Supreme Court of Ireland Decisions

Judgment Title: Competition Authority -v- O'Regan & ors

Neutral Citation: [2007] IESC 22


High Court Record Number: 2003 8680p

Date of Delivery: 08 May 2007

Court: Supreme Court

55

THE SUPREME COURT
2005 No. 077

Denham J.
Hardiman J.
Geoghegan J.
Fennelly J.
Macken J.

BETWEEN
THE COMPETITION AUTHORITY
Plaintiff
and
JOHN O'REGAN, MICHAEL GORMLEY, CARMEL DOWLING, SAMUEL ADAIR, MARK BAILEY, JEROME DAWSON, DAVID EASTMENT, GERRY FOLEY, KAY GEOGHEGAN, SELENA GILEECE, MARY GRIFFIN, ANN LLOYD, JIM McMAHON, CON O'BRIEN, ANN BYRNE AND TOMMY WEIR
Defendants


Judgment of MR JUSTICE FENNELLY delivered the 8th day of May, 2007.


1. This is the first appeal to come before this court, which involves application of substantive competition law. The appeal is from the judgment of Kearns J in the High Court, delivered on 22nd October 2004. The learned trial judge made a number of orders against the defendants pursuant to the Competition Act, 2002. The defendants, for the purposes of the action, represent the principal association of credit unions in the State. They have appealed the High Court judgment and orders.
2. The plaintiff is the Competition Authority (hereinafter “the Authority”). It is established pursuant to statute for the purpose generally of furthering and supervising competition in economic markets in the State. It is empowered by section 14 of the Competition Act 2002 (“the Act of 2002”) to bring actions, such as the present, to enforce section 4 (anti-competitive agreements, decisions and concerted practices) and section 5 (abuse of dominant position) of the Act of 2002.
3. The defendants were the elected members of the Board of the Irish League of Credit Unions (“ILCU”) as of the date of the last annual report prior to the commencement of these proceedings. ILCU is an unincorporated association, having a place of business in Dublin. The Authority has sued the defendants as being representative of ILCU. While there is a dispute on the pleadings as to whether the defendants are properly sued in that representative capacity, no serious issue was taken on the point at the hearing of the appeal. It was accepted that alternative persons could be nominated.
4. The Act of 2002 was enacted in order to provide, by way of analogy with Articles 81 (section 4) and 82 (section 5) of the Treaty establishing the European Community (“the Treaty”), for the prevention of activities which restrict or distort competition in trade in the State or which constitute abuse of a dominant position in trade.
5. The present case principally concerns section 5. Though findings and orders were also made pursuant to section 4, the case for the Authority essentially stands or falls on the section 5 case. I will first set out the terms of sub-sections (1) and (2) of section 5. Sub-section (3) concerns mergers and acquisitions and is not relevant. The relevant sub-sections provide:


“(1) Any abuse by one or more undertakings of a dominant position in trade for any goods or services in the State or in any part of the State is prohibited.

(2) Without prejudice to the generality of subsection (1), such abuse may, in particular, consist in—

(a) directly or indirectly imposing unfair purchase or selling prices or other unfair trading conditions,
(b) limiting production, markets or technical development to the prejudice of consumers,
(c) applying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage,
(d) making the conclusion of contracts subject to the acceptance by other parties of supplementary obligations which by their nature or according to commercial usage have no connection with the subject of such contracts.”

6. The present case principally concerns sub-paragraph (d): the Authority says that ILCU abuses a dominant position in the market for credit-union stabilisation services by requiring purchasers of that service also to purchase from it its representation services. Apart from adaptation of its wording to refer to trade within the State rather than the common market, section 5 of the Act of 2002 is identical with Article 82 of the Treaty.
7. Section 4 is a longer section. For the moment, it suffices to cite sub-section (1), which replicates, mutatis mutandis, Article 81(1)EC:

“(1) Subject to the provisions of this section, all agreements between undertakings, decisions by associations of undertakings and concerted practices which have as their object or effect the prevention, restriction or distortion of competition in trade in any goods or services in the State or in any part of the State are prohibited and void, including in particular, without prejudice to the generality of this subsection, those which—

(a) directly or indirectly fix purchase or selling prices or any other trading conditions,
(b) limit or control production, markets, technical development or investment,
(c) share markets or sources of supply,
(d) apply dissimilar conditions to equivalent transactions with other trading parties thereby placing them at a competitive disadvantage,
(e) make the conclusion of contracts subject to acceptance by the other parties of supplementary obligations which by their nature or according to commercial usage have no connection with the subject of such contracts.”

Credit Unions
8. The following account of the nature of credit unions is common to the judgment of the High Court and to the written submissions and is common case. Credit unions are individual autonomous savings and credit co-operatives established by groups of individuals who share a "common bond". The most usual type of common bond is that shared by individuals living or working within a particular community, with over 90% of all credit unions being based in a particular local community. The main objectives of credit unions are:

• the promotion of thrift,
• helping members to accumulate savings,
• providing loans at reasonable rates of interest, and,
• the control of members' finances for their own benefit.
9. Credit unions advocate an ethos of mutuality, volunteerism, self help and a “not-for-profit” philosophy. ILCU subscribes to those views and aims to promote unity and co-operation, the development of high quality financial services and the recognition of the value and role of both credit union volunteers and staff to the movement and to society.
10. In 1960, four credit unions came together, under the name of the Irish League of Credit Unions, for the purpose of trying to ensure that credit unions would conduct their affairs on the basis of a set of agreed operating principles and of presenting a united front to government and state agencies in relation to the legislative problems of credit unions. ILCU was and remains a voluntary unincorporated association.
11. When the Credit Union Act, 1966 was passed, there were approximately 339 member credit unions in ILCU. That Act provided that credit unions should have a set of operating rules in a form approved by the Registrar of Friendly Societies, who was to regulate credit unions under the Act. Pursuant to this requirement, the members of ILCU adopted a set of standard rules for adoption by each member credit union and under which they would operate. These standard rules were approved by the Registrar, in accordance with the Act.
12. The Credit Union Act, 1997 repealed the Act of 1966. Section 6(1) of the Act of 1997 restricts registration to credit unions whose members share one of the common bonds listed in section 6(3) such as: following a particular occupation; residing or being employed in a particular locality; being employed by a particular employer or having retired from employment with a particular employer; being a member of a bona fide organisation or being otherwise associated with other members of the society for a purpose other than that of forming a society to be registered as a credit union. Section 6(2) provides that a society may be registered if the Registrar is satisfied that it is formed for the objects of:

“(a) the promotion of thrift among its members by the accumulation of their savings;
(b) the creation of sources of credit for the mutual benefit of its members at a fair and reasonable rate of interest;
(c) the use and control of members' savings for their mutual benefit;
(d) the training and education of its members in the wise use of money;
(e) the education of its members in their economic, social and cultural well-being as members of the community;
(f) the improvement of the well-being and spirit of the members' community; and
(g) subject to section 48, the provision to its members of such additional services as are for their mutual benefit.”

13. ILCU, in 1998, adopted a new set of standard rules for credit unions pursuant to the 1997 Act. All its affiliated credit unions have adopted and follow those rules. ILCU now has some 437 members. It operates throughout the island of Ireland. It has 104 members in Northern Ireland.
14. Membership of ILCU is open to any credit union agreeing to adopt the standard rules and to abide by ILCU’s rules, bye-laws and resolutions.
15. Apart from the promotion of the ethos and vision already mentioned, the activities of ILCU include:

§ Representation of members in contact with Government and its agencies and other bodies, including representation in relation to legislative matters;
§ Provision of insurance (particularly loan protection/life insurance) for members of credit unions or for credit unions themselves;
§ Provision of a stabilisation fund (the key question in the case), including self-regulation for that purpose.
Insurance and savings protection service
16. The history of these two distinct aspects of the activities is closely intertwined.
17. ILCU has, since 1976, provided to its member credit unions an insurance service called loan protection/life savings insurance or LP/LS. This insurance must be provided by each member credit union to its own borrowers. ILCU obliges member credit unions to “carry such insurances in respect of itself and for and on behalf of its own members as may be laid down from time to time by members in general meeting.” This is done through an authorised insurance company wholly owned and controlled by ILCU, namely ECCU Assurance Company Limited, which was established in 1980. At the 1986 AGM, ILCU passed a resolution to the effect that each member:

"Shall and will insure with their own insurance company (ECCU Assurance Company limited) on its commencing operation … in all risks covered by our company (i.e. LP/LS)".

18. The policy is, firstly, for the benefit of any borrower from a credit union or rather for that of his estate. If a borrower dies, the loan will be wiped out and his estate will benefit to the extent of up to twice the level of any savings he may have with the credit union. However, LP/LS also benefits the credit unions. If such insurance did not exist, they would suffer loss and expense in pursuing claims against the estates of deceased members.
19. Prior to 2004, credit unions paid premiums for LP/LS to ECCU. That company, in turn, deducted commissions, called operational rebates, from those payments and paid them to ILCU in payment of the affiliation fees of the member credit unions. A final aspect of this scheme, to be discussed later, is that ILCU funded the Savings Protection Scheme (SPS) from its own general revenues, including these funds.
20. Although LP/LS is bound up with the development of the Authority’s case against ILCU, it is useful to note, at this point, that it is common case that ILCU or its members have a miniscule share in the market for that type of insurance. Although the Authority had at one point contended that the mandatory insurance requirement contained in the rules of ILCU infringed the Act of 2002, (an allegation to that effect was included at paragraph 16 of the Statement of Claim), that case was explicitly abandoned at the hearing in the High Court. It should be noted, however, that ILCU has the power to expel members in accordance with section 11 of rule 1:
"A member may be expelled from the League by resolution of a majority present at the meeting of the board of directors for any grave and sufficient reason including wilful and/or persistent breach of, or refusal to comply with, any of these rules".

