---

COMPETITION TRIBUNAL

REPUBLIC OF SOUTH AFRICA

Case no: 41/AM/Jun02

In the matter between:

Digital Healthcare Solutions (Pty) Ltd Applicant

and

The Competition Commission 1^st Respondent

Healthbridge (Pty) Ltd 2^nd Respondent

(intervenor)

Reasons for decision

Introduction

1. This case concerns the consequences of a putative breach of conditions imposed by the Competition Commission (“the Commission”) in respect of a merger approved by the Commission in 2001, subject to those conditions, as a small merger.

2. At issue is the validity of a notice of apparent breach (“the notice” or “the notice of apparent breach”) issued in regard to those conditions by the Commission on 28 May 2002 under rules 39 and 40 of the Rules for the Conduct of Proceedings in the Competition Commission (“the rules of the Commission”).

3. The recipient of the notice was Digital Healthcare Solutions (Pty) Ltd (“DHS”), which is the applicant in these proceedings. There are two respondents. The first is the Commission. The second is Healthbridge (Pty) Ltd (“HB”), which was authorised by this Tribunal in an earlier application to intervene as a respondent.¹

4. The background is briefly as follows.

Background events

5. The merger in question took place in terms of an agreement concluded in November 2000 when a company then named Persetel Q Data Trading (Pty) Ltd (“Persetel”) sold a part of its business, known as QEDI and conducted through a company named QEDI (Pty) Ltd, to a competitor, Mediswitch (Pty) Ltd (“Mediswitch”). Persetel has since changed its name to Comparex Africa (“Comparex”).

6. The QEDI business was absorbed into that of Mediswitch, which has changed its name and is now known as Digital Healthcare Switch (Pty) Ltd (“DH Switch”). DH Switch is a wholly owned subsidiary of DHS. Persetel received shares in DHS as the consideration for its sale of the QEDI business.

7. At the time of the merger Mediswitch was a subsidiary of Bytes Technology Group Ltd (“BTG”), a listed company once known as USKO, which is in turn a subsidiary of Allied Electronics Corporation Ltd (“Altron”). Altron is also a listed company.

8. Since the merger DHS has been owned by BTG and Comparex. DHS is a holding company, which operates DH Switch alongside another wholly owned subsidiary named Med-e-Mass (Pty) Ltd (“MM”). MM conducts the business of developing, licensing and maintaining software used by doctors and other medical service providers and known as practice management software or practice management applications (“PMAs"). PMAs help doctors and other suppliers of medical services to organise their business operations and to prepare medical scheme claims for onward transmission.²MM was in existence and in operation at the time of the merger.

9. The business of DH Switch is that of electronically transferring or “switching” information about patients’ claims between, on the one hand, doctors and other providers of medical services and, on the other hand, the administrators of medical schemes of which the patients are members. In this way the claims can be quickly processed and settled by the medical schemes, which are known in the industry as “funders”. Switching of information in this manner is a phenomenon of the computer age, supplanting the transmission of documents and the consequent paper-handling.

10. To allow switching to take place between switch entities such as DHS and the PMAs used by doctors, a so-called interface, known in the industry as an application programming interface (“API”), is necessary. The switch entities design and install APIs, for which they require access to the source code of the PMAs.³In order to gain access, a contract is entered into between the switch entity requiring access and the entity owning the PMA software. The contract covers, inter alia, the development cost and cost per site.

11. Two forms of switching were debated in the case. The first is known as batching, and refers to the habit of some medical providers to wait till the end of a working day and then switch all their claims in a batch. This is an earlier form of technology. The second, a more recent development, is real-time switching, by which medical providers switch every claim at the time when it arises rather than accumulating the claims for transmission in batched fashion at the end of a day, and moreover receive an almost immediate response from the medical scheme’s administrator. The two forms of switching require somewhat different versions of the relevant technology. There are also hybrid versions of switching technology.⁴

12. Switching technology came about through the investment of relatively large sums in past years. Mediswitch and Persetel were pioneers in this technology in South Africa and they incurred relatively large investment expenditures of this kind.

13. Before the merger between QEDI and Mediswitch, QEDI controlled approximately 95% of the switching market and Mediswitch 3%.⁵This led to a post merger market share of more than 95% in the switching market. At the same time, Mediswitch had a dominant market share of between 60% and 65% of the PMA market.⁶The net effect of the merger was that DHS became a dominant player in the switching market through DH Switch as well as in the PMA market through MM.

