import branded products from a state where the price is low to a state where the prevailing price is higher. This is particularly the case where in some southern EU Member States the price fixed by the state is much lower than the prices at which products are marketed by the manufacturer in the UK. A parallel importer will identify such products, seek a licence and then import them into the UK, making the necessary changes to the labelling and instructions which UK regulations require.

11. The major category of pharmaceutical product against which parallel imports are made are branded pharmaceuticals still subject to patent protection where there is no therapeutic alternative. Parallel imports provide the only competition in such cases.

12. There have been other factors which have made parallel importing attractive. First, the recent strength of Sterling relative to the Euro has increased price differentials. Secondly the system for reimbursing pharmacists encourages them to find the cheapest source of supply, as they are reimbursed on the basis of scales reflecting average prices and discounts; if a pharmacist can purchase at prices below the scale, he is entitled to keep the balance.

13. The evidence before the Court was that in 1994, parallel imports represented 3% of the market; by 1998, their share had risen to 7%. There was a 226% rise in the sales figures of the members of the API between 1995 and 1998. In the period from July 1999 (the inception of the PPRS) to August 2000, the market share of parallel imports has risen further from 10.5% to 12.4%.

14. The evidence filed on behalf of the API was that it was commercially worthwhile for a trader to import a parallel pharmaceutical product where the price at which the product could be purchased in another EU Member State was about 20% lower than that in the UK.

15. In many trades parallel imports are welcomed as they offer goods at a price lower than that charged by the appointed importer or the manufacturer; in de Peijper [1976] ECR 613, the European Court recognised the important role they can play in the market in pharmaceutical products. However the evidence of the Department was that parallel imports of pharmaceutical goods are not necessarily to be welcomed; the difference in price is not reflected in the price paid in the UK, as the importer, like any arbitrageur, retains a significant part of the price differential as profit. Parallel imports are also considered by some to act to the detriment of the pharmaceutical manufacturers and their ability to invest large sums in the research and development of their products.

The system of state intervention in the UK

16. The policy of controlling prices in the UK has primarily been based upon voluntary arrangements between the Department and the pharmaceutical industry. This has been so for two principal reasons:

i) to encourage the development of a successful pharmaceutical industry and a strong research and development base;

ii) to enable the government to negotiate with the pharmaceutical industry every 5 years or so to reach an agreement which meets the changing conditions in the market.

However the current voluntary arrangements in the PPRS which are made under s. 33 of the Health Act 1999 have a statutory fall back under other provisions of that Act and regulations made thereunder, to which it will be necessary briefly to refer.

All branded products are included whether or not they have patent protection. The scheme does not apply to the portion of branded "over the counter products" which are in fact sold to the public without prescription; nor does it apply to generics.

21. The evidence of the Department was that branded licensed medicines represented about 80% by value of pharmaceutical sales the NHS. This is, however, in terms of volume, under 50%.

22. Those companies which join the scheme are subject to two primary controls:

i) A control over their profits. Chapters 8-17 contain detailed provisions providing for financial returns and other information to be provided so that the Department can monitor profitability for the purposes of the PPRS which is based on target return on capital. The provisions regarding the supply of information to the Department also enable them to understand the market better.

ii) A reduction in prices.

23. API make no complaint about the provisions of the PPRS that impose controls over profits; it is not necessary therefore to set out details of that part of the PPRS. Their complaint is solely directed at one part of the PPRS dealing with the price reduction. It is therefore necessary to describe those provisions in a little more detail.

24. The PPRS provides that the prices of the medicines covered by the PPRS were to be reduced by 4.5% from 1 October 1999 (see chapter 18.1) and to remain at that level until 31 December 2000. New products are provided for in chapter 20 and generally can be priced on entering the market at the discretion of the company. From 1 January 2001, scheme members have been entitled to apply for price increases under the provisions of chapter 19.

25. The price reduction of 4.5% was applied to each product covered by the scheme under the provisions of chapter 18.1 and Annex C. However that overall price reduction is subject to provisions for "modulation". It is these provisions for modulation that are the subject of the complaint by API because they say they are targeted at and discriminate against parallel importers.

26. In the same way that the Department is concerned to protect its financial position by restricting the profits of pharmaceutical companies, its concern to achieve a 4.5% price reduction is a concern to achieve an overall price cut. As far as the Department is concerned, from the point of view of the effect on their budget and expenditure, it matters not how the scheme members cut prices on individual products, so long as an overall reduction of 4.5% is obtained. The ABPI and the Department therefore agreed the principles set out in chapter 21 of the PPRS which on specified terms entitle scheme participants to vary their price reductions, provided an overall reduction of 4.5% is achieved. It is necessary to set out the whole of this chapter:

"Principles underlying modulation as an alternative to a price reduction of 4.5% on all products

21.1 As an alternative to an across the board price reduction, scheme members may modulate the list price of their PPRS products by reductions that equate to an overall level of 4.5%. Product list prices may be increased or decreased during the period 1 October 1999 to 31 December 2000, provided that they do not exceed those that prevailed on 1 August 1999. Modulation will be deemed to have occurred where:

* List prices have been reduced by a percentage other than 4.5%.

* List prices remain unchanged from those that prevailed on 30 September 1999.

21.2 Companies can remodulate at any time from 1 October 1999, provided the Department is notified 21 days in advance of the implementation of the price change. The Department will have 14 days in which to respond to modulation notifications and will only withhold agreement where it can be shown that the effect would place the delivery of the price reduction in doubt.

21.3 Companies will not be permitted to substitute discounts or contract prices already in place before 30 September 1999. Price reductions made on products where the patent or supplementary protection certificate expires after 1 July 1999 and before 1 January 2001 will not be allowed in calculations of modulations or overall adjustments made to achieve the price reduction

Modulation principles after 1 January 2001

21.4 The Department is keen to minimise interference in the conduct of companies' commercial affairs consistent with safeguarding public expenditure. Companies are permitted to modulate the prices of products provided that the effect of the modulation is cost neutral.

21.5 From 1 January 2001 list prices may be increased to a level no greater than 20% above the level that existed on 1 August 1999 subject to the agreement of the Department. The Department will consider applications for increases of more than 20% for products with NHS sales of £100,000 or less where a medical need can be justified. The Department may withhold agreement to any price increase of 10% or more in any one year.

21.6 The prices of new products introduced after 1 October 1999 can be increased by up to 20% after 1 January 2001. Any reduction in the price of a new product cannot be used to offset price increases on other products until the new product has been on the market in the UK for two years.

21.7 Scheme members will not be permitted to use price reductions that may be necessary as a result of patent or supplementary protection certificate expiry to justify a price increase on other products. Consequently scheme members will not be allowed to include in their modulation proposals price reductions made on products where the patent or supplementary protection certificate has expired within one year before, or will expire within two years after, the proposed date for modulation. Where a competitor product enters the market within two years of patent or supplementary protection certificate expiry, the exclusion period for modulation purposes will be extended to a maximum of 2 years from the market entry of the competitor product."

27. Although each of the earlier schemes had imposed across the board price reductions (though not as deep as 4.5%), they had not contained express provisions for modulation within the formal scheme; however, modulation of prices was agreed between participants and the Department. The Department's Report to Parliament in December 1997 recorded that out of the £95m that had been saved by the 2.5% cut imposed by the 1993 scheme, £72m was accounted for by companies which modulated the reduction. Throughout the entirety of the scheme there were 914 modulations of which 569 were increases and 334 decreases; some of the price cuts were deep, 35.6% being greater than 20% of the previous price.

28. The modulation provisions under the PPRS are divided into two periods:

i) In the transitional period from 1 October 1999 to 31 December 2000, the scope for modulation was limited because a scheme member could modulate only if the price increase remained below the list price as at 1 August 1999.

ii) In the period from 1 January 2001 until the expiry of the PPRS in October 2004, a scheme member can increase prices up to a level 20% above that prevailing on 1 August 1999; the scope for modulation is therefore greater.

29. Under the interim stage of the 1999 PPRS prior to 1 January 2001, scheme members could modulate provided that the Department was notified 21 days in advance; its power to withhold agreement were limited by the terms of chapter 21.2. In the period on and after 1 January 2001, the policy is spelt out in chapter 21.4 and the express agreement of the Department is required under chapter 21.5.

The statutory provisions: ss 33 - 38 of the Health Act 1999

30. S. 33 of the Health Act 1999 (under which the PPRS was made) provides that the Secretary of State can enter into schemes with an industry body for the purpose of limiting prices or profits; a manufacturer or supplier to be bound by the scheme has to consent to the scheme being applied to him. The statutory provisions include requirements for the supply of information.

31. Under s. 35, the Secretary of State is empowered to adopt a statutory scheme for limiting prices and profits, but the provisions of the statutory scheme cannot be applied to a member of the PPRS, as s. 33(7) provides that statutory schemes do not apply to members of voluntary schemes made under s. 33.

32. On 24 January 2000, the Secretary of State made The Health Services (Control Prices of Branded Medicines) Regulations 2000 (SI 2000 No 123) imposing a 4.5% price cut in respect of branded medicines supplied to the NHS by reference to a price list published by the Department on its internet web site; other provisions enable manufacturers or suppliers to apply for a price increase. Like the PPRS it does not impose a price cut in respect of generics. However, unlike the PPRS, it contains no modulation provisions; it is an across the board measure and allows no commercial latitude. The evidence of the Department was that this omission was intended as an incentive for all suppliers and manufacturers of branded products to join the PPRS, as the statutory scheme does not apply to those who are members of the PPRS.

API's complaint

33. On 8 April 1999, the then Secretary of State, Mr Frank Dobson, made a speech at a dinner given by the ABPI during the course of which he referred to parallel imports in the context of the PPRS. It will be necessary to refer to that speech in more detail, but in a report of the speech in the Financial Times he was quoted as saying that the Government would

" "consider a system which gives extra rewards for the newer, research based products and will use the PPRS as a method of trying to reduce the incidence of parallel imports"

It was the arbitrageurs or middlemen, not the NHS or the industry, that gained from parallel imports, he said - the activity whereby drugs sold in foreign markets at lower prices are brought back into the UK to undercut the UK price.