21. I turn then to describe the background to the Savings Protection Scheme (“SPS”). It may be useful to indicate in advance why the SPS is central to the case: the Authority claims that it is a service provided by ILCU to member credit unions; that it is a tradable service distinct from its representation services; that ILCU enjoys a dominant position in the market for that service; finally, that ILCU has abused its dominant position in that market by refusing to provide access to SPS to credit unions which leave or are expelled from membership.

The Savings Protection Scheme (SPS)
22. In 1968, ILCU established a stabilisation scheme with an initial sum of £500. Its object was to provide some assurance of financial stability for credit unions. It had three principal elements:

(1) An inspection service whereby representatives of ILCU would monitor individual credit unions;
(2) An advisory service designed to improve efficiency of individual credit unions and highlight any problems at an early stage;
(3) Creation of a reserve or fund which ILCU could use in its discretion to assist credit unions in trouble where the very existence of such a reserve would help maintain public confidence.
23. This was an entirely non-statutory or voluntary scheme, though members of ILCU are required to be members. As explained later, adoption of an SPS did not become mandatory by law until 2001 and then only for credit unions established after that time. The essential object of the stabilisation scheme and of the SPS is the provision of assurance of stability and solvency to individual members as depositors with credit unions, to the public generally and, naturally, to the credit unions themselves. This entailed, as ILCU has emphasised especially, an indispensable element of voluntary submission by member credit unions to financial oversight or scrutiny. This extends to inspection, to which the member credit union may be obliged to submit, of its accounts and records. This form of prudential inspection is fundamental to confidence in financial institutions generally. Credit unions now form an important part of the national financial system.
24. In the 1980’s ILCU considered and took advice on whether to introduce a form of guarantee insurance. There was concern arising from a number of high-profile financial failures. Following expert financial and legal advice, it decided to develop the existing stabilisation scheme, which it called a Savings Protection Scheme. Its own report recorded that the existing scheme had saved three credit unions from financial collapse without loss of members’ savings. On 7th March 1989, the SPS was approved by the Registrar of Friendly Societies, who was then responsible for oversight and supervision of credit unions, subject to conditions. It received similar approval from the corresponding Registrar in Northern Ireland. Credit unions are now regulated by the Irish Financial Services Authority (IFSRA)
25. Under section 6(1)(f) of the Credit Union Act, 1997 (which was brought into force with effect from 1st August 2001), credit unions established after that date are obliged to operate or participate in a savings protection scheme. Section 46(2) of the Act of 1997 defines a savings protection scheme as “a scheme established to protect, in whole or in part, the savings of members of a credit union in the event of insolvency or other financial default on the part of the credit union……”
26. Rule 55(1) of the standard rules for credit unions formulated by ILCU, provides that: “The credit union shall participate in the Irish League of Credit Unions savings protection scheme.”
27. All member credit unions were members of the SPS.
28. Resolution 10 passed by the 1987 AGM of ILCU restricted participation in any guarantee scheme to members of ILCU and provides as follows:

"That this AGM resolves that, in any guarantee scheme set up now or in the future by ILCU, only credit unions affiliated to and under the supervision of the ILCU will be covered by or allowed participate in such a scheme."

29. As at 31st December 2002 the total value of savings protected under the ILCU SPS amounted to €8.25bn. The SPS fund that that date amounted to €73.5m or 0.88% of the total savings protected.
30. Since January 2004, ILCU has changed the funding system. At that time, some members expressed concern that the premiums for LP/LS were uncompetitive. They could get lower rates elsewhere. I will return to the circumstances in which some credit unions withdrew from or were threatened with expulsion from ILCU. At this point, I note that ILCU (actually ECCU) reduced LP/LS premiums by some 20%, introduced a new affiliation fee to pay for ILCU’s own services and, finally, that the SPS has been funded by a specific contribution, calculated by reference to the value of assets. This contribution is made directly to the SPS fund by the member credit unions. The monitoring and advice aspect of the SPS continues to be paid for out of ILCU’s own resources.
31. Kearns J, in his judgment correctly described the SPS as follows:

“The SPS is a stabilisation fund which does not offer guarantees or insurance to savers, or to individual credit unions, but rather enables discretionary assistance to be given in appropriate cases by way of emergency finance for an individual credit union which may be in difficulty or a sum of up to €12,700 for any individual saver on the failure of a credit union.”
32. Its essence is that it constitutes a stabilisation mechanism by which credit unions, through ILCU, monitor and advise credit unions as to their management and operation, provide remedial help to credit unions which are weak in these areas and may (at the discretion of ILCU) provide financial assistance to an individual credit union in trouble by allowing it to draw, by way of loan, guarantee or cash transfer, on the SPS fund. No credit union has any entitlement as of right to financial assistance, nor has any credit union any proprietary interest in the SPS. No credit union may withdraw funds from the SPS. Rule 6, section 2 provides:

“On cessation of membership no member shall have a claim on any assets of the League or any League affiliate other than investments but cessation of membership shall not relieve the former member from any liability to the League then existing under these rules or as may be determined by the Board."

33. The SPS is the only such scheme in the State.
34. About 2000/2001, as already mentioned, some credit unions were unhappy with the level of LP/LS premiums. There was also extensive criticism of the costly failure of a computer system called ISIS, which caused a loss of some €7m. to ILCU. Some credit unions voluntarily disaffiliated. Others, who sought LP/LS cover elsewhere, were threatened with expulsion from ILCU. A new association, CUDA, was formed. It had twenty-one members in 2002 and 19 at the time of the High Court hearing.
35. In January 2003 ILCU moved to disaffiliate 12 credit unions for not obtaining their LP/LS insurance from ECCU. These disaffiliation proceedings were postponed by undertakings given by ILCU to the High Court following initiation of the present proceedings by the Authority.

Pre-litigious history
36. The Authority has formulated, as presented in the High Court, a specific complaint against ILCU of alleged anti-competitive behaviour pursuant to section 5 of the Act of 2002. It is that ILCU is abusing its dominant position in the SPS market by imposing a condition on the supply of that service. The condition is that a purchaser of SPS must purchase ILCU’s representation services. The Authority has also suggested, but very much as a subsidiary argument, that this conduct is anti-competitive by reason of abusive refusal to supply by a dominant firm. In addition, it alleges that the same behaviour constitutes a breach of section 4 of the Act.
37. It is necessary to outline the background and history leading up to the formulation of that complaint, which is not the one advanced in the statement of claim and which emerged for the first time only in the report of the Authority’s expert economist.
38. On 28th September 1992, ILCU notified its rules to the Authority pursuant to the Competition Act, 1991, for a ruling as to whether they constituted an agreement involving distortion of competition. The Authority, in its decision dated 20th November 1995, considered the requirement that member credit unions purchase their LP/LS from ECCU. It ruled that the market for that service was not separate from that for life assurance generally. It noted the statement of ILCU that any member not complying would be open to disaffiliation proceedings. The Authority ruled that the rules did not offend against section 4(1) of the Competition Act, 1991, the predecessor of section 4 of the Act of 2002. No question of a dominant position was raised. It observed:

“Any credit union which did not want to participate in such arrangements could leave the ILCU and could continue its operations. The arrangements prevent other insurance companies competing for this particular business but, in reality, this only applies to 0.6% of total life assurance premiums. The Authority does not believe, given the tiny proportion of the market involved, and the fact that credit unions can opt out of these arrangements, that this provision can be said to prevent, restrict or distort competition.”

39. In about 2000 to 2001, there was discontent on the part of some credit unions, already mentioned, regarding the level of LP/LS premiums and some other matters. In 2001, a number of members and former members of ILCU formed the Credit Union Development Association (CUDA). A complaint was made to the Authority concerning the requirement to purchase LP/LS from ECCU. This came from a member which had ceased to purchase LP/LS from ECCU and had been threatened with expulsion. In fact, that member left voluntarily. By a letter of 22nd April 2002 addressed to ILCU, the Authority raised these issues and noted the threat of disaffiliation. The Authority had power to revoke its decision of 1995 provided there were had been a material change of circumstances: section 8(6) (a) of the Act of 1991.
40. In the course of its inquiry into this complaint, the Authority recorded for the first time the complaint of some credit unions that the real penalty was not disaffiliation as such, but the consequential withdrawal or denial of the protection afforded by the SPS. In its statement of objections dated 22nd May 2002, the Authority appeared to depart radically from the market definition it had adopted in its 1995 decision: LP/LS sold by ILCU represented a minute part of the general market in life assurance. It now proposed a “narrow market definition.” A narrower market definition naturally makes it easier to prove the existence of a dominant position. The market would be in LP/LS insurance for credit unions. The Authority did not, however, develop this reasoning. Certainly, there has been no suggestion in the present case that the market should be defined as that in LP/LS services for credit unions.
41. By its decision dated 28th June 2002, the Authority revoked its 1996 decision. This decision was based on two suggested changes of circumstances. One was the emergence of some ILCU members who objected to the joint buying arrangements for LP/LS. More notably, it relied upon the making of SPS cover compulsory by the introduction of Rule 55(1), which I have mentioned above. It concluded:

“……the member credit union contributes to the SPS fund. Should it leave the ILCU, voluntarily or otherwise, it has no claim on any of the accumulated fund. However, to be placed in a similar position when it leaves the ILCU it would need to set aside 1 per cent of its savings……This would impose a considerable financial penalty on any ILCU member credit union that wished to avail of lower priced LP/LS cover than that offered by ECCU……”
42. The Authority thus revoked its decision that the Rules of ILCU did not distort competition. All proceedings up to that time, of course, concerned the applicability of section 4 of the Act of 1991 (now section 4 of the Act of 2002). Counsel for the defendants has heavily criticised this decision. However, it is not before the Court by way of judicial review or any equivalent procedure. It is relevant only insofar as it provides a background to the emergence of SPS as the real issue in these proceedings and the tentative suggestion that there is a market in LP/LS services for credit unions.
43. A number of meetings took place between representatives of the Authority and of ILCU. The Authority has acknowledged that ILCU was both helpful and cooperative. On 4th April 2003, the Authority wrote to the solicitors for ILCU suggesting, admittedly rather tentatively, for the first time, that the conduct of ILCU (in relation to the supply of LP/LS) “might constitute a breach of section 5 [of the Act of 2002] in the market for representation of credit unions, where ILCU arguably holds a dominant position.” It should again be noted that this is not the complaint as ultimately formulated at the hearing of the action.
44. In a letter of 27th May 2003, the Authority formally advanced the new contention that the “relevant market is credit union representation” and that “there is a market for credit unions representation in the State.” This contention, which is still not the one advanced at the trial, was based on ILCU’s own Report of the Review Commission of 2002. The Authority contended that ILCU had a dominant position in that market, by representing a very high proportion of all credit unions. It was abusing that dominant position by “enforcing and imposing membership conditions on individual credit unions that discourage the emergence of rival representative credit union organisations.” Nonetheless, in the same letter, the Authority said:

“The Authority is not stating that ILCU cannot make LP/LS insurance mandatory or that they cannot disaffiliate a member credit union for not complying with this rule. The breach is rather the associated loss of access and/or no refund from the SPS.”

45. This led very shortly to the issue of proceedings. The Plenary Summons was issued on 22nd July 2003.

The Statement of Claim
46. The statement claim alleges that the defendants, as officers of ILCU, are responsible for its actions. ILCU is engaged for gain in the provision of various services, which are listed, to credit unions and that it is an undertaking for the purposes of section 3(1) of the Act of 2002. Alternatively, ILCU is an association of undertakings and the individual credit unions are undertakings.
47. The key allegation is in paragraph 5:

“There is a distinct market for credit union representation in the State, in which market ILCU has a dominant position.”
48. It defines that product market as comprising the entire composite of all ILCU services, including advocacy, lobbying, financial services and SPS. It is alleged that ILCU is dominant in the market for credit union representation, by virtue of the fact that nearly all of the credit unions in the State are long-standing members and that its members account for 90% to 95% of credit union assets.
49. The statement of claim recounts the history of the SPS, of the rules confining it to members of ILCU, of the threat to disaffiliate credit unions which had refused to purchase LP/LS from ECCU and the effect of this on access to SPS. Paragraph 15 pleads that the refusal of ILCU to supply access to SPS constitutes an abuse of its dominant position in the market for credit union representation. These allegations all arise pursuant to section 5.
50. The Authority repeats each of the particulars of alleged abuse of a dominant position, mutatis mutandis, as constituting a breach of section 4(1) of the Act of 2002. The Authority alleges that each of the decisions and practices of ILCU cause damage to the competitive process and to consumers.
51. The statement of claim claims consequential declarations that the acts complained of constitute breaches of sections 4 and 5. Once again, it is important that the Authority defined the relevant product market as being the entire composite of all ILCU services associated with its representation services. It did not suggest that the supply of SPS constituted a distinct product which existed in a separate product market.

Issues in the Case
52. The parties, pursuant to case management procedures under the direction of the learned High Court judge agreed, in a very lengthy document dated 18th June, 2004 upon the issues to be decided. These are set out at length in the judgment of the learned trial judge. I will set out only the headings (the document was much longer). They are:
(1) Whether the named defendants are proper defendants to these proceedings.
(2) Whether the Court has jurisdiction over the named defendants who reside outside of the State.
(3) Whether the ILCU is engaged in trade in any goods or services within the meaning of the Competition Act, 2002, and if so in which goods or services.
(4) Whether the ILCU constitutes an association of undertakings within the meaning of the Competition Act, 2002.
(5) Whether there is a market for credit union representation services in the State and, if so, the characteristics of the market.
(6) Whether the appropriate market(s) in respect of those activities of the ILCU which may take place in trade is the market for credit union representation.

(7) Whether the ILCU has a dominant position in the alleged market for credit union representation.

(8) Whether the threat of disaffiliation from the ILCU for breach of the ILCU rules/agreement which will result in the loss of access to and/or lack of refund from the SPS for credit unions on disaffiliation for breach of the ILCU agreement constitutes an abuse of the ILCU's alleged dominant position.

(9) Whether the ILCU's refusal to permit access to the SPS constitutes an abuse of the ILCU's alleged dominant position.

(10) Whether the threat of disaffiliation from the ILCU for breach of the ILCU agreement which will result in the loss of access to and/or lack of refund from the SPS for credit unions on disaffiliation constitutes a breach of s.4(1) of the Competition Act, 2002.

(11) Whether the Rules, decisions/agreements and practices of the ILCU constitute a breach of s.4(1) of the Competition Act, 2002.

(12) Whether the threat of disaffiliation and/or any such disaffiliation and the Rules, decisions and practices of the ILCU cause damage to the competition process and to consumers.

(13) Whether, if the Rules, decision and practices of the ILCU which will result in the loss of access to and/or lack of refund from the SPS for credit unions on disaffiliation and/or the ILCU's refusal to supply access to the SPS on an open basis constitute a breach of s. 4(1), they comply with conditions referred to in s. 4(5) of the Act.
(14) The reliefs if any to which the plaintiff is entitled.

53. Issues numbered five and seven concern whether there is a market for credit union representation services in the State and whether ILCU enjoys a dominant position in that market. Under those headings, many facts were agreed, but the existence of such a market was disputed.
54. Very strikingly, none of the agreed issues concerned whether there was a distinct market for the provision of stabilisation services (SPS) for credit unions, whether ILCU enjoyed a dominant position in such a market, whether ILCU had abused such a dominant position, whether by tying the provision of SPS to the provision of credit union representation services or otherwise.
55. Nonetheless, that is the case the Authority ultimately successfully made and which is the subject of the present appeal.
56. As counsel for the defendants (now appellants) said in intervention on the first day of the hearing in the High Court, the only case pleaded was in respect of the market for credit union representation services. The concept of a market for SPS was first introduced in the expert report of Professor Seabright on behalf of the Authority. That report was dated 24th May 2004 and was delivered shortly before the hearing. The learned trial judge recognised that Mr Massey, the expert economist called on behalf of the appellants, had had “no opportunity to address the contention advanced so late in the day by the Competition Authority that savings protection formed a separate product market.” One result of the late change of direction by the Authority was that Mr Massey was unable, in his report, to address the issue upon which the Authority based its case. This difficulty was compounded by the fact that, for reasons of availability, Mr Massey gave evidence for the defendants before Professor Seabright gave his expert evidence on behalf of the Authority.
57. The learned trial judge explained the manner in which this radical shift in the case made by the Authority ultimately came to be dealt with as follows:

“……the suggestion that the SPS fund formed part of a separate product market was not a case advanced by the Competition Authority in the statement of claim. Nor was that issue an issue defined as such in either the summary of issues agreed between the parties, or in the memorandum of issues 'agreed and in dispute' dated 18th June, 2004. Quite clearly, the notion of the SPS as a separate product market emerged only following the production by Professor Seabright of his economic report in this case. However, Mr. Collins SC, counsel for the defendants, accepted that he was not prejudiced by this late adjustment to the plaintiffs claim, preferring to assert that it merely underlined the weaknesses inherent in the claims and assertions advanced on behalf of the plaintiff and served to undermine the credibility of the case generally.”

The High Court Hearing
58. The action was heard in the High Court over a period eleven days in June and July 2004. Evidence was heard from witnesses as to fact and from expert economists: Professor Seabright for the Authority and Mr Pat Massey for ILCU. With the consent of the parties the High Court appointed its own expert economic assessor pursuant to Order 36, Rule 41 of the Rules of the Superior Courts.
59. The learned trial judge considered in some detail whether the case fell to be determined by reference to sections 4 and 5 of the Competition Act, 2002 or by reference to Articles 81 and 82 of the EC Treaty. He noted that the parties had “been able to agree that the same economic analysis is required and the same legal principles apply……” under each dispensation. On analysis, he came to the conclusion that the impugned agreements or alleged abusive practices probably did have the potential to affect trade between Member States. That should mean that the case should be assessed by reference to Articles 81 and 82. In the end, it does not appear that this conclusion has any material effect on the relevance of evidence, the economic analysis or the outcome of the case.
60. The substance of the High Court judgment should, in my view, for the purposes of this appeal, be considered under three principal headings:

(1) Whether credit union representation services and SPS should be considered as distinct products and in different relevant product markets;
(2) Whether ILCU enjoys a dominant position in the market for either of those services (especially SPS) assuming them to be in separate relevant product markets;
(3) Whether ILCU has abused its alleged dominant position in the market for SPS, principally by tying its provision to the provision of a separate product, namely credit union representation services but alternatively by abusive refusal to supply.