14. HB is another switch entity. Two of its three main shareholders, each with a holding of some 29%, are large administrators of medical schemes. They are known as Medscheme and Discovery Health. A third shareholder with a similar holding is Dimension Data Ltd (“Didata”), a supplier of information technology services. Medscheme and Discovery Health are the two biggest administrators in the market.

15. HB does not supply PMA software.

16. It is a characteristic of certain contractual arrangements prevailing between HB and its shareholders, or some of them, that Medscheme and Discovery Health only permit HB, and not any other switch entity, to transmit claims made in real time to them.

17. In summary therefore, we have a situation where post merger the dominant PMA vendor and the dominant switch provider reside in DHS. HB, its major rival in the switching market is not in the PMA market and has to enter into access agreements, so called front-end access, with PMA vendors to enter this market. On the other hand HB enjoys the advantage of real-time access with the two dominant medical scheme administrators, Medscheme and Discovery, who collectively hold approximately 60% of the equity in HB. This so-called back-end access is only possible if the switch provider has an agreement with the relevant medical scheme administrator.

18. The merger had not originally been notified by the parties to the Commission, but after its implementation the Commission required notification in terms of s. 13(3) of the Competition Act (“the Act”).⁷

The Commission’s conditions for the merger and reasons for its decision

19. Attached to the Commission’s merger clearance certificate, dated 9 April 2001, were the following reasons and conditions:⁸

“The proposed transaction raises significant competition problems. The parties involved are in the business of conveying claims electronically between medical practitioners and healthcare funders. The parties’ market shares, post-merger, for the electronic conveyance of claims for medical practitioners will be high. Furthermore, the merger raises vertical integration concerns, more fully discussed below.

In order for the Electronic Medical Claims Switching process to work, private medical practitioners require specific Practice Management Software (“PMS”) to be loaded on their computer systems as well as certain functionality to be built into their systems, in the form of an Application Program Interface (“API”). Med-e-Mass (Pty) Ltd (“Med-e-Mass”), a subsidiary of Mediswitch controls a significant portion of the PMS market. Furthermore, Mediswitch is already vertically integrated and Mediswitch has allowed QEDI to interface with its PMS.

All independent PMS systems are fully integrated with competitors in the switching segment that require the transportation of claims from private medical practitioners to the healthcare funders. The PMS package controlled by Mediswitch group, (Med-e-Mass), is only allowed to integrate with Mediswitch and QEDI. The vertical integration concern that the Commission has with this transaction is that there has been foreclosure by Mediswitch even prior to the proposed merger. If this is tied to the switching service rendered by QEDI, which will be acquired as a result of the proposed transaction, it is believed that the transaction will give rise to a substantial lessening of competition in the electronic claims switching market for private medical practitioners.

This will have a negative effect on the choice that the end consumer, which in this particular case is the private medical practitioner, will have with regard to the type of electronic switch he/she is allowed to interface.

However, the Commission is of the view that the conditions set out below would significantly reduce those concerns. The merger is, therefore, approved, subject to the following conditions:

1. The merged entity shall, on reasonable written request by any healthcare switch entity, integrate the applicable latest versions of PMS packages which it owns or controls, with an API which enables an interface with the switching technology of the healthcare switch entity requesting such integration, in accordance with an agreement referred to in paragraph 2 below.

2. Pursuant to such a reasonable written request, the merged entity shall use all reasonable endeavours to conclude a written agreement with the requesting healthcare switch entity concerned, within a period of 60 days after receiving such request, containing commercially, financially and technically reasonable terms.

3. The merged entity shall provide a quarterly report to the Commission, for a period of 12 months after the date of approval, detailing all requests by third party healthcare switch entities to integrate their API and functionality with the PMS packages owned or controlled by the merged entity, as well as detailing the agreements and time frames concluded with such third party healthcare switch entities in respect of the integration process.”

20. The meaning and interpretation of these conditions (“the conditions”) lay at the centre of the debate in this case.

Relief sought by DHS

21. The prayers in DHS’ notice of motion, as amended in the course of the proceedings, are for orders as follows:

1. The Notice of Apparent Breach issued by the First Respondent against the Applicant dated 28 May 2002 be set aside.

2. It is declared that the Applicant and its subsidiaries have complied, alternatively substantially complied, with the merger conditions set out in the First Respondent’s merger clearance certificate dated 4 April 2001.