...... "For every pound that the NHS saves through parallel imports, the industry loses £6... That's bad for industry and bad for Britain. We are determined to help deal with the problem..." "

The report commented that what the Minister appeared to have in mind was to persuade the industry to make price cuts on popular drugs which were parallel imported as a way to save the NHS cash and cut out the middlemen; this would involve differential price cuts on drugs rather than an across the board reduction. As a result of this newspaper report, the API entered into correspondence and had meetings with the Department about the PPRS which was then being negotiated with the ABPI. The negotiations for the PPRS were concluded and the PPRS signed on 14 July 1999.

34. In October 1999, the API commenced these proceedings. In their grounds lodged with the court, they alleged that the modulation provisions enabled (and were specifically designed to enable) scheme members to target price reductions on products on which they faced a competitive challenge from parallel imports so as to reduce or eliminate the viability of parallel imports of competing products. They further alleged that scheme members had taken advantage of the modulation provisions to reduce parallel imports by cutting prices of products subject to competition by more than 4.5%. The modulation provisions of the PPRS therefore had as their object and effect the restricting of parallel imports from other EU states into the UK and were therefore contrary to Article 28 EC or Article 81 EC.

35. In response to this, a substantial amount of evidence was filed on behalf of the Department and ABPI; some of the evidence was directed at showing that the provisions of the PPRS were not targeted at parallel imports but simply facilitated ordinary market competition.

36. In the skeleton argument lodged by API with the court prior to the hearing, their contentions were developed primarily in reliance on the evidence filed in relation to the significance of parallel imports in the negotiations that led to the PPRS and the effect of the scheme in practice. However, in his very clear opening submissions, made orally and in writing, on the first day of the hearing, Mr Nicholas Green QC advanced a further argument which can be summarised as follows:

* There are three sources of competition to branded licensed products that are the subject of the PPRS:

i) competition from other branded products within the same therapeutic class;

ii) competition from a generic where the patent on the branded product had expired;

iii) competition from parallel imports of an identical product.

* However, the price modulation provisions by their terms (chapter 21.3 and chapter 21.7) exclude modulation in respect of competition from the preponderant proportion of generics. Although it is possible to modulate in respect of competition from branded products within the same therapeutic class, that competition is limited; there is little scope for modulation and the modulation provisions were not needed in respect of that competition.

* Therefore parallel imports were the only major source of competition against which the modulation provisions could be used to a material degree. The provisions were therefore plainly discriminatory.

37. At the end of the second day of the hearing, the Department and the ABPI raised an objection to the further argument put forward by Mr Green in his opening on the basis it had not been advanced in the grounds or Form 86A or otherwise notified to them in sufficient time in advance of the hearing. It required a factual enquiry on two principal matters for which the Department and the ABPI had not adduced evidence:

i) whether the PPRS in effect excluded the use of modulation against the preponderant proportion of generic products;

ii) whether the competition from branded products within the same therapeutic class was so limited as to be of no practical importance.

Such evidence was, it was contended, essential if the inference was to be drawn that the modulation provisions of the PPRS were discriminatory by reason of the fact that these provisions of the PPRS could, in practice, only be directed to a material degree at parallel imports.

38. It was agreed that the better course, in the time available, was for me to determine that issue as part of my judgment.

The issues

39. In the course of the hearing the parties helpfully formulated the issues that arose, although they were not agreed as to the precise formulation of those issues or the order in which they arose. Having regard to the whole of the evidence and argument, it seems to me convenient to consider first the scope of API's complaint, then the arguments under Article 28 and finally the arguments under Article 81.

II: THE SCOPE OF API's COMPLAINT

40. API did not allege that the profit control provisions of the PPRS or the 4.5% price reduction of the PPRS made across the board contravened Articles 28 or 81. They contended that only the modulation provisions contravened those Articles. Thus they contended it was not necessary to examine the PPRS as a whole, but only the modulation provisions. The Department submitted that it was not permissible to adopt this approach and the PPRS as a whole should be examined.

41. API initially made it clear that they did not seek to strike down the PPRS as a whole and would not want to do so. It was their contention that if I came to the view that the modulation provisions contained in Chapter 21 were contrary to Article 28 or Article 81, then the PPRS as a whole was not affected and the modulation provisions could be severed. However, at the conclusion of their case, they did not address any argument on this issue and seemed to accept that if the provisions of chapter 21 were illegal, the whole of the PPRS was illegal.

42. ABPI and the Department both contended that the modulation provisions could not be severed; API did not address any specific argument to the contrary. It might have been important to decide whether, for the purposes of severance, because the PPRS is an agreement, the contractual test was applicable or whether, because of the nature of the PPRS, the public law test set out in cases such as Kent County Council v Kingsway Investments [1971] AC 72 and DPP v Hutchinson [1990] 2 AC 783 applied. In R v North Hertfordshire District Council ex p Cobbold [1985] 3 All ER 486, Mann J, after setting out a number of the authorities in public law, applied the test of whether a discrete part could be excised without altering the character or substance of the rest. In contract the test can be formulated as whether severing the provisions would alter entirely the scope and intention of the agreement (see the authorities cited in Chitty 28^th edition at paragraph 17-189). However in the argument before me it was not suggested that there was any material difference in the private and public law tests for severance as applied in the circumstances of this case.

43. In my view whatever test is applied it is not possible to sever the modulation provisions. It is clear on the evidence that they are an integral part of the agreement made; they are an essential part of characterising it as "light touch regulation", as they leave substantial room for free market competition. Without them, the provisions of the agreement would have been structured differently. Indeed agreement may not have been reached, as I accept that the ABPI would probably not have agreed to a 4.5% price cut if modulation had not been included. Without the modulation provisions, therefore the character and the substance of the scheme would have been fundamentally altered; the substance would have been materially changed.

44. Thus, in considering the submissions of API, it is necessary to bear in mind that the modulation provisions are an integral part of the PPRS and that their attack must necessarily be on the scheme as a whole, though this does not mean that one cannot focus on the modulation provisions. The question of severance also give rise to an issue on the form of relief, but it is more convenient to deal with that issue after determining whether there has been a breach of Article 28 or 81.

45. I will therefore consider the two principal arguments raised on the basis that it is necessary to examine the PPRS as a whole.

III: ARTICLE 28 EC

46. Article 28 (formerly Article 30) provides:

"Quantitative restrictions on imports and all measures having equivalent effect shall,..., be prohibited between Member States."

47. It was API's primary case that the modulation provisions of the PPRS were a breach of Article 28. An examination of their case is most conveniently considered by asking five questions:

i) Do the price modulation provisions of the PPRS constitute a state measure?

ii) Are the modulation provisions a restriction on, or hindrance to, imports within the scope of Article 28?

iii) Have API shown that there has been an actual effect on imports under Article 28?

iv) If so, do the modulation provisions, as a selling arrangement within paragraph 16 of the decision of the European Court of Justice in Keck [1993] ECR1 - 6097 apply

a) to all relevant traders?

b) without discrimination between domestic goods and imports?

v) If there was a breach of Article 28, can it be justified under Article 30?

(i) Do the price modulation provisions of the PPRS constitute a state measure?

48. API advanced four main contentions:

* The PPRS as a whole is a state measure and therefore the modulation provisions in chapter 21 must be.

* The modulation provisions are a state measure as they are a facility through which modulation can be made.

* They also embody a decision by the state systematically to pay more on some products than it would ordinarily be prepared to pay so that scheme members could fund price competition with parallel importers on other products.

* The provisions of chapter 21.2 (which were applicable until 31 December 2000) required notification to the Department and gave it power to withhold agreement; the provisions of chapter 21.5 (which have been applicable since 1 January 2001) require the consent of the Department to price increases. Whether the Department acts or does not act, each modulation is a state measure, as it is subject to the exercise of governmental power.

49. It was not disputed that the PPRS as a whole is a state measure within the meaning of Article 28; API's complaint was however directed by them, not against the scheme as a whole, but against the modulation provisions.

50. API relied on the evidence of Mr Brownlee of the Department that, although the intention of the PPRS was that members should be free to regulate their own commercial affairs (consistent with the need to safeguard public expenditure), the Department was aware of its legal obligations, both under domestic public law and EC law; that these obligations could override the provisions of the PPRS and entail a somewhat broader scope for refusing requests for modulation than was apparent on the face of the scheme. API contended that the Department therefore must, on the basis of this evidence, have accepted that it had a role in the modulation of prices, though, as scheme members were not required to give reasons in their notifications of or requests for modulation, it was difficult to see how the Department could ever monitor the use made of modulation by the scheme members.

51. I do not think that Mr Brownlee's evidence assists API in its contention; it is only a statement of the general duty of HM Government. It is necessary, in my view, to consider the nature of the PPRS and of modulation provisions in Chapter 21. The modulation provisions should, in my view, be characterised as permissive provisions enabling price alterations, provided that they are cost neutral to the Department; they permit price competition within the market provided that the overall price reduction is achieved by the scheme member making the modulation.

52. The question therefore is whether the PPRS as a whole or the modulation provisions on their own are a state measure within the relevant authorities. In Commission Directive 70/50/EEC, Recital 1, measures are defined to mean for the purposes of Article 28:

"laws, regulations, administrative provisions, administrative practices, and all instruments issuing from a public authority, including recommendations

...for the purposes of this directive, "administrative practices" means any standard and regularly followed procedure of a public authority; whereas "recommendations" means any instruments issuing from a public authority which, while not legally binding on the addressees thereof, cause them to pursue certain conduct."

53. In Commission v Ireland [1982] ECR 4005, the European Court of Justice held that a "buy Irish" campaign was, on the facts, within Article 28 because of the role of the state in the sponsoring of the advertising campaign and a "Guaranteed Irish" symbol. The Court said at paragraphs 27-29 :

"In the circumstances the two activities in question amount to the establishment of a national practice, introduced by the Irish Government and prosecuted with its assistance, the potential effect of which on imports from other Member States is comparable to that resulting from government measures of a binding nature.