61. However, for reasons which will appear later, it is an essential precondition to the case for the Authority that the first question be answered affirmatively. If it is not, the second and third questions do not arise.
62. The learned trial judge considered whether credit union representation services and stabilisation services (the supply of SPS) are distinct products or services and whether they are in separate product markets. He answered both these questions affirmatively. It is common case that this step in the analysis was vital for an assessment of the case being made by the Authority, namely that ILCU holds a dominant position in the market for SPS and abuses that position by tying its provision to the supply of representation services. It is also accepted that there are two principal means of analysing the existence of separate product markets. Firstly, there is a test known under the acronym of the SSNIP test: this involves statistical analysis of whether a small but significant (usually taken to be 5 to 10%) increase in price will cause a significant number of consumers to switch to a close substitute. Secondly, there is the “intuitive” test, whereby the products are considered by reference to their “innate characteristics.” The learned trial judge preferred the latter test.
63. Since it will be necessary to consider these issues in some detail when deciding the issues on this appeal, I will summarise the rival contentions as put before the High Court in bare outline.
64. Mr Paul Gallagher, Senior Counsel, for the Authority said that the SSNIP test and the innate-characteristics test should be considered as alternatives rather than as rivals. There was clear demand for representation services, in a distinct product market, comprising a complex bundle of products and services to individual credit unions. The Authority had not carried out any statistical analysis for the purpose of applying the SSNIP test, due to the absence of quantitative data. He criticised the SSNIP analysis which had, in fact, been carried out by Mr Pat Massey, the expert called on behalf of ILCU. I should interpolate, once more, that Mr Massey naturally based all his analysis on the case pleaded by the Authority and not on the one advance very late in the day in the report of Professor Seabright. Mr Gallagher went on to argue that there was a distinct product market for SPS, for the following reasons:
§ There is only one SPS available in the State;
§ Participation in a stabilisation scheme is vital for confidence in credit unions;
§ This might take the form of deposit insurance;
§ It was extremely unlikely that a 5 to 10% increase in price for SPS would bring about a switch to close substitutes;
§ There was little or no possibility of entry into the market for SPS taking place;
§ There is a separate market for the provision of SPS and ILCU is dominant in that market.

65. Mr Michael Collins, Senior Counsel, for ILCU said that Mr Massey had correctly applied the SSNIP test, employing three different definitions of price to alternative bundles of services. Consumer credit unions had switched to another product, namely LP/LS services, where it was accepted that there was a highly competitive market. He criticised the failure of the Authority’s expert, Professor Seabright, to provide any analysis as to whether SPS was in a distinct product market. The Authority had merely asserted at the last moment that SPS was in a distinct relevant product market. The evidence showed overwhelmingly that the demand for representation services and SPS go hand in glove. There is no pattern of independent demand for either.
66. There was a separate discussion regarding the definition of the relevant geographic market. Mr Gallagher contended that it is natural to identify the State (excluding Northern Ireland) as constituting the relevant geographic market. The credit unions in Northern Ireland are subject to a different regulatory regime. Mr Collins, on the other hand pointed out that a substantial number of members of ILCU are based in Northern Ireland and subject to the regulatory regime in force there. There are more than 80 credit unions on the island of Ireland who are not affiliated to ILCU.
67. One issue on which the expert economists disagreed in their evidence was as to how to characterise the supply of the SPS. Mr Massey was of the view that ILCU’s member credit unions should be regarded as being engaged in supplying SPS to themselves, a type of self-supply amounting to vertical integration. Professor Seabright disagreed. He thought that SPS should be considered as a form of joint venture, the participants being the member credit unions, admittedly an unusually large number of participants.
68. The learned trial judge held that SPS constituted a product distinct from representation services. ILCU is the seller and the credit unions are the buyers. SPS forms a “backdrop resource akin to deposit insurance…” It could be made available to members of CUDA but for the rules of ILCU. It is “vital” to credit unions and it is highly likely that some form of savings protection scheme will be demanded from any supplier of credit union representation services. Having referred to Mr Collins’ analysis based on the Commission decision in the Microsoft case (referred to later), he thought the conditions there mentioned had a resonance for the present case. As part of his analysis of relevant product market, the learned trial judge referred to the market share of ILCU as being 80%, stating that it “enjoys a presumption of dominance in that market.” Since the start of 2004, ILCU charges its members separately for the SPS service. The learned trial judge did not consider that any product market description (for the purpose of the SSNIP test) which characterises the bundle as including SPS was capable of yielding an accurate market definition. Mr Massey’s application of that test delivered a distorted result. I would observe, in anticipation of more detailed consideration of this issue, that Mr Massey had applied the SSNIP test to the precise product market defined by the Authority in its statement of claim.
69. The learned trial judge explained that he preferred the approach of Professor Seabright, even though he lacked quantitative data. He though that “common sense must surely indicate that ILCU as the sole supplier of SPS would be immune from a 5 to 10% price increase given the absence of any alternative product such as a deposit insurance equivalent.” He added that there was a “separate market for credit union representation services to include advocacy, provision of insurance and financial services and self-regulation.” He explained that he preferred, having considered both the SSNIP and the economic evidence, to adopt the innate-characteristics test to find that there are two markets at work. The learned trial judge also preferred the view of Professor Seabright on the question of self-supply. The thought that acceptance of Mr Massey’s analysis would provide a wide range of activities with a shield against competition-law scrutiny.
70. He also concluded that the State constituted the relevant geographic market, because there was no legal requirement for savings protection in Northern Ireland, stating that he had not heard any evidence of equivalence of competitive conditions in Northern Ireland. He added that nothing of significance hinged on this particular point.
71. He went on to consider the issue of dominance. It was common case that, on the assumption of a separate product market for credit union representation services, ILCU supplied 80% of that market in the island of Ireland and 85% in the State. In respect of SPS, the learned trial judge held that it went without saying that ILCU had 100% of the market. Relying on the presumption that large market shares (over 50%) are, save in exceptional circumstances, evidence of dominance, he concluded that ILCU had a dominant position both in the market for representation services and for SPS. (Case 27/76 United Brands v Commission [1978] ECR 207; Hoffman La Roche v Commission [1979] ECR 461.)
72. The above conclusions led on to consideration of the Authority’s argument that the ILCU abused its dominant position.
73. The Authority’s case was and still is that an undertaking, which enjoys a dominant position in one product market and uses that position to insist that purchasers of that, first, product also purchase another product also supplied by the dominant firm is engaged in abuse of its dominant position. This is a “tying” abuse. The dominant firm uses its market power to leverage an effect on a parallel or downstream market. (Case 333/94 P Tetra Pak v Commission [1996] ECR I-5951; Case 311/84 CBEM v CLT and IPB [1985] ECR 3261 (usually called “Télémarketing”)) It made an associated case that the actions of ILCU constituted an abusive refusal to supply by a dominant firm. (Joined Cases 6/73 and 7/73 Istituto Chemioterapico Italiano and Commercial Solvents v Commission [1974] ECR 233.)
74. Mr Collins, on behalf of ILCU did not deny that, on the essential underlying assumptions of dominance in a separate relevant product market, the tying activity was capable of amounting to an abuse of dominance. He contended, however, that, assuming the Authority’s case to be correct, the rule applied by ILCU was objectively justified. The SPS service was, by its nature supplied by credit unions to themselves on a collective basis and had to be binding, including by expulsion, if necessary, to be effective. In particular, the credit unions must permit monitoring by their fellow credit unions to minimise any call on the stabilisation fund.
75. Mr Collins also submitted that, although the Authority disavowed any reliance on that doctrine, in essence, the case for the Authority was that the SPS constituted an essential facility. He relied on the limitations on that doctrine recognised in recent literature and above all in the opinion of the Court of Justice in Case 7/97 Oscar Bronner GmbH v Media Print [1998] ECR I-7791.
76. The learned trial judge rejected the argument that the case was to be considered as one involving essential facilities. Although various witnesses had referred to the SPS service as “vital,” it was not essential in the sense that credit unions could not carry on business without it. He accepted the characterisation of the application of their rules by ILCU as a tying arrangement. There was a “double tie,” namely the requirement on credit unions who seek access to the SPS to avail of ILCU’s representation services and also its mandatory LP/LS insurance arrangements. It should be recalled, however, that the Authority had abandoned any suggestion that it was anti-competitive to insist on member credit unions to purchase LP/LS from ILCU. As already noted, the Authority has expressly stated that it is “not stating that ILCU cannot make LP/LS insurance mandatory or that they cannot disaffiliate a member credit union for not complying with this rule……”
77. The learned trial judge rejected the defendants’ arguments based on objective justification. He concluded that it was incontestable that ILCU’s conduct may in time foreclose the market.
78. Turning to the case based on section 4 of the Competition Act, 2002 (corresponding to Article 81EC), the learned trial judge noted that it was not contended that the ILCU arrangements were restrictive of competition by object. The court was concerned only with their anti-competitive effect.
79. The learned trial judge found that the defendants were in breach of section 4 “for much the same reasons as have already been offered by the court for its findings in relation to alleged infringement of s. 5 of the Competition Act, 2002.” He held that the “tie” has the effect of distorting and restricting competition in the market for credit union representation services. He referred again to leverage exercised in the market for representation services. He said: “Tying imposes costs on the credit union members concerned that discourage them from purchasing on the most favourable terms the representation services they might wish to purchase from rival credit union representative providers.”
80. For the purposes of section 4 of the Act of 2002, the Court was, for reasons more or less identical with those which led the learned trial judge to conclude that ILCU had abused its dominant position in the market for SPS by tying its supply to the purchase of its representation services, satisfied that the tie had the effect of restricting and distorting competition in the market for credit union representation services.
81. The learned trial judge gave careful consideration to the remedy which should be granted to the Authority, on the basis of his findings. He recognised that there could be circumstances where ILCU would legitimately be entitled to disaffiliate a credit union, but regarded resort to that procedure as unlawful when used for the particular reason at issue, namely refusal of access to SPS. Therefore, he considered that the resolution of ILCU which restricted participation in any future guarantee scheme to ILCU members to be void. He decided to grant an injunction against ILCU from disaffiliating credit unions which have been threatened with disaffiliation and consequential loss of access to the SPS fund because they have sought LP/LS cover other than through ECCU. Finally, he considered that non-affiliated unions or newly formed credit unions should, should they so wish, have access to SPS.
82. Prior to the adoption of his judgment the learned trial judge provided a copy to the Commission of the European Union pursuant to Article 11(4) of Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty. That provision obliges the competition authorities of the Member States to notify the Commission at least thirty days in advance of the adoption of any decision “requiring that an infringeent be brought to an end.” The Commission duly notified the High Court that it had no observations to make with regard to the envisaged judgment.