3. It is declared that the Applicant and its subsidiaries are not obliged to continue to negotiate and conclude an agreement with the Second Respondent for the integration of the latest versions of practice management software packages which they own or control, with an application program interface which enables an interface with the switching technology of the Second Respondent.

4. As an alternative to prayer number 3, it is declared that the merger conditions set out in the First Respondent’s merger clearance certificate dated 4 April 2001 be amended to include the following sentence at the end of the second condition: ‘including, without limitation, that such other healthcare switch entity shall cancel any agreement which it may have with healthcare funders or their administrators, relating to the exclusive use by such funders or administrators of the switching technology of such healthcare switch entity for the electronic conveyance of claims, whether by batch or real time.”

22. There was also a prayer for costs.

23. Prayer 4 was abandoned at a hearing on 21 November 2003 and no more will be said about it.

24. The amended notice of motion stated that the application was brought in terms of rule 39(2)(b) of the rules of the Commission, read with rule 42 of the Rules for the Conduct of Proceedings in the Competition Tribunal (“the rules of the Tribunal”); alternatively the same rule 42 of the rules of the Tribunal, read with s. 27(1)(c) of the Act.

25. Voluminous papers were filed on behalf of DHS and HB, and there were also substantial affidavits and annexures on behalf of the Commission. Relevant passages will be referred to below.

Preliminary observation on the issue of jurisdiction

26. In its papers DHS has raised two preliminary issues for consideration. These are that the notice of apparent breach has been served on the wrong party and secondly and alternatively that the Commission had failed to apply its mind properly to the question of whether there had been a breach of the conditions before issuing the notice because the Commission entirely misconceived the conditions. It was agreed between all parties that these two issues could best be disposed of by being argued as points in limine.

27. DHS had originally framed its application in terms of rule 39(2)(b) of the rules of the Commission. That rule states:

Within 10 business days after receiving a Notice of Apparent Breach, a firm referred to in sub-rule (1) may –

1. submit to the Commission a plan to remedy the breach; or

2. request the Competition Tribunal to review the Notice of Apparent Breach on the grounds that the firm has substantially complied with its obligations with respect to the approval or conditional approval of the merger.

28. The review contemplated by rule 39(2)(b) of the rules of the Commission is however not an untrammeled review having its roots in the common law but is confined to the enquiry whether there has been compliance with the terms on which the merger was approved by the Commission.

29. DHS seems to appreciate that Rule 39(2)(b) is too narrow to encompass the points in limine that it seeks to raise. For this reason DHS amended its notice of motion, as we indicated above, so that it could rely on other procedural provisions to found this attack on the Commission’s notice of breach. Indeed counsel for DHS stated in their heads of argument, referring to the original notice of motion that had relied on rule 39(2)(b), that it was:

“ technically too narrow to permit of a decision on the two points in limine.”⁹

30. Thus what DHS has sought to achieve through the amendment is to found jurisdiction for the Tribunal to entertain two points of law that relate to classic grounds of review in addition to the power we enjoy under rule 39(2)(b). We have some doubts about whether it has been able to achieve that. Nevertheless our jurisdiction to consider the points in limine has not been challenged by the two respondents. ¹⁰

31. For the purposes of this case we have assumed in DHS’ favour that we have adequate powers of review to hold the necessary enquiry and issue such an order. This should not be regarded as a precedent for other cases involving comparable claims for relief going beyond the review contemplated in rule 39(2)(b) of the rules of the Commission.

The Commission’s notice of apparent breach

32. The notice, set out on form CC19 as specified in the rules of the Commission, was addressed to Mr HW Du Plessis, Digital Healthcare Solutions (Pty) Ltd (Mr Du Plessis is the chief executive officer of DHS) and contained a reference to case no. 2001Mar13, that of Mediswitch (Pty) Ltd and QEDI (Pty) Ltd.¹¹The reasons for the apparent breach were stated as follows:

“The Commission’s conditional approval of the merger specifically provided that DHS should conclude an agreement with any party that initiated negotiations with it within sixty days of receipt of a written request by the aforesaid party. In the instant case, DHS received a formal request from Healthbridge on 4 June 2001, erroneously dated 4 July 2001. In terms of the Commission’s conditions, DHS should have concluded the agreement with Healthbridge on or before 3 August 2001.

DHS has been in breach of these conditions since 4 August 2001. The Commission had allowed DHS to continue with its negotiations with Healthbridge in the hope that DHS and Healthbridge would conclude an agreement before the 23^rd of May 2002.