Such a practice cannot escape the prohibition laid down by Article [28] of the Treaty solely because it is not based on decisions which are binding upon undertakings. Even measures adopted by the government of a Member State which do not have binding effect may be capable of influencing the conduct of traders and consumers in that State and thus of frustrating the aims of the Community as set out in Article 2 and enlarged upon in Article 3 of the Treaty.

That is the case where, as in this instance, such a restrictive practice represents the implementation of a programme defined by the government which affects the national economy as a whole and which is intended to check the flow of trade between Member States by encouraging the purchase of domestic products, by means of an advertising campaign on a national scale and the organisation of special procedures applicable solely to domestic products, and where those activities are attributable as a whole to the government and are pursued in an organised fashion throughout the national territory."

On the facts of that case, the actions of the Government, though not binding on private undertakings, were state measures because they were attributable to the state.

54. An omission to act can also, in certain circumstances, amount to a measure; in Commission v France [1997] ECR I-6959, the European Court of Justice held that the government was obliged to take action against protesters blocking imports, as it was the government's duty to see that the fundamental freedom under Article 28 was respected in their territory. The Court stated at paragraphs 31-32:

"31. The fact that a Member State abstains from taking action or, as the case may be, fails to adopt adequate measures to prevent obstacles to the free movement of goods that are created, in particular, by actions by private individuals on its territory aimed at products originating in other Member States is just as likely to obstruct intra-Community trade as is a positive act.

32. Article [28] therefore requires the Member States not merely themselves to abstain from adopting measures or engaging in conduct liable to constitute an obstacle to trade but also, when read with Article 5 of the Treaty, to take all necessary and appropriate measures to ensure that that fundamental freedom is respected on their territory. "

55. In my view, as the PPRS is accepted to be a state measure, the modulation provisions which are an integral part of it must be a state measure; it is a facility provided by the state within the scheme as a whole. That is sufficient to answer this question in API's favour; I can therefore consider API's other arguments more briefly. To the extent API's argument proceeded on the basis that the modulation provisions have to be considered independently (which for the reasons I have given they cannot), then I do not accept those other arguments.

* The individual decisions to make price modulations are decisions of the individual members of the scheme, not of the state. The scheme is permissive; it is not like the scheme in Commission v Ireland where the state sponsored the campaign which was directed at the promotion of domestic products. Modulations are not, in my view, attributable to the state.

* I do not consider that there is any element of state subsidy or recoupment for reasons which I give below at paragraphs 70 to 75. In short the ability of the scheme members to make deeper reductions on certain products by charging more on other products is no different to what happens in the market where the ability of a supplier to charge a lower margin on one product may often depend on his ability to charge a higher margin on another.

* Consent to a price increase does not in my view make that price increase a state measure within the meaning of the authorities. The act of increasing the price remains the act of the scheme member and not the act of the state.

(ii) Are the modulation provisions a restriction on, or hindrance to, imports within the scope of Article 28?

56. The next issue was formulated differently by the parties; the Department formulated the issue in terms "was the reservation of the freedom to compete on price a restriction on or hindrance to imports within Article 28?" It is more convenient to take API's more neutral formulation.

The case law

57. At the heart of API's submissions was the submission that the modulation provisions would have an effect or potential effect on parallel imports; it mattered not that they might not be able to identify the effect directly, as an indirect effect was enough. Furthermore there was no de minimis requirement. They relied on the line of decisions beginning with Dassonville [1974] ECR 837, where the European Court of Justice held in respect of Article 28 that :

"All trading rules enacted by Member States which are capable of hindering directly or indirectly, actually or potentially, intra-Community trade are to be considered as measures having an effect equivalent to quantitative restrictions."

58. In Van der Haar [1984] ECR I -6197, when the European Court of Justice again considered Article 28, it held that it was sufficient if the measures which infringed Article 28 only had a slight actual or potential effect:

"12. Article [28], on the other hand, belongs to the rules which seek to ensure the free movement of goods and, to that end, to eliminate measures taken by Member States which might in any way impede such free movement. Thus the Court has held that a national provision which is capable of hindering intra-Community trade, directly or indirectly, actually or potentially, must be regarded as a measure having an effect equivalent to a quantitative restriction."

13. It must be emphasised in that connection that Article [28] of the Treaty does not distinguish between measures having an effect equivalent to quantitative restrictions according to the degree to which trade between Member States is affected. If a national measure is capable of hindering imports it must be regarded as a measure having an effect equivalent to a quantitative restriction, even though the hindrance is slight and even though it is possible for imported products to be marketed in other ways.

14. ... A court called upon to consider whether national legislation is compatible with Article [28] of the Treaty must decide whether the measure in question is capable of hindering, directly or indirectly, actually or potentially, intra-Community trade. That may be the case even though the hindrance is slight and even though it is possible for imported products to be marketed in other ways.

59. The sufficiency of potential effect was emphasised by the European Court of Justice in Commission v France [1997] ECR 1-6197 - see paragraph 17 where the court stated that Article 28 applied even if there were no actual cases where the measure had had an effect; the potential effect was enough.

60. It is, however, clear that an across the board price reduction of 4.5% under the PPRS (without modulation) would be capable of restricting or hindering imports, as the state's action in lowering prices would make importing less attractive. However API accepted that such a price cut would not be a breach of Article 28. Such a price reduction would only infringe Article 28 if the price had been cut to such a level that the sale of imported products became either impossible or more difficult than that of domestic products so that they could not be sold for a profit: see Roussel Laboratoria v Netherlands [1983] ECR 3849. In Commission v Belgium [1991] ECR I - 1275, the European Court of Justice summarised the case law:

"15. On this specific point, which is the one at issue, the Court has held on several occasions that price control systems applicable to domestic products and imported products alike, although not in themselves constituting measures having an equivalent effect to a quantitative restriction contrary to Article [28] of the Treaty, may nevertheless have such an effect when the prices are fixed at a level such that the sale of imported products becomes either impossible or more difficult than that of domestic products (see in particular the judgment in Case 181/82 Roussel Laboratoria v Netherlands [1983] ECR 3849, paragraph 17).

16. It should be added that the same applies in the case of a system for fixing the prices of individual products, such as that challenged by the Commission. Even if on account of the policy followed by the Government of ensuring a moderate level of prices for pharmaceutical products and imported products, it is sufficient that the circulation of the latter be impeded for that system to constitute a measure having equivalent effect to quantitative restriction.

17. Such a hindrance occurs when imported products cannot be sold at a reasonable profit on the market of the State of importation.

61. On the basis of the case law, it is accepted by API that, in this market for branded pharmaceutical products, a simple 4.5% across the board price cut without modulation would not have amounted to a hindrance.

62. Something more was therefore required as the case law makes clear. In Cullet v Leclerc [1985] ECR 305, a question was raised as to whether a national system imposed by the Government of France for fixing minimum prices on fuel was capable of having an adverse effect on imported products, primarily because the lower price of imported fuel was not reflected in the retail selling price. The European Court of Justice set out the applicable principle at paragraph 23:

"It should be noted in the first place that, as the Court has consistently held, the prohibition laid down in Article [28] of measures having an effect equivalent to a quantitative restriction covers any measures which are capable of hindering, directly or indirectly, actually or potentially, imports between Member States. As regards the application of those principles to State systems of price control, the Court has repeatedly stated that such systems, if applicable to domestic products and imported products alike, do not in themselves constitute measures having effect equivalent to a quantitative restriction but may have such an effect when the prices are fixed at a level such that imported products are placed at a disadvantage compared to identical domestic products, either because they cannot profitably be marketed on the conditions laid down or because the competitive advantage conferred by lower cost prices is cancelled out."

63. In Commission v Italy [1987] ECR 2919 the European Court of Justice considered an Italian scheme for fixing the prices of pharmaceutical products which specifically permitted the allowable costs which were to be taken into account in fixing prices to be increased if the work or research was done in Italy. The Court stated that:

"6.... Article [28] of the Treaty precludes national price control schemes in so far as such schemes fix price components differently for domestic and imported products, to the detriment of the latter, preclude any increase in the price of the imported product corresponding to the supplementary costs and charges inherent in the importation or fix the prices of products solely on the basis of the cost price or the quality of domestic products at such a level as to create a hindrance to importation.

...

8....It must therefore be recognised that the new method introduced .. is likely to promote domestic products to the detriment of imported products and, therefore, that it constitutes a measure having an effect equivalent to quantitative restrictions on imports within the meaning of Article [28] of the Treaty."

64. Whichever test is applied from the case law, it is clear that a price control measure will infringe Article 28 if it is capable of promoting domestic products to the detriment of imported products or of putting imported products at a disadvantage to domestic products or is in some other way capable of being discriminatory against imports. Thus it is necessary to see whether, on the facts, the PPRS can have this potential effect and to scrutinise the PPRS accordingly.

65. It was submitted by API that the modulation provisions did have this potential effect and were capable of promoting domestic products to the detriment of imported products; the provisions were discriminatory and did not permit simple competition.

API's case on the discriminatory nature of the modulation provisions

66. API contended that the modulation provisions were different to ordinary competition in two respects:

i) They were an artificial stimulant to certain types of competition by making the price lower than would otherwise be the case through a state sponsored price subsidy.

* Where there was normal market competition, a seller would drop its prices to compete; the 4.5% reduction required by PPRS did not in any way inhibit further competition on price, as scheme members were free to reduce their prices further.

* However, in normal market competition a buyer would not usually agree to pay more for unrelated products.

* A buyer would also ordinarily welcome parallel imports, because they are cheaper than the self same product provided by the manufacturer or his supplier in the UK.

* However, under the modulation provisions the Department agreed in effect to subsidise a scheme member who cut prices on one product by agreeing to pay a higher price for an unrelated product supplied to it. The extra payment would offset the loss sustained by the scheme member when it reduced its prices to compete.

* This cross subsidy was an artificial and government funded boost to the ability of the scheme member to drop prices and compete with a rival source to a greater degree than would otherwise occur. It entailed the state subsidising deeper price cuts than would otherwise occur.