Identifying the Status of the Defendants on the Appeal
83. The parties have furnished extremely comprehensive submissions covering every aspect of the High Court decision. The essence of the case for the Authority has been well explained on a number of occasions. Whether by reference to section 4 or 5 of the Act of 2002 or Article 81 or Article 82 of the Treaty, it is a “tying” case. The named defendants are sued as representatives of ILCU, not of the individual credit unions. The statement of claim says that they are officers of and responsible for the actions of ILCU, an unincorporated association. It is pleaded that they have been nominated as defendants on behalf of ILCU to represent the members of ILCU. The defence pleads that the named defendants were the named members of the Board of ILCU as at the date of its last annual report prior to the issue of proceedings. The defendants have agreed that appropriate alternative names may be substituted. Nonetheless, the named defendants represent ILCU, an unincorporated association and not its member credit unions.
84. The defendants accept that individual credit unions are undertakings for the purposes of competition law generally. They are engaged in activities for “gain,” a term considered in Deane v Voluntary Health Insurance Board ([1992] I.R. 319). Nor is it in dispute that ILCU is an association of undertakings for the purposes of section 4 of the Act of 2002 or Article 81EC. However, section 5 and Article 82EC concerns abuse by “one or more undertakings.” Mr Collins, raises the question whether ILCU is itself an undertaking. He submits that a particular body may be an undertaking in respect of some of its activities but not others. While he refers to a number of authorities concerning the nature of an undertaking for the purposes of competition law (Joined Cases C-159/91 and C-160/91 Poucet v Assurances Générale de France and Pistre v Caisse Autonome etc. [1993] ECR I-637; Case C-67/96 Albany International IBV v Stichting; Case C-309/99 Wouters v Algemene Raad van de Nederlandse Orde van Advocaten [2002] ECR I-1577), the only authority cited to support the suggestion that ILCU was not engaged in economic activity was a statement from my opinion as Advocate General in Sodemare v Regione Lombardia [1997] ECR I-3395. He extrapolates from that the proposition that ILCU is organised on a principle of solidarity, that its activities involve subsidisation by larger credit unions of smaller credit unions which is an expression of the principle of solidarity and self-help.
85. Section 3(1) of the Act of 2002 defines an undertaking as meaning:

“a person being an individual, a body corporate or an unincorporated body of persons engaged for gain in the production, supply or distribution of goods or the provision of a service.”

86. I am satisfied that ILCU is an undertaking for the purposes of the activities of which the Authority complains in the present case. That is in accordance with constant case-law. In Case C-41/90 Höfner and Elser v Macrotron [1991] ECR I-1979 at paragraph 21, the Court of Justice held:

“It must be observed in the context of competition law …… that the concept of an undertaking encompasses every entity engaged in economic activity, regardless of the legal status of the entity and the way in which it is financed……”

87. As the learned trial judge correctly observed, “the mere fact that the ILCU involves co-operation between individual member credit unions does not prevent it being an undertaking.” He referred to Case T-61/89 Dansk Pelsdyravlerforening v Commission {1992] ECR-II1931, where the Court of First Instance reiterated the principle laid down in Höfner and Elser. ILCU is an unincorporated association, as distinct from also constituting an association of undertakings. The fact of its not being incorporated does not, as section 3(1) demonstrates, prevent it being an undertaking. The Authority claims that ILCU supplies distinct economic services. Assuming that it supplies its representation services for gain, whether or not including SPS, that is an economic activity. Whether it does so is, of course, a substantive issue on the appeal. It would beg the question to decide that matter at this point. It is clear to me that ILCU, if it provides the distinct services identified by the Authority, does so for gain. It is indisputably engaged in economic activity insofar as the supply of LP/LS is concerned. It is not engaged in the sort of uneconomic activity identified in some of the cases. Therefore, ILCU should be treated as an undertaking for all purposes of the appeal.
88. At the same time, insofar as it may be relevant, it is ILCU and not its member credit unions that is the effective defendant. It is ILCU that is alleged to be the supplier, as distinct services, of the representation services and SPS. On the other side, it is the Authority, and not any individual credit union or association of credit unions, which is the plaintiff. In a certain sense, nonetheless, it has appeared to stands in the shoes of potential complainant credit unions or, at any rate, to articulate their complaints. Specifically, the tying complaint is that ILCU refuses to supply SPS to credit unions generally or to any credit union in particular unless the purchaser also purchases its credit union representation services.

ARGUMENTS ON APPEAL
A. The Appellants; ILCU
89. I think it is fair to state that the first, principal and potentially determinative issue on the appeal is the identification of the relevant product market. If there are truly distinct markets for the supply of the two services identified by the Authority, then it is not in dispute that ILCU enjoys an exceptionally large market share in each of them. According to the learned trial judge, that share is 80 to 85% of the market for representation services and 100% of the market for SPS. In principle, that may not decisive. A very high market share may, in certain circumstances, exist without the seller having that market power which is the essential characteristic of a dominant position. However, it is clear that a large market share is always regarded as strongly indicative of dominance. Assuming that ILCU enjoys that position of dominance in a distinct product market for SPS, it is probably inescapable that it is engaged in abusive tying activity.
90. However, the entire superstructure of the Authority’s case is laid on the substratum or foundation of relevant product market.
91. ILCU disputes the findings of the learned trial judge in respect of relevant product market both in law and on the facts. It addresses, as separate issues, whether representation services and SPS are separate products and whether they are in separate product markets.
92. ILCU emphasises that no question of tying arises unless the representation services and SPS are separate products. ILCU noted that the Authority had abandoned its original case that ILCU provided a complex bundle of services, including SPS, which created a distinct product boundary, in favour of the identification of SPS as a distinct product in a distinct market in which ILCU was dominant. ILCU submits that, by its very nature, SPS is provided by ILCU, as a collective body, to its members on a collective basis. It cannot be provided by one credit union to itself. It necessarily entails close mutual monitoring and supervision to which member credit unions voluntarily submit themselves. This is the invariable practice throughout the world. The SPS fund and the accompanying regulatory regime are mutually inter-dependent. Thus, member credit unions do not purchase stabilisation services from ILCU as an arms-length seller.
93. Moreover, ILCU says that there is no evidence either in Ireland or anywhere else in the world of the existence of an independently-supplied SPS or stabilisation service. Firstly, no insurance company provided such a service, even in the United States, where there has been a strong credit union movement for many years. Secondly, in the United States, stabilisation schemes were set up by leagues like ILCU and confined to their members. No anti-trust enforcement action had ever been taken against any such league on the basis that this activity was anti-competitive. Thirdly, precisely the same situation prevails in Ireland.
94. ILCU also recalls the evidence given by representatives of a number of credit unions. The principal thrust of that evidence was that those credit unions claimed to have some form of proprietary, beneficial or equitable interest in the SPS fund which they had built, a claim not supported by the Authority. Furthermore, many of those witnesses objected strongly to being potentially monitored or supervised by ILCU. They had lost confidence in ILCU. What they really wanted was that SPS would be independently controlled and supervised on a nationwide basis. Finally, ILCU points to the existence since the date of the High Court judgment of a situation where any credit union may, as of right purchase SPS from ILCU, but none have done so.
95. ILCU complains that the learned trial judge, having decided that SPS was a separate product, simultaneously, and with little analysis, conclude that it was in a separate product market. It argues that the Authority has proffered little or no analysis as to whether SPS is in a separate product market and no quantitative analysis at all. In particular, it had made no attempt to use the SSNIP test.