In light of DHS failing to conclude a contract by the 23^rd of May 2002 and its shareholders insistence that Healthbridge’s shareholders, Discovery and Medscheme, give DHS back end access to their systems in order to reach an agreement represents a breach in the conditional approval and this has necessitated that the Commission issue the CC19.“

DHS’ points in limine

33. The two points in liminewhich had been raised by DHS were argued at a hearing on 20 August 2003.

34. In the first, DHS asserted in effect that the Commission had cited the wrong party in the notice. DHS claimed that the error lay in the fact that the notice had been addressed to DHS whereas the merged entity comprised Mediswitch (Pty) Ltd, now DH Switch, which had absorbed the business of QEDI (Pty) Ltd. DHS was not, it contended, the merged entity.

35. Several factors were relied upon by DHS in support of this objection.

36. First, it was pointed out that the Commission’s notice requiring notification of the merger was directed to Mediswitch and QEDI (Pty) Ltd, not to DH Switch’s holding company, DHS. The merger was notified by Mediswitch and QEDI (Pty) Ltd, not by DHS. The conditions for approval of the merger were moreover directed at Mediswitch, not DHS. The notice was, in DHS’s submission, correctly addressed to DHS as the “spokesman” or “messenger” for DH Switch, and correctly referred to the original parties to the merger, but wrongly alleged a breach by DHS. DHS, Mr Burger contended, was “never in a debate” with the Commission. As a matter of company law, he contended, there had been an error. He contended that the case of Dadoo v. Krugersdorp Municipal Council 1920 AD 530 at 550-1 was in point in ruling that a company is a separate entity from its shareholders and that property vested in the company cannot be regarded as vested in the shareholders.

37. The evidence showed, said Mr Burger, that DH Switch and MM are under the control of different chief executive officers and boards for purposes of their day-to-day operations. The Commission’s merger clearance document had wrongly identified Mediswitch as the entity which controlled the PMAs, whereas the correct entity was MM. The Commission’s merger clearance document and notice of apparent breach had been directed specifically to named companies and not to an “acquiring firm” in terms of the definition of that term in the Act, which would have covered a wider group than merely the companies named in these documents.

38. The “merged entity” to which the Commission referred in its documents could only be DH Switch and not DHS. He refuted the suggestion that, since the evidence showed that the conditions for the merger had been drawn up collaboratively by the Commission and the merging parties, his argument about the non-involvement of DHS in the activities alleged by the Commission to amount to a breach had the implication that the Commission might have been given incorrect information about the circumstances of these entities by one or more of the relevant companies at the time when the merger was approved, and this might not have led to approval of the merger subject to the particular conditions imposed. All that the conditions required, he contended, was that DH Switch use its best endeavours or reasonable endeavours to get its fellow subsidiary, MM, to comply with the conditions. That fact that MM might be “difficult” was not to be blamed on the “sometimes competing” entity, DH Switch.

39. The Commission’s attitude to this point in limine, advanced by Mr Coetzee, was that it was irrelevant how the acquiring and target entities of the merger organised themselves internally, and the Commission was entitled to address the complaint about an alleged breach to DHS as the holding company of the entity which had committed the alleged breach. DHS was the “controlling mind” of the two entities, DH Switch and MM. The Commission, he said, looked to the entity which could give effect to the conditions. If it had expected MM to on its own give effect to the conditions it would not have issued the conditions in the wording in which they were in fact issued.

40. HB’s approach, argued by Mr Unterhalter, was that DHS was on its own evidence, as set out in a supplementary affidavit by Mr Du Plessis, the entity which represented DH Switch and MM in matters of strategic importance, such as in “dealing with lawyers” and “ensuring compliance with regulatory authorities”. On this basis the regulatory entity to be addressed by the Commission was DHS. If the objection to the notice had been anything other than opportunistic, the correct entity to have brought the review application would have been DH Switch, not DHS, since DH Switch would have considered the obligations of the conditions to rest on it, not on DHS.

41. However, said Mr Unterhalter, DHS had understood fully that obligations contained in the conditions had rested on it since DHS had substantively sought to answer for the merged entities in the review application under Rule 39. In any case, DH Switch (then Mediswitch) was never in a position to control, and did not own, the PMAs referred to in the conditions, so other entities have obviously been implicated from the start. DHS had at the relevant time been named MediSwitch Holdings (Pty) Ltd, so the term “Mediswitch” in the Commission’s merger clearance document and in the notice of apparent breach was always “potentially an ambiguous term”, and what was clearly intended in the notice was a group of companies aligned together for the purpose of the deployment of the PMAs.