* As API's submission emphasised, the essence of their contention was that modulation differed from ordinary competition, because the Department, as the payer, agreed to pay more for an unrelated product to ease the consequences to scheme members that would otherwise result. The Department's state sponsored price was state intervention that made parallel imports more difficult and often impossible.

Indeed they suggested that any significant form of flexibility in delivering a price cut was capable of promoting domestic products over imports, because it would enable producers to discount against imports.

ii) The provisions were targeted at parallel imports, as they were not neutral in their intended effect.

* There was no logic in excluding the possibility of modulation against generics and permitting it against parallel imports.

* Construed objectively in their context the modulation provisions were inherently more likely to impact on parallel importers than upon the generality of branded competition.

The modulation provisions were therefore discriminatory on their face and by their very nature.

67. It is necessary to examine the arguments in more detail.

Are price movements under modulation different from ordinary competition? Are they a state subsidy?

68. It is clear from the evidence before the court that ABPI's aim in the negotiations which led to the PPRS in 1999 was to achieve complete de-regulation in the pharmaceutical market. ABPI contended, despite the characteristics of the market to which I have referred, that de-regulation allowing free market competition would prevent pharmaceutical companies making excessive profits at the expense of the NHS. They knew, however, that they would not achieve their aim of complete deregulation, but made clear in the negotiations the importance of competition in the market, particularly that provided by parallel importers. Their evidence was that they regarded modulation as critical. It would enable members to set their own prices so as to be able to respond on price to competitive pressures from generics, branded products within the same therapeutic class and parallel imports, with the objective of winning more market share or defending their market share. They viewed modulation, whilst preserving cost savings to the NHS, as allowing companies some commercial freedom to price their products competitively and therefore as ameliorating the worst restrictions on free competition imposed by the PPRS.

69. It is clear that under the modulation provisions, members of the PPRS were enabled to make deeper price cuts on some products by charging more for others, so long as the overall reduction was achieved.

70. API contended that it did not matter if the provisions could be used against other competition provided that they could be used against parallel importers, as the PPRS involved the agreement by the Department to a price cut greater than would otherwise be the case to prevent or reduce competition from parallel imports; the Department was paying for that price cut by agreeing to pay more on other products. They were thus subsidising price cutting that could be directed at parallel imports. This was different in the effect it was likely to have than an across the board price cut, though as regards the Department the effect was cost neutral.

71. I do not agree. In effect, what the PPRS did was to establish a mechanism to allow scheme members to deliver the 4.5% price cut, subject to certain restrictions, in a manner of their choosing. It is not and cannot be contended that the requirement that there be an overall 4.5% price cut is unlawful. Nor it seems to me could it be contended that, if the Department had merely stipulated that there was to be an overall reduction of 4.5% by each supplier and left it to each of the pharmaceutical companies, subject to rules on predatory pricing and other anti-competitive behaviour, how to achieve that, such a stipulation would have been unlawful. Allowing the members to deliver that price cut in a manner of their choosing could not, it seems to me, be unlawful, for it would simply be allowing them the commercial freedom of the market subject to the ordinary rules on anti-competitive behaviour.

72. Indeed this is confirmed by the fact that in the Transparency Directive, to which I referred in paragraph 9, it is implicit that a scheme based on a profit ceiling is permissible.

73. The modulation provisions are, in essence, the means for allowing scheme members the commercial freedom to achieve an overall reduction of 4.5%, subject to the controls and restrictions necessary in this market and to the provision of information. Thus it follows that, if the PPRS is to be shown to be unlawful, there must be some feature of the PPRS that is actually or potentially discriminatory and which marks it out from ordinary market competition.

74. It is against that background that API's argument on "subsidy" must be examined. As the scheme members are allowed the market freedom to set their prices in a way that meets competition from scheme members and others, it is inevitable that price changes on products will differ; that is what happens in any market. A margin may be greater on one product than another because of ordinary market forces; in that sense a person buying a product with a higher margin can be said to be "subsidising" the person who buys the product with the lower margin. It seems to me that it is only in that sense that the Department is "subsidising" different prices; it is an ordinary aspect of a competitive market where the same person buys several products from the same supplier. If there was no PPRS, then a pharmaceutical company facing competition from a parallel importer on a product would, subject to legal constraints such as those against predatory pricing, cut his price against that product, but would seek to maintain the overall profitability of his business by trying to obtain a higher price on other products. The ability to seek to charge higher and lower prices to the purchaser is an ordinary feature of a market. If there was no PPRS, the suppliers of pharmaceutical products to the Department would be in exactly the same position to raise or lower prices.

75. I therefore reject the argument that there was any form of "subsidy" from the Department; the fact that there would be different margins on products supplied is an ordinary feature of free market competition which the scheme preserves. There is no potential discriminatory promotion of domestic over imported products in this way.

76. It was also argued by API that scheme members would always be free to reduce prices; that is so, but there is a limitation. Save where temporary price reductions are permissible under Chapter 19.7, a scheme member, without the modulation provisions, can only reduce prices in the knowledge he cannot increase them again, as he would in the ordinary market place to respond to the pricing of his competitors.

77. Thus it is necessary to examine API's second contention and to consider whether there are aspects of the market in pharmaceutical products or of the provisions of the PPRS that make the PPRS or the modulation provisions discriminatory.

78. It is necessary first to consider whether the provisions are, on the objective evidence before the court, targeted at or more likely to impact on parallel imports.

Were the modulation provisions targeted at or more likely to impact on parallel imports?

79. As set out in paragraph 36, a key part of API's case was establishing that the modulation provisions were, on their face, designed to target parallel imports; API contended that this followed from the fact that there was nothing else on which the modulation provisions could operate, as was clear from an examination of the scope of the scheme in respect of the other areas of competition. Alternatively they were more likely to impact more on parallel imports.

80. It was accepted by the Department and ABPI that, as set out at paragraph 11, parallel imports are the only competition for branded products still subject to patent and where there is no therapeutic alternative. Modulation can therefore clearly operate against parallel imports in respect of those products. But is that the only category against which modulation can be used, or can it be used to a material degree against either (i) generic products or (ii) branded products within the same therapeutic class?

(i) Competition from generic products

81. The Department has for sometime been encouraging prescriptions to be written in the generic name of the product; in the period 1984 to 1994, the share of generics increased from 18% to 42% by volume. It is continuing to encourage this by providing more information to general practitioners, by continuing education, and by the monitoring of performance of allocated budgets. The evidence before the court was that 60-70% in volume of all prescriptions written within the NHS are written by the generic name of the product, though (as I have set out at paragraph 21) 80% of the expenditure on pharmaceuticals (by value) is on branded products.

82. Where a prescription is written by the generic name, then the pharmacist is entitled to dispense either the branded product or any of the generics. Because of the way pharmacists are reimbursed, as described in paragraph 12, the pharmacist will wish to obtain the cheapest supply; there is therefore keen price competition between the branded product and its generic equivalent.

83. When a branded product loses its patent protection (or the protection of a supplementary protection certificate), so that it is open to competition from its generic equivalent, the manufacturer of the branded product will reduce its price; often this reduction is, as Professor Jones, the Director General of the ABPI, stated, dramatic to enable the owner of the brand to maintain an acceptable market share.

84. Because of this, chapters 21.3 and 21.7 of the PPRS specifically exclude the ability of scheme members to count, in calculations of modulations, price reductions made within a specified period of the expiry of the patent protection (or the supplementary protection certificate). As such a price reduction is an inevitable consequence of the operation of market forces at about the time of the expiry of the patent, if this exception had not been made, the Department would have left open a way to negate their aim of an effective overall price reduction. Furthermore the prohibition against modulation during this period is a complete prohibition; it operates against all competition, including parallel imports and branded products within the same therapeutic class and not merely generics.

85. However, the exception provided for in chapters 21.3 and 21.7 does not mean that price reductions against generics outside the specified period cannot be brought into account in modulation. For example, in the case of a branded product where the patent expired in the early 1990s, any price reduction in respect of that branded product can be brought into account.

86. API asserted that the preponderance of the price reductions that could and would be made were reductions excluded by the exception. Although in theory reductions could made against generics outside the specified period, the periods specified in chapter 21.3 and 21.7 were designed to exclude modulations and would do so in practice, as the evidence of the ABPI was that reductions would generally be made in the specified period.

87. However, there was no evidence which supported that assertion. Although it was clear on the evidence that a dramatic price reduction would usually be made at the time the patent protection expired, a branded product could still command a premium over its generic product, as a premium is inherent in the value of a brand. Thus although the price would move sharply downwards at around the time the patent protection was lost and thus within the specified periods in chapter 21.3 and 21.7, there could remain a relatively large price gap between the branded product and its generic competitor; there was therefore scope for further price cuts under the modulation provisions. There was some evidence from Mr Brownlee that there were price cuts outside the periods specified in Chapter 21.3 and 21.7 where a branded product had lost market share to generics or where the manufacturer gradually reduced prices over a period of three years before patent expiry to lessen the impact expected when the patent expired. Similar general evidence was given on behalf of ABPI, through Mr Bailey, President of the ABPI and corporate affairs director of Glaxo Wellcome and Ms Charlesworth of ABPI; each gave two specific examples of branded products where the price had been reduced to meet competition from generics - Beconase and Zantac in the case of Mr Bailey and Voltarol and Planquenil in the case of Ms Charlesworth. There was some other specific evidence to this effect. Thus on the evidence some price reductions are in practice made outside the periods specified in chapters 21.3 and chapter 21.7.

88. In my view, on the evidence before the court there is, in practice, material scope for the modulation provisions to operate against generic competition in the circumstances I have specified. Although the Department put forward some evidence in relation the number of products where there was competition between branded products and generics and as to the number of products where the patent expired during the first phase of the PPRS, it was not possible on the evidence to determine the extent of the scope for modulation to be used against generics.