B. The Respondents; the Authority
96. The Authority fully supports the findings of the learned trial judge: he had explicitly found that there can be independent demand for SPS as distinct from representation services. The Authority, in its written submissions complains that, while the learned trial judge had correctly treated the case as a tying case, the ILCU continue to seek to have the issues analysed by reference to the principles applicable to refusal to supply, particularly of essential facilities. The Authority says that this mischaracterises its case. Even if, as argued by ILCU, SPS and deposit insurance form one market, that fact emphasises that both are in a distinct market from representation services. Since 2004, ILCU has separately priced its SPS service. All credit unions formed since 1st August 2001 are required to belong to an approved SPS. Furthermore, it is highly desirable, from the point of view of the stability of the industry that credit unions formed prior to that date should belong to one. The fact that SPS and representation services were not traded separately prior to 2001 cannot determine whether SPS should be regarded as a separate product. The reality is that representation services are provided by CUDA, though it is unable to supply SPS.
97. The Authority contests the interpretation and emphasis placed by ILCU on the evidence of representatives of the individual credit unions. Accepting that many of them would prefer, for reasons principally related to lack of confidence in ILCU, that the SPS be placed under independent control, it is clear that their principal concern was to have access to SPS in the meantime. The Authority objects to consideration of the evidence submitted on affidavit for the appeal that no credit union had applied for access to the SPS. This evidence had not been tested in the High Court. In any event, the absence of such application from credit unions could be readily explained by the uncertainty created by the existence of the appeal.
98. The Authority also disagrees with ILCU’s summation of the position in the United States. In that jurisdiction, the evidence showed that stabilisation funds fell away over time to be replaced from 1971 by Federal Deposit Insurance.
99. The Authority contests the suggestion made by ILCU that the learned trial judge had insufficiently differentiated between the issues of separate products and separate product markets. The evidence established that there was no SPS available to credit unions in Ireland, save on the basis that the members of CUDA could establish one for themselves. Accordingly, once it was established that SPS constituted a separate product, it necessarily also constituted a separate product market.

The Courts and Competition Law
100. Cases, such as the present, requiring direct assessment of the market effects of allegedly anti-competitive conduct are a virtual novelty for our courts. The present appeal is the first occasion on which this Court has been faced with the need to assess the economic effects of market behaviour and to decide substantive issues of the law relating to competition.
101. Articles 81 and 82 of the EC Treaty embody in legal form the objective of prevention of two classic forms of anti-competitive behaviour, namely collusive or horizontal conduct between enterprises, which form cartels designed to fix prices or other market conditions to the benefit of the participants and unilateral conduct of one or more undertakings enjoying undue market power. The Oireachtas has chosen, since 1991, to incorporate the objectives of prohibiting such conduct in Irish Law. In turn, they constitute a European inheritance from the anti-trust law of the United States, in particular sections 1 and 2 of the Sherman Act.
102. Courts are thus required to integrate economic principles into law, which is not an easy task. Judges are not all familiar with economic principles. There is a substratum of generally accepted principles of classical economics. However, economic theory is in a state of constant development. Techniques of economic analysis vary and are sometimes contested between economists. The boundaries between economic theory and public policy vary over time and from one Member State to another. In particular, States differ with regard to the economic role that is accorded to government and its agencies.
103. There are, furthermore, two fundamentally different ways of incorporating competition principles into law. The first model assigns investigation and decision-making power to an independent expert administrative body designated as a competition authority. Under that model, primary decisions are made by the authority. Consequently, the role of the courts is normally limited to judicial review. This is the model of the European Union, where the Commission investigates and makes decisions, which may be reviewed by the Court of First Instance pursuant to Article 230EC. It is also the model in some Member States. The second model entails assigning to the courts the task of making the substantive decisions. It is true that, under the direct effect of Articles 81 and 82 of the Treaty, courts may, from time to time, be asked to resolve competition disputes between private undertakings. There has been a small number of such cases in this jurisdiction. The present case arises from the special role assigned by section 14 of the Competition Act 2002 to the Authority. The Authority has not made any decision other then to institute the proceedings. It identifies market conduct and invites the Court to condemn it. There is no prima facie legal presumption in favour of the Authority’s view. The Authority carries the normal civil burden of proof.
104. The Court is not, of course, entirely bereft of sources of guidance. The Court of Justice considers questions of interpretation, on references from the courts of the Member States, of Articles 81 and 82 (formerly Articles 85 and 86) of the Treaty. It has done this for almost fifty years. Until 1989, it performed the function of review of decisions of the Commission, a function thence taken over by the Court of First Instance. Thus, there is a rich body of case-law from which important underlying principles can be derived. In addition, the Court has the possibility of consulting and taking the advice of the Commission in accordance with Article 10 of the Treaty.
105. In embarking on this novel exercise of jurisdiction, the Court needs to identify some basic points of reference, to understand the essence of competition law.
106. The entire aim and object of competition law is consumer welfare. Competitive markets must serve the consumer. That is their sole purpose. Competition law, as is often said, is about protecting competition, not competitors, even if it is competitors who most frequently invoke it. Its guiding principle is that open and fair competition between producers of goods and services will favour the most efficient producers, who will thereby be encouraged to satisfy consumer demand for better quality products, wider choice and lower prices. Their reward is a greater market share. Production of better and newer products may necessitate expensive market research, involving a degree of economic resources and market power. Competition law does not outlaw economic power, only its abuse. Economic power may, indeed should, be the reward of effective satisfaction of consumer needs. It would be inconsistent with the objectives of free competition that successful competitors should be punished. It is not the existence but the abuse of a dominant position which offends principles of free and open competition.
107. It is obvious that these principles are in frequent and general tension and that competition authorities must strike a balance between competing considerations. Complex economic relationships need analysis before a conclusion may be reached as to whether particular impugned behaviour is injurious to or beneficial to competition and to the consumer.
108. Undertakings compete in the provision of goods and services, so that the notion of the product is necessarily central to market analysis. Identification of the relevant products and markets are the necessary starting point in every case.
109. The debate on the appeal illustrates two different ways in which the product, here taking the form of a service, dictates the debate. Mr Collins insisted on the need to answer two distinct questions, namely whether, firstly, a good or service constitutes a distinct product and, secondly, whether that distinct product is in a separate relevant product market. I do not think such a sharp distinction necessarily presents itself in all cases. Possibly that is because it is unusual to have a dispute about the identity of the product. The focus in competition writings is normally on the need to identify the relevant product market.
110. Mr Collins’s concern is that a first or preliminary question needs to be answered when the debate concerns whether there are, in truth, two distinct products at all or merely one. This type of issue sometimes arises where a supplier is accused of abuse of market power by “bundling” two or more distinct products. It does not always arise, but it is certainly central to the present case.
111. The second question arises where there are indisputably separate identifiable products, but where the economic question, which is of interest for competition purposes, is whether those products are in competition with each other, i.e., whether they are bought and sold in one market and are in competition with each other, or are in separate economic markets. The economic concept of elasticity of demand provides a conceptual basis for determining the issue of relevant product market. This is where the SSNIP test comes in. It postulates the existence of close substitutes. If the seller, deemed for this purpose to be a hypothetical monopolist, has the ability to increase his prices by the notional small non-transitory amount of 5% to 10% without significant loss of sales, he is in a separate relevant product market. However, this technique is designed to assess the degree of competition between similar or potentially similar products, which are capable of competing with each other. A simple example would be whether fruit juices of different flavours were in separate markets or in a single market for fruit juices. Whether or not bananas constituted a distinct market from other fruit was, famously, the subject-matter of the decision of the Court of Justice in Case 27/76 United Brands v Commission, cited above. This type of market definition for parallel markets is not relevant to the first question. The SSNIP test determines the degree of competition between products, not between competitors selling the same product. It is patently irrelevant to the case of complementary products, i.e., non-competing products.
112. In my opinion, the first question is the one that is most relevant-- to the present appeal. It is certainly the first question to be addressed.
113. The European Commission in Microsoft (Commission Decision of 24.03.04 relating to a proceeding under Article 82 of the EC Treaty. Case COMP/C-3/37.792; the decision of the Court of First Instance on Microsoft’s challenge is pending) decided that Microsoft was engaged in abusive “tying.” It had to decide whether Microsoft’s streaming media player and its operating system were two separate products. The Commission summarised (paragraph 794) what all parties agree to be the elements of tying abuse succinctly as follows:

“Tying prohibited by Article 82 of the Treaty requires the presence of the following elements: (i) the tying and tied goods are two separate products; (ii) the undertaking concerned is dominant in the tying product market; (iii) the undertaking concerned does not give customers a choice to obtain the tying product without the tied product; and (iv) tying forecloses competition.”

114. The first element mentioned in that passage, whether the tying and tied goods are two separate products, is the same as the first question postulated by Mr Collins, namely whether there are two separate products. The Commission went on to say (paragraph 800): “Products that are not distinct cannot be tied in a way that is contrary to Article 82.”
115. The Commission’s approach to distinctness (paragraph 803 of the Decision) is:

“The distinctness of products for the purposes of an analysis under Article therefore has to be assessed with a view to consumer demand. If there is no independent demand for an allegedly tied product, then the products at issue are
not distinct and a tying charge will be to no avail.”