42. It was clear from the reference to “control” in the conditions, said Mr Unterhalter, that what was being referred to was a group of entities going wider than DH Switch. In that group, DHS exerted control in the classical sense of owning the shares in the operating companies. All these companies were implied in the term “the merged entity” as used in the Commission’s merger clearance document and in the notice of apparent breach. This was confirmed by the fact that DHS had supplied the Commission with quarterly reports on the process of integrating other switch entities, and had used in relation to its reports the words “in compliance with the third condition set out in the merger clearance certificate.”¹²This, it was contended, was a clear admission that the merged entity included DHS.

43. Mr Unterhalter further referred to the definition in the Act of “acquiring firm”, which includes the upstream and downstream associates of the parties to a merger, and submitted that the term “merging party” had to be interpreted in the same context, wider than the identity of the individual company which was a party to the merger agreement. The key point in the conditions was the control of the software, and it would have been inappropriate and illogical to impose the conditions only on Mediswitch, which did not technically control the PMA software. Consequently, it was artificial and wrong, HB contended, to interpret the conditions and the notice in the manner argued for by DHS.

44. The Tribunal is satisfied that the stance of the Commission and of HB on this issue is correct, and that the reasons advanced by them in their arguments are essentially sound. We consider that the conditions imposed by the Commission must be read as referring to and being addressed to entities, which could give proper effect to the conditions.

45. The Commission, as noted, imposed the conditions in its merger clearance decision on the “merged entity”. The words “merged entity “ are not used in the Act, which refers instead to a “party to a merger”. The Act defines a “party to a merger” as an “acquiring firm” or “a target firm”. These two concepts are further defined but as these definitions are lengthy they need not be reproduced here. It suffices to say that DHS would meet the definition of an acquiring firm in the Act. It follows that this would make DHS a “party to the merger” as defined, and thus subject to the consequences in the Act for firms of that status – inter alia, the duty to notify a small merger when so required by the Commission under section 13(3) and the duty to comply with conditions imposed on that merger (section 59(1)(d)(iii). The question then is: when the Commission imposed the condition on what it termed the “merged entity” did it restrict itself only to the legal entity that absorbed the target firm (DH Switch)? In our view both the legal regime created by the Act for merger notification, which contemplates a wide range of firms as parties to the merger, as well as the logic of the condition, suggest that DHS was contemplated within the notion of “merged entity”.¹³(Would the Commission have intended to impose a condition on a firm incapable of implementing it?)

46. In the circumstances of the group structure of DHS, DH Switch, and MM, and the changes in name and shareholding which took place in these companies as a result of, or after, the merger, this means that these companies are included in the concept of the merged entity affected by the conditions. It seems clear from the evidence that this was understood by DHS in the same way. In the post-merger activities of relevance in this case those entities were represented by DHS in dealings with the Commission and with HB. There is uncontradicted evidence by Mr Van Zyl, the CEO of HB, that at the very outset of HB’s negotiations to secure integration for HB with the PMAs supplied by MM he approached the CEO of DHS, Mr Du Plessis, who insisted that the negotiations proceed with him rather than with the managing director of MM, and that he (Du Plessis) had the necessary authority to take the matter further.¹⁴

47. It is also clear to the Tribunal that the point in limineis a comparatively novel weapon in DHS’ armoury. In a letter dated 28 August 2001 (p. 96 of the record) sent by Du Plessis on behalf of DHS to HB – clearly a considered and legalistic letter – Du Plessis stated unequivocally that “DHS is acutely aware of its obligations in terms of the ruling issued by the Competition Commission in Notice 919 of 2001 and will comply with that ruling to the letter.” Notice 919 contains the Commission’s reasons for approving the merger and its conditions for the approval.

48. It lies ill in the mouth of DHS and its counsel to argue at this stage that DHS was somehow not affected by the conditions and was wrongly addressed by the Commission when it issued the notice of breach.

49. Consequently, the first point in liminefails.

50. The second point in limineconcerns the approach adopted by the Commission when it considered whether there had been compliance with the conditions. DHS contended that the Commission had been inconsistent in that, while the conditions stipulated that the merged entity should use all reasonable endeavours to conclude the agreement contemplated in the conditions, the notice of apparent breach reflected a decision postulating an absolute obligation on DHS to conclude an agreement with any party initiating negotiations within 60 days from receipt of the request to negotiate, regardless whether such request was reasonable.