89. I cannot go beyond that because of the state of the evidence before the court. That evidence was put before the court by reference to the issues the parties thought that they would be dealing with at the hearing, based on the nature of the case of which they had been given proper notice. I consider that the arguments made by the Department and the ABPI set out at paragraph 37 are essentially correct. A full factual enquiry would be necessary if I were to draw the inference API seek to draw as to the scope for modulation to be used against generics. This aspect of API's case was not foreshadowed in Form 86A; that formal defect may not have mattered provided it was otherwise notified to the Department and ABPI. It was not, but it should have been. It would therefore not be fair to draw inferences from the absence of detailed evidence from the Department or ABPI on these issues or to conclude that the examples given by ABPI are no more than isolated instances. On the present evidence, I cannot therefore go further than I have; I cannot assess the extent to which the operation of the modulation provisions impact upon competition from generic products There is material scope for modulation against such competition, but beyond that I cannot go. I cannot assess its extent.

(ii) Competition from branded products within the same therapeutic class

90. There is no doubt, for the reasons set out in paragraph 6, that competition between branded products within the same therapeutic class is limited. API contended that a rationale of the PPRS and its predecessor schemes was that the ordinary forces of competition between branded products within the same therapeutic class is not significant enough to guarantee that prices are fair and reasonable and therefore a scheme is necessary. They submitted that it therefore followed that, as price competition between branded products is by its very nature limited, then the inherent scope for modulation within the PPRS by one branded supplier against another must also be limited and of no material significance. It is necessary to consider first the argument based on the rationale for the PPRS.

91. In my view, it is necessary to emphasise the nature of the distinction drawn between (1) branded products protected by patents where there is no therapeutic alternative and (2) branded products where there is an alternative within the same therapeutic class. In the former case, there is no competing brand (though there may be competition between the parallel importer of the branded product and the manufacturer or appointed distributor of that product) whereas in the latter there is, limited though it is.

92. I do not accept the API's argument that the rationale of the PPRS is to be found in the lack of inter-brand competition of products within the same therapeutic class during the period of patent protection. The position is more complex. The need for the PPRS arises also because there are branded products where there is no therapeutic alternative during the period of patent protection of such products. As the Second Report to Parliament on the PPRS (December 1997) observed at paragraph 1.2

"While patents are granted for medicines, Government must nevertheless ensure that prices to the NHS remain fair and reasonable."

Another reason for the PPRS is that the decisions to acquire products within the NHS are not made, as explained at paragraph 5, on ordinary market principles. Therefore although I accept that the limited inter-brand competition between products within the same therapeutic class may well be a reason why the PPRS was established, it is only one reason in this complex market. It is thus not the rationale of the PPRS and does not demonstrate that competition between products within the same therapeutic class is as limited as API contend. It is therefore necessary to consider API's argument on the competition between branded products within the same therapeutic class.

93. As I have already stated, the competition between branded products within the same therapeutic class is limited for the reasons to which I have briefly referred at paragraph 6. These reasons also appear from the Department's Report to Parliament on the 1993 PPRS and can be summarised:

* Doctors prescribe on the basis of the needs of a patient; the cost is met by the Department; although the Department has attempted to bring cost into the decision making, it is clear that price sensitivity is very much more limited that in an ordinary market.

* Branded products may not be interchangeable for certain patients because of the patient's tolerance to the product, the side effects and the risk of side effects; this is recognised by the Commission in its decision on the Ciba-Geigy/Sandoz merger (Decision 97/469/EC; OJ 1997 L201/1). Such products are therefore not competitors in absolute terms.

94. The question is whether the extent of that limited competition between branded products within the same therapeutic class is insignificant or immaterial. The evidence of Mr Bailey (of Glaxo Wellcome) in his witness statement made in December 1999 is that price changes which are permitted to take place as a result of modulations are much more likely to be attractive to companies in competing against branded products within the same therapeutic class; this is because the price differences between these products are likely to be smaller than between branded products and parallel imports. He did not elaborate or support that statement by specific instances. Mr Elford in his second witness statement made in August 2000 answered this statement by setting out his analysis of the price modulations within the top seven therapeutic categories which accounted for 35.7% of the total drugs bill in the UK:

* Acid pump inhibitors

* Calcium antagonists

* Ace inhibitors

* HMG-COA reductase inhibitors

* SSRIs (anti-depressants)

* B2- stimulants inhalants

* Corticoids inhalants

His analysis concluded that in most cases price reductions did not alter the competitive position of the products within the same therapeutic class; that the majority of price reductions did have a detrimental effect on parallel imports and price reductions seem to have been targeted at parallel imports.

95. There was a very detailed response to this by Professor Jones, in addition to his general evidence that there was strong competition between branded products within the same therapeutic class. He set out an explanation of the reasons why price changes had been made on products within these seven categories. It is quite clear from his account that the reasons why there were modulations were complex. For example within the first category of products listed by Mr Elford - acid pump inhibitors - there were at least five products - Losec 10mg, Losec MUPS 10mg, Pariet 10mg, Protium 40mg and Zoton 15mg. Although these were within the same therapeutic class, it was not clear what were the limitations on the interchangeablility; assuming that they were interchangeable, the evidence on these five products was:

* Losec: Mr Elford contended that the price of Losec had only been reduced by 5% and therefore had not been reduced to compete with Pariet and Protium. The explanation for the price reduction given by Professor Jones was that, as Losec is the biggest prescription product in the UK, a small price reduction on this product beyond 4.5% would go a long way to meeting the overall requirement of a 4.5% price reduction.

* Zoton 15mg The evidence given by Ms Charlesworth was that the price was reduced by 25% on 1 July 1999, before PPRS came into effect and was not made under the modulation provisions. Mr Elford accepted in his second witness statement that the price was reduced to make it more competitive.

* Pariet 10mg: Mr Elford's evidence was that Pariet was lower in price and competitive; it did not need to be reduced to compete with the other branded products; nonetheless it was reduced by 14.38% and the effect was to make parallel importing unattractive. Professor Jones' evidence was that the manufacturer had reduced the price to compete with Zoton 15mg and made the 14.38% reduction to meet the overall requirement of the 4.5% price reduction. There was then a further reduction in the price of Pariet 10mg, when the price of Zoton 15mg was reduced

* Protium 40mg: Mr Elford's evidence was that the price of protium 40mg was reduced by 10.95% under the modulation provisions; this made parallel importing less attractive, but allowed it to compete more with Losec 20mg (which was its therapeutic alternative); Mr Elford claimed that the manufacturer did not need to reduce the price by 10.95% to compete with Losec 20mg but it did need to do this to compete with parallel imports. Professor Jones' evidence was that the manufacturer had made the price reduction to compete with the other proton pump inhibitors, not just Losec 20mg.

In a further witness statement served on 24 November 2000, Mr Elford provided further evidence in relation to these and other products. However it is not necessary for me to set out his evidence, as it is clear that there is a highly complex factual enquiry with disputed evidence which cannot easily be resolved on the written evidence before the court. It is, in my view, not necessary to attempt such a resolution, as it is clear from these witness statements that, although there is competition between branded products within the same therapeutic class and competition with parallel imports, an assessment of the extent of the competition is simply not possible on the evidence before the court.

96. API argued that, on the basis of the evidence before the court, inter-brand competition between products within the same therapeutic class is immaterial; I have already concluded that I cannot accept API's contention that this conclusion can be derived from the rationale of the scheme. Nor can I draw any inferences as to the minimal scope of the competition, as I was requested by API, from the fact that, although the material evidence would be within the knowledge of the ABPI and its members, they had not adduced the necessary evidence. I do not consider that that is an inference that I should draw as the ABPI did not receive fair notice of the significance of the argument made by API until the opening of the hearing, as I have set out at paragraph 36. There may or may not be significant competition; on the evidence I cannot draw any conclusion and for the reasons given, I am not prepared to draw an adverse inference against ABPI.

97. Moreover, there are other reasons, on the evidence, why the scheme members might wish to modulate their prices. These include the wish to test the price level which the market for a particular product will bear and variation of prices to manage a drug portfolio; in the latter example, a company may vary prices of different drugs it produces to induce purchasers to move from older to newer products.

98. Thus on the evidence before me I cannot conclude from the structure of the scheme that the PPRS was targeted at parallel imports; there was scope for modulation under the PPRS to operate on other forms of competition which, on the evidence available to the Court, I am satisfied existed to a material extent. For similar reasons, I cannot conclude that the PPRS and the modulation provisions were more likely to impact on parallel imports

Subjective intention

99. There was one further argument that was advanced by API as to the discriminatory nature of the PPRS and its targeting on parallel imports.

100. API contended that the clear intention to target parallel imports could be inferred from the speech of Mr Frank Dobson, the Secretary of State at the time during which the PPRS was negotiated and agreed between the Department and the ABPI; from that speech, it was clear that the expectation of the Department and the intention of the ABPI was that it would be used against parallel imports. I have referred to the report of his speech at paragraph 33; the accuracy of that report was covered in the evidence before the court.

101. A speech had been prepared for Mr Dobson to make which stated:

"5....We want a system that delivers prices for medicines that reflect good value for money for the NHS as purchaser and prices which represent a reasonable return for the developer and manufacturer. I said last year, and repeat it, that the prices we pay have to give you the wherewithal to invest in further R&D. It is the need to ensure that you get a reasonable return that has led me to agree if we can make progress on parallel trade within the PPRS negotiations. Any action we do take will need to be legal practical and proportionate."

102. The policy of the Department was to try and take account of the ABPI's view that the parallel trade had to be addressed within the negotiations. However Mr Frank Dobson in his speech, according to the evidence of Mr Barnes who was present, did use words to the effect that the Department would use the PPRS to reduce the attraction of parallel imports; he did not say which features of the PPRS he had in mind or refer to price modulations. Mr Barnes' evidence was that, to the extent that Mr Dobson implied that the negotiations on the PPRS were concerned with devising new measures for the specific purpose of dealing with parallel trade, his comments did not reflect the reality of the negotiations; the Department's view was that the PPRS was a mechanism for controlling the profits of the pharmaceutical companies and only that.

103. As I set out at paragraph 148 below, the intention of the Department is best inferred from the evidence of the officials who had conduct of the negotiations and not from an unscripted departure by the Secretary of State from his prepared speech. He was not familiar with the negotiations and there is no reason to doubt the clear evidence of the officials who conducted the negotiations. I therefore cannot conclude that it was the subjective intention of the Department that the modulation provisions were designed to target parallel imports.