116. It is clear, therefore, that the threshold question and a precondition to the application of the “tying” analysis is whether the representation services and the SPS provided by ILCU to its members are economically to be regarded as separate products.
117. The following passage from a textbook, O’Donoghue and Padilla, The Law and Economics of Article 82 EC (Hart Publishing) (2006), page 101, provides useful insights into the issue of distinctness of products:

"Effect on market definition: tying and pure bundling.
The effect of tying and bundling practices on market definition varies according to the type of bundling at issue. Consider first tying and bundling. The first key question in cases involving allegations of illegal tying and bundling is to establish whether A and B are "separate products" from the viewpoint of consumer demand or whether instead they should be treated as components of a single product. Two products can only be tied if they are genuinely distinct products. That is, when an independent product market exists for each of them; or, in other words, when there are separate product markets for both A and B.
118. As noted by Professors Areeda, Elhauge and Hovenkamp:
'However under the competitive market practices test, a distinct market for the tied item does not imply separate products absent widespread sales of the tying item in unbundled form. For example, an independent market for carburettors does not make a car with carburettor installed two products because no significant independent market exists for cars stripped of their carburettors. Nor does an independent market for television tubes prove that a television and its installed tube are separate products because we have no significant independent market for television lacking tubes. Two items constitute one product under the market practices test unless each could efficiently be sold without the other.'
That is, one cannot determine whether the bundle AB is a single product or the combination of two separate products by looking solely at the demand for product B. In fact, once it is established that B is a separate product, the relevant question is whether there is demand for A as a standalone product. Are there consumers prepared to pay a price to acquire product A with product B attached? If so, then A and B are separate products; otherwise there are two products AB and B, and A is just a component of the first of the two products.
A recent case that considered this issue is BT Analyst. In that case, the Office of Fair Trading (OFT) was concerned about a product (BT Analyst) which British Telecom (BT) was giving to multi-line customers free of charge. BT Analyst provides a retail telephony electronic bill provision of service. A rival company, Magic Telecom complained alleging that BT was attempting to foreclose the market. The OFT decided that BT Analyst did not constitute a separate product. Instead, it concluded that there was a single market for retail telephony services, which should be considered as a "cluster" and which included inter alia an electronic bill provision service:
'In a cluster market, consumers choose suppliers on the basis of the most competitively priced cluster of prices offered. Once a supplier is chosen on this basis, the consumer will purchase all products/services within the cluster from the chosen supplier. This means that purchasing decisions are not made on the basis of the individual prices of products. Consequently, the practice of "bundling" these services together is not in itself anti-competitive.'" (Emphasis added)
119. The three underlined phrases emphasise the need to establish the existence, from
the viewpoint of consumer demand, of an independent product market for each of the products in question, adding that there should be “widespread sales of the tying item in unbundled form.” The Authority, in its written submissions, criticises these authors for belonging to “the Chicago/Professor Bork school of anti-trust law.” It is suggested that their work relates to “refusal to supply jurisprudence” and that “they ignore the fundamental fact that this is a tying case.” However, the above passage is almost entirely concerned with “tying and bundling practices.” I find its analysis clear and compelling and reject the criticism of it advanced by the Authority. I also note that the unfavourable view held by the Authority does not appear to be shared by the leading English writer on competition law, Mr Richard Whish. In his review of the work of O’Donoghue and Padilla in Competition Policy International (Vol 2 N0 2 Autumn 2006), he described it as a work of “high class and great interest,” stating that “the law and economics are appropriately interwoven.”
120. This text does not, I accept, enjoy the status of a judicial decision. Its emphasis on the role of consumer demand is, nonetheless, entirely consistent with the long-established case-law of the Court of Justice. In Case 6/72 Europemballlage Corporation and Continental Can v Commission [1973] ECR 215 at paragraph 32, the Court emphasised “those characteristics of the products in question by virtue of which they are particularly apt to satisfy an inelastic need and are only interchangeable with other products.” (Emphasis added). The Court used almost exactly similar language in United Brands, cited above. (paragraph 22). The Continental Can test has been consistently applied both by the Court of Justice and the Court of First Instance, including in Case T-219/99 British Airways v Commission [2003] II-5917, upon which the Authority relies particularly.
121. Apart from cases concerning competing products or parallel markets, there are also many decisions of the Court of Justice resolving disputes about the existence of separate or distinct products. Foremost among these are Case C-333/94 Tetra Pak Rausing v Commission [1996] ECR I-5951 (also Case T-51/89 [1990] ECR II-309) (hereinafter Tetra Pak) (tying sale of machines to cartons) and Case C-53/92 Hilti AG v Commission [1994] ECR I-667 (also Case T-30/89 [1991] ECR II-1439 (hereinafter Hilti) (tying the sale of Hilti nails and cartridges to sales of its nail guns). At the risk of repetition, I believe the entire passage at paragraph 91 of the judgment in British Airways v Commission, cited above, is worth quoting:

“According to settled case law,…………for the purposes of investigating the possibly dominant position of an undertaking on a given product market, the possibilities of competition must be judged in the context of the market comprising the totality of the products or services which, with respect to their characteristics, are particularly suitable for satisfying constant needs and are only to a limited extent interchangeable with other products or services. Moreover, since the determination of the relevant market is useful in assessing whether the undertaking concerned is in a position to prevent effective competition from being maintained and behave to an appreciable extent independently of its competitors and, in this case, its service providers, an examination to that end cannot be limited to the objective characteristics only of the relevant services, but the competitive conditions and the structure of supply and demand on the market must also be taken into consideration.”
122. In the particular case, the issue was whether there was a separate market for air travel agency services. British Airways provided travel agents with commissions, discounts or other incentive schemes based on sales of tickets. The Court held (paragraph 93) that they “constitute independent intermediaries carrying on an independent business of providing services ……” The appeal by British Airways against that decision was dismissed by the Court of Justice on 15th March 2007 (Case p5/04 P British Airways v Commission). British Airways did not, on that appeal contest the aspects of the decision of the Court of First Instance concerning the definition of product markets.
123. The Commission, in its Microsoft decision (paragraph 802) explained the approach of the Court of Justice in these cases as follows:

“The Courts rejected these integrative approaches. In both cases, it pointed out that there existed independent manufacturers who specialised in the manufacture of the tied product, a fact which indicated that there was separate consumer demand and hence a distinct market for the tied product.” (Emphasis added)
124. It follows from all of the foregoing that the existence of SPS as an independent product must be assessed from the perspective of consumer demand. With this in mind, I turn to consider the applicability of these principles to the evidence regarding SPS.

SPS as an Independent Product: The Evidence
125. SPS is entirely the creature of the membership of ILCU, established under the control of ILCU. It consists of a fund contributed to and built up by the member credit unions over the years under rules, which expressly exclude any proprietary interest. It is a mechanism by which ILCU monitors and advises credit unions as to their financial management and operation, provides remedial help to credit unions which are weak in these areas and may provide discretionary financial assistance from the fund to an individual credit union, which is in need of it. By means of SPS, the members of ILCU collectively provide a stabilisation service to themselves. It has, as a matter of fact, assisted three credit unions. The purpose of the SPS is to provide assurance of stability and solvency to individual members (as depositors with credit unions), to the public generally and to the credit unions themselves. It is an essential and inescapable feature of the operation of the SPS that members submit themselves to a regime of monitoring and inspection of their financial affairs, books and records by ILCU.
126. All members of ILCU are members or subscribers to the SPS. No non-member has ever, prior to these proceedings, been permitted to have access to it. It is and always has been available only to its members. Thus, it constitutes to a significant extent a service provided by an association of credit unions to those credit unions themselves. It has never been offered or provided to any other person or body. No credit union, on cessation of membership has any claim on the assets of ILCU or the SPS.
127. Evidence was given by representatives of a number of credit unions who have ceased to be members of ILCU and who had since participated in the formation of CUDA in 2001. A fair summary of that evidence would be as follows. The departing credit unions were very concerned that their disaffiliation from ILCU would result in loss of access to the SPS. Some thought it “vital” or “a key issue” to have access to a stabilisation fund. Many of them, however, maintained that they had some type of proprietary or beneficial interest in that fund, though one fairly acknowledged that the rules excluded such a claim. Two witnesses spoke of a “double whammy” in the fact of losing an interest in the SPS, while having to pay for their own fund. Several were unwilling to submit to monitoring by ILCU. Almost all had as their real objective that the SPS fund should be put on an independent statutory footing and that it would be available to all credit unions. In effect, it would be taken out of the control of ILCU. They stated that there was no other fund on the market. The general effect of this evidence was that: (a) most credit unions which had left ILCU or were threatened with expulsion were dissatisfied at losing some inchoate proprietary interest in the SPS fund; (b) they had a preference for the fund to continue under independent control; (c) they were reluctant to embark on establishing a new SPS under CUDA auspices in the light of these considerations. It does not appear to me that any of them gave evidence that it was impossible or even extremely difficult to do so though it will not be necessary to reach a definitive ruling on that point.
128. It is common case that there is not and never has been any other stabilisation fund in Ireland other than the SPS. No such fund has ever been available for sale on a commercial basis.
129. Mr Przybelski, Vice-President of CUNA Mutual Insurance Group for Ireland and the United Kingdom and with extensive experience in the United States and world-wide, was a witness for the Authority. CUNA Mutual is a world-wide financial services and insurance enterprise. The objective of CUNA Mutual, a financial services company, is to make credit unions stronger financial institutions around the world. Mr Przybelski gave evidence that he had been asked, on behalf of CUDA to try to replicate on the market the specific type of SPS product, supplied by ILCU. That is the service which the CUDA members had been able to enjoy up to then. Following extensive inquiries, he discovered that neither his own company nor any company of similar size or greater provided such a service. Inquiries through CUNA’s reinsurance broker and via Lloyds produced a similar negative answer. Thus, neither his own company, CUNA mutual, nor any other company in the market provided or, it appears, was willing to provide, a stabilisation service such as the SPS. It simply was not there. This evidence seems to me to point very strongly to a conclusion that SPS is not a commercially saleable product.
130. This evidence gains support, moreover, from the history of credit union stabilisation services in the United States. Mr Michael Riley, a consultant in matters relating to credit unions, gave evidence on behalf of ILCU. He had spent 22 years with the National Credit Union Administration, a federal institution with responsibility for supervision and examination of credit unions both at federal and state level, some 8,000 in number. At one time there were some 30 to 40 stabilisation funds in the United States. These funds were operated by leagues of credit unions. None was provided by bodies other than leagues. They were “virtually identical” to the SPS at issue here. Access was confined to league members. About 1971, stabilisation services began to be replaced by a Federal Insurance Fund and stabilisation funds gradually died out.
131. I turn then to consider the evidence of Professor Seabright, the expert witness for the Authority. There is no significant difference of opinion between Professor Seabright and Mr Massey, ILCU’s expert economist on the principal elements of tying abuse or indeed on the applicable economic principles generally. However, they parted company on the crucial question of market definition.
132. To begin with Professor Seabright expressed the opinion that representation services are in a distinct relevant product market. It may not be absolutely necessary to reach a final conclusion on this issue, since the crucial point, for the Authority’s case, is whether SPS is a distinct product. Nonetheless, it is part of the Authority’s case that representation services constitute a distinct product. I will discuss it later. SPS is the alleged tying product; ILCU is alleged to be dominant in the market for its supply. It is an accepted precondition for the application of the tying theory that SPS constitute an independent product. Professor Seabright’s evidence on this point was quite short. He was of opinion that the economics of mounting a stabilisation fund were not really any different from those of an insurance fund. The purpose is to diversify risk. An individual credit union is not in a position to provide insurance against its own insolvency. In order to provide either a stabilisation fund or formal insurance, it is necessary to have a reasonably large number of credit unions. On all these points Professor Seabright’s evidence was in agreement with that of Mr Massey. He was asked by counsel for the Authority whether SPS was something bought and sold in an open way like any other commodity or service. From what I have said, it is clear that I consider it essential for the Authority to establish an affirmative answer to that question. In fact, Professor Seabright said:

“There is clearly just one monopoly provider of SPS cover in the State at the moment. ……it is not my understanding that the fact there is a monopoly supplier means you don’t have a relevant product market. The service is, I would say transacted in a relevant market precisely because the body offering it at the moment, the ILCU, is a distinct body from its individual members. We have heard this argument before and I think it is distinct in a very clear way so my answer to that would be yes.”
133. On the other hand, Professor Seabright was not convinced by Mr Massey’s view that SPS was an in-house service provided by the members to themselves, i.e., an example of vertical integration. He thought that the relevant analogy was that of a joint venture, which has a very large number of parent companies. I agree with the view of the learned trial judge on this issue. Counsel for the Authority rightly points to the distinctness of ILCU from its members. It owns its own assets (including the SPS fund). The member credit unions pay into the SPS fund and do not acquire any proprietary interest in it. However, Professor Seabright, when asked to answer the direct and central question of whether SPS was something bought and sold in an open way like any other commodity or service made no reference to any pre-existing evidence of such sales either in Ireland or anywhere else in the world. This is not surprising: the evidence showed clearly that there was no market. It is not necessary to decide whether SPS constituted a joint venture. If so, it is an unusual one, certainly in the sense of a piece of commercial co-operation designed to bring profit to the participants.

Conclusion on whether SPS is an Independent Product
134. I have reached the clear conclusion that the Authority has not established that SPS constitutes a distinct product, still less that it is in a separate product market. To consider that the representation services and SPS, which have always been provided to their own members, are distinct products is counter-intuitive. In my view it is artificial. The fact is that SPS has always existed as an integral part of the bundle of services that ILCU has provided to its own members and has never been provided independently. Monitoring and inspection of the finances of the member credit unions is the quintessence of the stabilisation fund arrangements. The member credit unions submit themselves to supervision and control by their fellows; they contribute to and build up the fund. Nobody has produced any evidence whatever of the existence of an independent commercially provided SPS anywhere. It is common case that it has never existed in Ireland. Mr Riley showed, without contradiction that many stabilisation funds had existed in the United States (prior to the establishment of federal insurance in 1971): all were provided by leagues of credit unions only to their members. Mr Przybelski showed that, not only did no such service exist on the insurance market, but that no insurer was willing to provide it.
135. It is not altogether surprising that the Authority has failed to provide a convincing analysis of ILCU’s activities as being anti-competitive. The history shows that it has changed its position in relation to ILCU on several occasions. It was permitted finally to change its stance from that advanced in the statement of claim only because Mr Collins decided not to object, believing that this radical change of position demonstrated the lack of credibility in the Authority’s case. It certainly seems to me to undermine confidence in the Authority’s consistency.

Evidence regarding events subsequent to High Court Judgment
136. I regard these considerations as decisive. However, ILCU has sought leave to file an affidavit, with a schedule showing that no credit union has sought to avail of the right conferred on all credit unions by the High Court order to have access to the SPS. The Authority objects to the use of this evidence and argues that the facts deposed to are “unexceptional, given the fact that all such credit unions would doubtless have been aware of the existence of this appeal.” While this evidence could be admitted pursuant to the principles mentioned by thus Court in Fitzgerald v Kenny [1994] 2 I.R. 383, it does not add anything significant to what has already been clearly demonstrated, namely the absence of an independent product market for SPS and I prefer not to rely on it.

Consideration of Representation Services as an Independent Product
137. It is not strictly necessary, in these circumstances, to consider whether the credit union representation services provided by ILCU consist of a distinct product in the economic sense, or whether, to use the language employed in the High Court, they are tradable services at all. On the basis of the adjusted case made by the Authority, it was first necessary to determine whether SPS, being the alleged “tying” product, was in a separate product market. Once it is not, it is not strictly necessary also to determine the character of the representation services. Nonetheless, the entire case for the Authority is that these are economic services provided by ILCU. ILCU abuses its dominant position in the SPS market by refusing to supply SPS unless the buyer also purchases what it characterises as its representation services. It thereby forecloses the market for the latter.
138. I confess to finding it troubling that any and every association of business undertakings should be held, for the purposes of competition law, automatically to be engaged in a business consisting of the provision of services for reward. That, in reality, was the contention advanced persistently on behalf of the Authority in the cross-examination of Mr Massey. Mr Massey consistently declined to accept that there was a market for the provision of representation services. That activity seems to me more naturally to meet the description of common pursuit of common interests. A members’ association, like a members’ club is simply that it is not in the business of supplying the service of “membership.” When invited to comment on this issue at the hearing, Mr Gallagher cited a decision dated 17th September 2001 of the Appeal Tribunal of the (UK) Competition Commission, (Sir Cristopher Bellamy being President) in The Institute of Independent Insurance Brokers v Director General of Fair Trading. (Cases Nos. 1002/2/1/01(I.R.); 1003/2/1/01; 1004/2/1/01.) In that case, the Tribunal held that there was a market for the provision of regulatory or certification services to independent insurance brokers as an alternative to, and in competition with one established by the General Insurance Standards Council. The principal reason for that conclusion (paragraph 232 of the decision) was that one way in which independent brokers are able to compete with insurers’ direct selling operations is by promoting themselves as offering higher quality standards of professional competence and consumer protection. That is not, to my mind, at all the same as saying that there is a market for representation services of the sort provided by ILCU and, no doubt, many other representative bodies.
139. One of the most readily foreseeable implications of the Authority’s analysis is that any trade association representing a significant percentage (which need not even amount to 50%) of a particular trade or profession might be held to enjoy a dominant position in the market for representation services for the trade or profession in question. A person denied admission to membership of the association, for whatever reason, might plausibly mount a case of abuse of a dominant position based on refusal to supply the association’s services. Naturally, proper analysis of the conditions of competition in the market would have to be carried out. Nonetheless, it seems at least conceivable that competition law could be deployed in pursuit of complaints of disappointed applicants for membership.
140. Moreover, Mr Massey, in his evidence in the High Court made the telling point that, if it were to be assumed that there was a market for credit union representation services, one would expect that there would be, in that market, suppliers of representation services generally. While there are very many representative organisations for various trades, industries and professions, in fact no such market exists in fact. In cross-examination, Mr Massey defended this view tenaciously, saying that the situation, “overwhelmingly is that representative services, however one chooses to find them (sic) [presumably meaning “define”]…… are provided by representative organisations comprising their own members.” He said: “All sorts of organisations have concluded that it is not possible to buy adequate representation services in the market… that is why there is no market.” This seems to me to represent common sense observation. Certainly, no evidence was presented of the existence of any market consisting of sellers and buyers of representation services.
141. It was not suggested, on behalf of the Authority that there was a market, but rather that it was possible, in principle, for such a market to exist and that credit union representation services are capable of being a distinct product market.
142. For reasons already given, it is not necessary to form a concluded view on this issue. It is preferable to leave this issue over for another day. If it is to arise in another case, I would hope that the Authority would produce cogent factual evidence of the existence of such a product market in representation services. It was not debated with the intensity of the SPS issue.

Conclusion
143. If SPS and representation services are not distinct products in distinct product markets, the Authority’s case for tying abuse fails. I do not, therefore, need to consider the issue of dominant position or its abuse, foreclosure or justification. For the sake of clarity, I would add that there can also be no question of an abusive refusal to supply.
144. Finally, the Authority’s case under section 4 of the Act of 2002 fails for the same reasons. The learned trial judge found in favour of the Authority under section 4 “for much the same reasons” as he had decided under section 5. By parity of reasoning, I conclude that the Authority’s case fails under section 4.
145. I would allow the appeal, set aside the order of the High Court in its entirety and substitute an order dismissing the claim of the Authority.