Conclusion

104. As set out in paragraph 61, API accepted that a 4.5% across the board price cut would not have infringed Article 28, though it was inevitable that such a reduction would have an effect on imports, both imports by manufacturers and by parallel importers. It followed that API had to point to some actual or potential discriminatory hindrance or restriction on imports arising from the modulation provisions if there was to be an infringement of Article 28. It seems to me that if the PPRS had merely imposed a 4.5% price cut and allowed the pharmaceutical companies complete commercial freedom to achieve this in the way they wished, then it is difficult to see how, in the absence of evidence that this might potentially discriminate against imports, such a provision would have been contrary to Article 28. API thus had to show that there was some potential effect of the modulation provisions that would hinder imports in a discriminatory way. The difficulty that API faced was that they advanced this case without the evidential basis to support it, for the reasons I have given. API maintained that, as they only had to show that there was a capability to hinder or restrict imports, they did not have to show the quantitative effect. However their argument proceeded on the basis that the PPRS could in effect only operate against parallel imports, as there was no scope for it to operate in a material way against any other source of competition. For the reasons, I have given that particular argument fails; they had to show that there was no substantial scope, so that the inference could be drawn that the only real scope for the use of the modulation provisions was against parallel imports and thus establish it was a measure having a potential effect on imports in a manner contrary to Article 28. Again on the evidence they have failed.

105. I therefore conclude that, on the evidence before the court, the modulation provisions were not a restriction on or hindrance to imports within the scope of Article 28 and were not capable of being such a restriction or hindrance.

106. I should emphasise that this is a conclusion which I have reached on the evidence before the court. I accept that the argument advanced by API is one of substance on this issue. I also accept that the evidence as to the operation of the PPRS is not within their knowledge. As the Transparency Directive recognises, there is an important public interest that detailed information about the operation of the market in branded pharmaceutical products is made available and published on a regular basis; this is particularly true of the operation of the PPRS so that its operation can be the subject of detailed scrutiny to ensure that it is not having a potential discriminatory effect on parallel imports.

(iii) Have API shown that there has been an actual effect under Article 28 on imports?

107. API submitted that:

* it was the effect on trade that was relevant not on competition; it was measured in terms of profitability and turnover.

* that it was sufficient if there was a potential effect.

* it was enough that the effect was slight; there was no de minimis rule as Van der Haar made clear.

* the effect should be determined by an analysis of the measure - the modulation provisions of the PPRS.

108. API relied on a number of matters; they pointed to the evidence of Mr Brownlee that the prices of 14% of the top 50 medicines for which there were parallel import licences had been reduced by 4.51-10% and 16% of the top 50 by more than 10%.; and that in 20% of the cases where there had been price reductions, the Department was unable to isolate a reason other than competition with parallel imports as the motivating factor. They pointed to the evidence of Ms Charlesworth of the ABPI to the effect that 26.5% of products facing parallel import competition had price reductions of over 4.5%. API submitted that it could not really be disputed that modulated price reductions had been made against parallel imports.

109. Indeed as Professor Jones stated in his third witness statement:

"The ABPI accepts, without any embarrassment at all that, since parallel imports are part of the competitive environment within which the pharmaceutical industry in the United Kingdom operates, modulations have taken place to compete with parallel imports. "

However they did not accept that the majority of modulations were applied to compete with parallel imports.

110. Thus, in so far as this issue is concerned with the question as to whether the PPRS had some effect on parallel imports, the answer is clear that it has had some effect; plainly some price modulations must have been made to enable products to compete with parallel imports; in consequence the parallel importer has had to reduce his price and/or has suffered a decline in the volume of business in respect of these products. But that, it seems to me, is inevitable from the form of competition the PPRS permits and is self evident. That may well be sufficient to answer the question. Some of the modulated price reductions under the PPRS have had an actual effect on parallel imports.

111. However, if the question is whether the effect has been discriminatory or otherwise infringed Article 28, then it is necessary to examine the evidence in more detail.

112. It was the evidence of Mr Elford that it was no longer viable for Dowelhurst to import certain products because of the drop in the UK prices and identified 21 such products.

113. API also produced a detailed schedule abstracting the evidence before the court in relation to 87 products where Dowelhurst claimed that there had been modulation in relation to their products of more than 5% in the period October 1999 to January 2000. In the relevant period there had been 2634 modulations, 364 of them over 5%. The schedule was produced during the course of the hearing; after the conclusion of the hearing I received written comments on it from ABPI and a response from API.

114. From that schedule API advanced three propositions:

* In 30 out of 87 of Dowelhurst's products where the modulation had been more than 5%, the evidence indicated that the sole reason for modulation was to combat parallel imports; the specific target of the price cut in such cases was the parallel import.

* Regardless of the reason for modulation, modulation of 5% and over had had a significant effect on Dowelhurst's products.

* Modulation allegedly targeted at branded competitors has had a significant effect on parallel imports.

115. There are three specific matters that must be borne in mind when examining the schedule:

* It covered only the products imported by Dowelhurst or in respect of which they had applied for an import licence; it did not cover other parallel importers. Mr Elford of Dowelhurst stated in his second witness statement that he believed their product range was representative of parallel importers; there were statements from other parallel importers - Waymade Health Care Limited, Doncaster Pharmaceuticals Limited, Freemans Pharmaceuticals Limited, Munro Wholesale Medical Supplies Limited and Europharm of Worthing Limited to the effect that each of them had a similar product profile and the effects had been similar. Although there was no statistical or comparative evidence, on the evidence of these companies, I accept that the product range of Dowelhurst was representative of parallel importers.

* The schedule summarised the evidence available to the court about the reasons for modulation in respect of the products; but there was no direct evidence in many cases from the particular pharmaceutical company that made the decision. Some of the information was contained in confidential exhibits which the manufacturers of the products concerned had not seen, though the ABPI had.

* The schedule did not cover other products in the market where there had been a reduction in price under the modulation provisions. For the purpose of showing that price reductions had had an effect on parallel imports, this was a permissible approach; but, as I have said, that is not disputed, as some modulations must have taken place for the purposes of meeting competition from parallel imports. However, if the question is whether the effect has in practice been discriminatory, the schedule did not assist as it would be necessary to look at least at a representative sample of the entirety of the modulations made.

116. As to the first of the three propositions advanced by API, they sought to draw the inference that price modulations had been targeted at parallel imports from the fact that ABPI had not provided evidence of reasons in relation to many of the products in respect of which it is alleged that the object of modulation was to target parallel imports. In none of the original witness statements served was any allegation made as to the reason for the modulations; the evidence served related solely to the effect of modulations. Specific allegations of targeting were made in Mr Elford's second witness statement served in August 2000 in relation to branded products within the same therapeutic class and in relation to generics; these were answered by ABPI.

117. API contend that the schedule set out some cases where they could not identify any competitor branded product within the same therapeutic class or any generic competitor; these were therefore parallel imports where there was only a branded product competitor protected by patent; they identified the products as De-nol, Duphaston 10mg (with its concept generic Dydrogesterone 10mg), Cacit 500mg, Pergolide (in 250mcg, 1000mcg and 50mcg strength),Zovirax (in 200mg and 800mg strength), Stugeron forte, Ikorel (in 10 and 20 mg strengths with its concept generic Nicorandil in similar strengths), Prepulsid 10mg (in two different presentations and with its concept generic Cisapride), Salazopyrin, Zirtek ( and Zyrtek which is said to be an overseas spelling of Zirtek), Gonal (in three different strengths or presentations), Ventide, Salbutamol (Ventolin), Zyprexa (in three different strengths) Risperdal and Motival. In respect of these products, they contended that ABPI had not challenged their explanation.

118. I have set out at paragraphs 94 to 96 some of the evidence relating to branded products within the same therapeutic class and set out my view that the position is more complex and that I cannot conclude the extent to which there has been targeting to make the measure discriminatory. As to generics, the position is also complex. Mr Elford referred in his second witness statement to six branded products where there is a generic equivalent where the price had been reduced by more than 4.5% - Imodium, Maxolon, Zantac, Serc, Nitrolingual and Voltarol; he contended that the reductions had not enabled the owner of the brand to compete wtih the manufacturer of the generic equivalent. The schedule contains references to three different presentations or strengths of Voltarene (which it is said is known as Voltarol) and two different strengths of Zantac. The evidence put forward by Professor Jones was that in the case of Zantac, the level of generic prescribing was 90%; the price reduction enabled the manufacturer to compete with the generic equivalent and parallel imports were not an issue. In the case of Serc, there were two strengths - 8mg and 16mg. The manufacturers had modulated against Serc 16mg where there was competition from two generic manufacturers and very little parallel importing; they had not modulated in respect of Serc 8mg where there was substantial parallel importing. As to Imodium, the evidence given by Professor Jones was that the parallel imports were not a factor in the decision to reduce the price, as the level of parallel imports are small. In relation to Voltarol, Ms Charlesworth's evidence was that the manufacturers thought that the price reduction would help them compete with generic competition. There was no evidence before the court in relation to the reasons for modulation in respect of Maxolon and Nitrolingual as the manufacturers were not members of the ABPI and not parties to the proceedings. On the basis of the evidence in relation to generics, again I cannot conclude that there have been price reductions made to compete with parallel imports to an extent greater than would be the case in ordinary market competition and such as to make the modulation provisions discriminatory or otherwise contrary to Article 28.

119. This therefore leaves the products which I have identified in paragraph 117 where API contend that no explanation has been advanced by ABPI to rebut their contention. However it seems to me that there was nothing to put ABPI on notice that API were going to make a case on the reasons for the price reductions on these products as opposed to a case on the effect of the price reductions on these products. As I understand ABPI's position, they provided evidence in answer to the case made by API on the reasons for modulated price reductions on branded products within the same therapeutic class and generics (to which I have referred in the two preceding paragraphs), but did not appreciate a case was being advanced on the reasons for the reductions in respect of the products identified in paragraph 117. They therefore submit that no inference should be drawn. I have very considerable sympathy for their position and accept that they did not appreciate that this case was being made until the hearing. As I have mentioned, the schedule was produced in the course of the hearing and ABPI's detailed response was delivered after the conclusion of the hearing. A complex factual enquiry into the reasons for price reductions in relation to a number of different products should have proceeded by the applicants setting out their case in a schedule and the respondents then answering it; evidence should then have been being deployed to support the contentions so identified. That did not happen in this case. It would not be fair to ABPI to draw an inference against them in such circumstances. Plainly, as I have set out in paragraph 110, price modulations have been made to meet competition from parallel imports, but I cannot, on the present state of the evidence, draw any conclusion as to the scale of such price reductions so as to draw the further conclusion on discrimination.

120. As to the second of API's propostions, the schedule sets out the effect on the volume of products sold by Dowelhurst where there had been price modulations greater than 5%; this showed that in respect of the 87 products, there had been a decrease in the volumes of 51 products sold after modulation and price reductions in the case of 69 such products. ABPI contended that the evidence on which Dowelhurst relied, as set out in the confidential exhibits to Mr Elford's second witness statement, was unreliable for a number of reasons set out in Professor Jones' third witness statement.

121. In assessing the evidence, it is first important to note that sale volumes and price fluctuations in this market are determined by a number of factors including the prescribing habits of doctors, recommendations, marketing campaigns, competition and the availability of new products. Parallel imports will be affected not only by these factors and price changes in the products with which they compete in the UK but also by the prevailing price in the state where the product is purchased by the parallel importer and the availability of the supply. Before it would be safe to conclude that modulated price cuts had caused the precise changes identified by API in the schedule, it would be necessary to examine the circumstances surrounding each product. Thus it seems to me that, although it would follow from the fact that there have been modulated price reductions which must have adversely affected the volumes and/or prices of the parallel imports made by Dowelhurst (as I stated at paragraph 110), it is not possible without a much more detailed enquiry to determine the extent. Furthermore the enquiry would have to examine other detailed facts; for example, Zoton 15 mg, to which I referred at paragraph 95, appears in the schedule as a product where the price and the volume decreased; it is a product which API say cannot viably be imported. On the evidence of Ms Charlesworth, the price cut occurred before the PPRS was agreed; the concept generic of Zoton (Lansoprazole) also appears on the schedule as a product where the volume and price have fallen; concept generics are branded products imported into the UK and re-boxed by the parallel importer in its "house style" and sold under the generic name. In respect of this concept generic, there is not only the evidence that the price reduction was made before the PPRS was agreed but also the consideration that as it is a generic, it can only be supplied when prescribed as a generic and therefore the way in which it competes with Zoton is different to the way in which the parallel import of Zoton competes with Zoton. In the case of Losec, a branded product within the same therapeutic class, the volume of sales increased, but the sale price decreased. All of these matters would need examination to see whether the effect was discriminatory or otherwise contrary to Article 28.

122. The third of API's propositions was directed at Mr Bailey's evidence (to which I referred at paragraph 94) to the effect that price changes were more likely to be attractive to companies competing with branded products within the same therapeutic class. API said that of the 23 products on the schedule where modulation was said to have been carried out to compete with products within the same therapeutic class, modulation had caused a reduction in the volume of parallel imported products sold or in the price of such products; this suggested that price cuts necessary to compete with other branded products also had a significant effect on competition from parallel imports. It seems to me that such a conclusion would probably be expected, irrespective of the evidence. It was accepted by ABPI in argument that both sets of competitors would be affected. If the manufacturer reduces the price of his branded product to compete within another product within the same therapeutic class, the parallel importer of the product in respect of which the price reduction is made will have to adjust his price to maintain his competitive position; if the manufacturer of the other product within the same therapeutic class reduces his price, then that also may affect the parallel importer. Thus, although it seems to me clear that the effect for which API contend is self evident, it is difficult to understand how it can be contended that this effect, on its own, is discriminatory or otherwise infringes Article 28, unless there is some other feature of the PPRS that is discriminatory. In their reply submissions on the schedule, API seem to accept this and pointed to a discriminatory feature being the state subsidy or government sponsored price cut, a submission which I have addressed at paragraphs 70 to 75. The fact that a modulated price cut directed at a branded competitor within the same therapeutic class will also have an effect on a parallel import is an ordinary consequence of competition between brands within the same therapeutic class.

123. Since 1 January 2001, the members of the PPRS have been able to modulate prices to a greater extent, if they take advantage of the ability to make greater price increases. It would seem clear, as API submit, that the potential effect on parallel imports from this will be greater than the effect during the interim period; as there will be an ability to charge higher prices on some products, greater price reductions can be made on others; the effects of such reductions will be on all forms of competition, including parallel imports. Although the effect may therefore be potentially greater, the issues of principle remain the same and it will not be possible to say whether the operation of these provisions has had any discriminatory effect until evidence is available as to the effect of their operation. As I observed at paragraph 106, it will be important for information to be made available and published and the operation of the PPRS scrutinised.

(iv) If so, do the modulation provisions as a selling arrangement within paragraph 16 of Keck, apply

(i) to "all relevant traders"?

(ii) without discrimination between domestic goods and imports?

The decision of the European Court of Justice in Keck

124. In Keck and Mithouard [1993] ECR I- 6097, the European Court of Justice re-considered its case law on Article 28 in the context of re-sales of products at a loss, "loss leaders", contrary to a French law prohibiting sales at a price lower than the purchase price of the goods. The Court referred at paragraph 11 to the consistent case law that any measure that was capable of directly or indirectly, actually or potentially, hindering intra-Community trade constituted a measure having equivalent effect to a quantitative restriction. The core of the Court's judgment is at paragraphs 13-17:

15. It is established by the case-law beginning with Cassis de Dijon (Case 120/78 Rewe-Zentral v Bundesmonopolverwaltung fûr Branntwein [1979] ECR 649) that, in the absence of harmonization of legislation, obstacles to the free movement of goods which are the consequence of applying, to goods coming from other Member States where they are lawfully manufactured and marketed, rules that lay down requirements to be met by such goods (such as those relating to designation, form, size, weight, composition, presentation, labelling, packaging) constitute measures of equivalent effect prohibited by Article [28]. This is so even if those rules apply without distinction to all products unless their application can be justified by a public-interest objective taking precedence over the free movement of goods.

16. By contrast, contrary to what has previously been decided, the application to products from other Member States of national provisions restricting or prohibiting certain selling arrangements is not such as to hinder directly or indirectly, actually or potentially, trade between Member States within the meaning of the Dassonville judgment (Case 8/74 [1974] ECR 837), so long as those provisions apply to all relevant traders operating within the national territory and so long as they affect in the same manner, in law and in fact, the marketing of domestic products and of those from other Member States.

17. Provided that those conditions are fulfilled, the application of such rules to the sale of products from another Member State meeting the requirements laid down by that State is not by nature such as to prevent their access to the market or to impede access any more than it impedes the access of domestic products. Such rules therefore fall outside the scope of Article [28] of the Treaty."

125. In Schutzverbrand v TK-Heimdienst (13 January 2000), the European Court of Justice considered the application of the principles in Keck to an Austrian law requiring bakers, butchers and grocers who wished to sell their wares from vans in the Tyrol to have a fixed place of business in the administrative district of the Tyrol or an adjoining district. The Court held that this provision did not affect in the same manner the marketing of domestic products and products from other member states, as goods from other member states had to bear the extra costs of a fixed establishment, even though the same extra costs would have to be borne by nationals in other parts of Austria wishing to sell from vans in the Tyrol; the measure was discriminatory. The Court said at paragraph 26-27:

"26. .. Consequently for goods from other Member States to enjoy the same access to the market of the Member State of importation as domestic goods, they have to bear additional costs.

27. That conclusion is not affected by the fact that, in each part of the national territory, the legislation affects both the sale of products from other parts of the national territory and the sale of products from other Member States... For a national measure to be categorised as discriminatory for the purposes of the rules on the free movement of goods, it is not necessary for it to have the effect of favouring national products as a whole or of placing only imported products at a disadvantage and not national products."

The questions on this issue

126. It was common ground that, if the price reduction and modulation provisions of the PPRS were within Article 28 and API had been successful on the earlier issues, those provisions of the PPRS were "selling arrangements" of the type referred to by the Court in paragraph 16 of its judgment in Keck. It is therefore not necessary to examine the arguments relating to the application of the principles in Keck.

127. The questions which therefore arose were whether:

(i) the provisions applied to all relevant traders; and

(ii) they affected in the same manner the marketing of domestic products and those imported from other Member States.

ABPI drew my attention to a passage in Introduction to the Law of the European Communities by Kapteyn, Verloren van Themaat and Gormley (3^rd edit) at p 635 which, after referring to the two conditions, states:

" The first of these seems to recall the statement in Dassonville that the means of proof had to be accessible to all Community nationals, but it may be wondered what the added value of the statement is. As to equal applicability in law and in fact, this simply reflects one of the criteria already applied in relation to the rule of reason. Again this represents nothing new."

The submission was made by ABPI that the first condition was inserted into this case concerning the free movement of goods to ensure Member States do not discriminate on the grounds of nationality. As there is no question on the facts of this case of discrimination on the grounds of nationality because of the nature of the market in pharmaceuticals, this submission seems to me to have very considerable force. But it is not necessary for me to decide whether the first condition is so limited as on the facts of this case, the question can also be answered in the same way on a broader reading of the first condition.

(i) All relevant traders

146. The Department contended that the test for determining the object was an objective test determined by looking at the provisions of the PPRS to see what the object of the agreement was; they submitted that the subjective intention of one of the parties was immaterial. They relied on the decisions of the Court in Miller International v Commission [1978] ECR 131 and NV IAZ Belgium v Commission [1983] ECR 3369 which made it clear that it was not the subjective intention of a party that was relevant, but the object determined by an analysis of the agreement in its legal and economic context and the conduct of the parties.

147. Although I accept that the statement by one party of that party's intention cannot be relevant in determining the object of an agreement, I do not see why a statement by one of the parties cannot be evidence from which the common aim or object of the agreement and the parties to it can be inferred. Such a statement would simply be part of the evidence and it would be necessary to establish that the statement was a statement of the common object and intention.

148. The remarks by the former Secretary of State, Mr Dobson, in the after dinner speech in April 1999, to which I have referred at paragraphs 33 and 101, may therefore be taken into account, if it is evidence of the common objective intention. In my view, on the evidence, I do not, however, consider that it is evidence of the common object. In the first place, the agreement was negotiated by Mr Brownlee and Mr Barnes. Their evidence as to the common object of the PPRS is quite different from the reported remarks of Mr Dobson. In my view, their evidence is of much greater weight as to the common intention. They had a command of the detail and the understanding which Mr Dobson did not; he did not participate in the negotiations. He was briefed as to the position and provided with a draft speech (set out at paragraph 101). From the report of his remarks (set out at paragraph 33), it is clear he departed from it. I can only conclude that what he said was not what those at the Department understood from their dealings with the ABPI was the common object. I therefore do not consider it reliable evidence of the common intention. I consider the evidence of Mr Barnes (set out at paragraph 102) more reliable.

149. It was also submitted by the Department that, if any view of a Secretary of State was relevant, it was the views of the current Secretary of State that were relevant. I do not accept that submission. If a view was relevant, it was the view of the Secretary of State at the time the agreement was made; it would be his views, if they were evidence of the common intention, that would assist the Court and not the views of the current Secretary of State who could not give direct evidence of the common intention at the time that the agreement was made.

150. For the reasons I have given in relation to Article 28 I do not regard the modulation provisions of the PPRS as having as their object the distortion or restriction of competition. Their overall object was to promote competition within the constraints inherent in the market for the supply of branded pharmaceutical products to the NHS.

(iii) Do the modulation provisions have as their effect a restriction or distortion of competition?

151. API contended, that, if they had not established that the object of the provisions was the restriction or distortion of competition, then the provisions had that effect. Although the modulation provisions had had, they submitted, an actual effect, they did not have to establish that; it was sufficient that they had a potential effect. A potential effect existed where it was predictable to a reasonable degree of probability that the provisions would exert an impact upon competition. API relied upon the same arguments as they had advanced in respect of Article 28 that the modulation provided a form of state subsidy to be used against parallel importers and that their potential effect would be greater after 1 January 2001.

152. However, for the reasons I have given in respect of Article 28, these contentions also fail. I do not consider, on the present evidence, that API have shown, for the reasons I have given, that the modulation provisions have had the effect of distorting or restricting competition.

153. API then contended that it was sufficient that the effect was indirect; if contrary to their primary case price reductions under the modulation provisions were not aimed at competition from parallel imports but at competition from other products, it was sufficient that they had a potential effect on competition from parallel imports.

154. As I have set out at paragraph 122, it is self evident that a price reduction of a branded product to compete with other products within the same therapeutic class is bound to effect parallel imports of the product, but that is not a distortion or restriction of competition, but an effect of ordinary competition. I therefore consider that this alternative argument also fails.

(iv) Is there an effect on trade from the PPRS as a whole which is more than de minimis?

155. I can deal with this issue briefly in view of the conclusions to which I have come on the earlier issues. It was common ground that for the purposes of Article 81 API had to establish that the PPRS as a whole had an effect on the trade between member states; that effect had to be real and it was not enough that it was de minimis. The test set out in the case law is (see Societe Technique Miniere v Maschinenbau Ulm [1966] ECR 235 at 249) :

"It must be possible to foresee with a sufficient degree of probability on the basis of a set of objectives factors of law or of fact that the agreement in question may have an influence, direct or indirect, actual or potential on the pattern of trade between member states."

156. In Windsurfing International v Commission [1986] ECR 611, the European Court of Justice made clear at paragraph 96:

"Article [81.1] of the Treaty does not require that each individual clause in an agreement be capable of affecting intra-Community trade. Community law on competition applies to agreements between undertakings which may affect trade between Member States; only if the agreement as a whole is capable of affecting trade is it necessary to examine which are the clauses of the agreement which have as their object or effect a restriction or distortion of competition. "

157. API submitted that, as (1) the PPRS covered the supply of branded pharmaceuticals to the NHS, (2) suppliers were domiciled in most EU states and (3) supplies came from EU states and (4) there was an active trade in parallel imports, the PPRS affected trade between the Member States. They relied principally on three cases. First, in Verband der Sachversicherer v Commission [1987] ECR 405, where the European Court of Justice was considering a pricing arrangement between domestic German fire insurers, the argument was made by the fire insurers that the arrangements were internal to Germany and had no effect on trade between Member States. The Court, on the facts of that case, was prepared to find that the arrangement did, because foreign insurers with branches in Germany were affected by the agreement and because the agreement made access to the German market more difficult. In Commission v Italy [1998] ECR 1-3851 the European Court of Justice, when considering an agreement relating to the tariff charges of customs agents applicable in Italy which set minimum and maximum rates and laid down scale charges, said at paragraphs 48 and 49:

"As regards the question whether intra-community trade is affected, it need merely be pointed out that an agreement extending over the whole of the territory of a Member State has, by its very nature, the effect of reinforcing the compartmentalisation of markets on a national basis, thereby holding up the economic interpenetration which the Treaty is designed to bring about..

The effect is all the more appreciable in this case because the various types of import and export operations, within the Community, as well as transactions between Community traders, require customs formalities to be carried out and may, in consequence, make it necessary for an independent registered customs agent to be involved..."

The Court made clear that Member States must not introduce measures that might render inapplicable competition rules applicable to undertakings and said at paragraph 54:

"Such would be the case if a Member State were to require or favour the adoption of agreements, decisions or concerted practices contrary to Article [81] or to reinforce their effects, or to deprive its own rules of the character of legislation by delegating to private economic operators responsibility for taking decisions affecting the economic sphere."

They also relied on Ets. Consten SARL and Grundig GmbH v Commission [1966] ECR 299, an early case which decided that an agreement between a manufacturer and his distributor in another state which tended to restore national divisions in trade between Member states would be contrary to Article 81.

158. It is clear that the organisation and scope of the pharmaceutical trade is such that the PPRS, relating, as it does, to the supply of branded pharmaceutical products to the NHS, is capable of affecting trade between Member States; it would also seem clear that the PPRS as a whole must have had an effect on trade between Member States which was more than de minimis.

(v) Should API be prevented from challenging the modulation provisions in judicial review on the grounds that there are more appropriate alternative remedies, in a complaint to the European Commission or a civil action?

159. The Department and the ABPI submitted that it was not in the circumstances of this case appropriate to raise an Article 81 challenge by way of judicial review. They relied on a number of matters:

* The relevant parties were not before the court; it would be necessary to have the individual companies

* There were significant factual issues in relation to anti-competitive behaviour which were not suitable for determination on written witness statements.

* The proper forum was a complaint to the Commission or a civil action against the companies against whom they had a complaint.

They relied on the observations of Dyson J in R v Ministry of Argiculture ex p Dairy Trade Federation Ltd [1995] COD 3, made on the application for permission, that the Court should, in determining whether there were alternative remedies more appropriate than judicial review, look at the real issues involved and the nature of the enquires needed.

160. In view of the conclusion to which I have come it is not necessary for me to decide this issue. I would observe that, in any event, the point should have been raised at a much earlier stage and not left until the substantive hearing.

161. If similar issues arise in future in relation to the PPRS or a similar scheme where significant and complex issues of fact arise, then consideration should be given at the outset as to whether there is a more appropriate forum than this Court for the determination of the issue. If a determination of the extent of the scope of the operation of the modulation provisions in respect of competition from generics and from patented products within the same therapeutic class has to be undertaken, then it may be that that complex exercise should be carried in another forum. No doubt if there is a further challenge to the PPRS or a similar scheme, it would be appropriate to consider this question on or immediately after the application for permission, when the issues and the scope of the evidence necessary to adjudicate upon those issues can be examined. It is not a point that should be left to the hearing of the application.

CONCLUSION

162. I therefore have come to the view on the present state of the evidence before the Court that the applications fail and that the PPRS and the modulation provisions are not in breach of Articles 28 or 81. However, as I have noted at paragraph 106, it is in my view important that information about the operation of the market is made available and published, particularly so that the operation of the PPRS can be scrutinised to ensure that it is not having a potentially discriminatory or anti competitive effect on parallel importers.

163. Had I come to a different view, then an important question would have arisen on the issue of remedy. In view of its potential significance, I will refer briefly to the issue.

The issue on the form of relief

164. If API had been correct in their contentions, on the initial stance that they adopted that they did not wish to strike down the PPRS as a whole, then a determination of the arguments would have raised the question of whether, in circumstances in which the Court had found the modulation provisions of the PPRS contrary to Article 28 or 81, the PPRS could continue. However, in their closing submissions API seemed to accept that the consequence of their argument might be to affect the validity of the entire PPRS. In those circumstances, the Department contended that they had not had a proper opportunity of putting forward evidence as to why the court should not strike down the whole of the PPRS. I would not have been very sympathetic to the Department's position, as it was they who had made the point that chapter 21 could not be severed. If therefore I had determined the application in favour of API, the fairest way of dealing with the matter would have been to have held a further hearing on this issue.

165. Alternatively, API contended that I could make a declaration to the effect that the modulation provisions were contrary to Article 28 or Article 81 or declare that it would be unlawful for the Department to accept any application for modulation that would have an effect on parallel imports. The remainder of the PPRS could operate as before.

166. I cannot see how a Court could have declared the modulation provisions were contrary to Article 28 or Article 81, as the inevitable consequence of such a declaration would have been that the whole of the PPRS failed. Nor could I see how a Court could make the declaration API sought as to the operation of the modulation provisions. As currently formulated such a declaration would have had precisely the same effect as the declaration that the PPRS was unlawful. Thus, had I come the view that the modulation provisions were contrary to Article 28 or Article 81, the result would have been, subject to any further submissions, the striking down of the whole of the PPRS.