51. DHS argued that the Commission had been obliged to consider the reasonableness of the parties at three separate stages. First, when a written request was made to the merged entity, the request had to be reasonable. If the request survived this test, the second stage required the merged entity to use all reasonable endeavours to conclude an agreement within 60 days of receiving the request. There was no rigid requirement to conclude an agreement; merely to use such endeavours to conclude an agreement. Finally, the reasonableness was to apply only in regard to agreements containing commercially, financially and technically reasonable terms. The obligation to conclude the agreement would arise only if an offer were made to DHS that contained terms that were commercially, financially and technically reasonable.

52. The Commission, so DHS contended, had merely taken the fact that no agreement had been signed by 23 May 2002 to be a breach of the conditions, and did not consider the reasonableness of both parties at each stage mentioned above. It therefore did not apply its mind to the real issues.

53. The evidence shows that HB directed a request to DHS for an integration agreement in a faxed letter of 4 June 2001 (erroneously dated 4 July 2001), sent by Van Zyl to Du Plessis. This followed the telephone call from Van Zyl to Du Plessis mentioned above in which Van Zyl had broken the ice on the issue of the proposed integration agreement and had established that Du Plessis, and not the CEO of MM, was the person with whom HB was to negotiate.

54. Enclosed with HB’s letter of 4 June 2001 was a proposal for an integration agreement, in the sense of an agreement to connect HB via an API to the PMAs installed by MM with medical service providers, and also a proposal for a so-called interswitch agreement, to which reference will be made later. The proposal dealt at some length with the technical issues, as well as others, and it is accepted by DHS that the technical proposals were reasonable. There is a dispute about the reasonableness of the financial and commercial aspects. DHS did not, it seems, submit a detailed appraisal of HB’s proposals but instead, on 18 June 2001, sent HB a set of counterproposals for the integration agreement. Significantly, they contained nothing related to the interswitch issue. It is clear to the Tribunal that DHS did not at that stage, and in fact in several rounds of negotiation that followed in the ensuring three months or more, raise the issue of interswitch, or link its proposals for an integration agreement to any form of interswitch agreement. That came later.

55. There is a lengthy account in the papers of the negotiations that followed. The negotiations became acrimonious and accusations of unseemly conduct appear to have been exchanged. In about September 2001 DHS came forward with the suggestion that DHS and HB might merge, and for some five months the two groups concerned engaged in a discussion about this possibility. It led nowhere since in the end it was concluded that there were insuperable competition objections to such a merger. Complaints were made on both sides about allegedly unreasonable requirements which were being raised by their counterparts. The negotiations were foundering.

56. HB approached the Commission and alerted it to the state of the negotiations. DHS lodged a complaint with the Commission that there was anti-competitive activity in the HB camp in that Medscheme and Discovery Health would not allow DHS the access to their systems which it needed to be able to switch claims in real time directly to those administrators – so-called “back-end access”. This is the interswitch topic referred to above.

57. The negotiations were protracted, intense, and complex. Several hundreds of pages of documents detailing the negotiations and what they encompassed were filed. It is not easy for the Tribunal to summarise them adequately.

58. A few conclusions have however been drawn by the Tribunal from this welter of documents and the large amount of information they contain. These conclusions may be summarised as follows:

1. The Commission was told at numerous stages during the negotiations of the progress, or lack of progress, which had been achieved. The Commission was in touch with DHS and HB and received reports independently from both of them. It considered the details of the proposals, which were exchanged at the time when these reports were made. The Commission’s investigator in charge of the matter, Mr Chetty, was in touch with both sides by telephone and in other ways and actively followed the development of the negotiations. It is clear that he was applying his mind to the issues under discussion between DHS and HB.

2. It was not inattentiveness or indifference which governed the Commission in allowing the 60-day time limit stated in the conditions to be overtaken, but rather the belief that there was a prospect that the negotiations would succeed and would result in an agreement which would make further action by the Commission unnecessary.

3. The Commission was aware of the restrictions imposed by Medscheme and Discovery Health on back-end access and examined these restrictions through the lens of anti-competitiveness, but was not perturbed by them. In any case the Commission regarded this issue as extraneous and irrelevant to the question of integration of the HB switching technology with the PMAs supplied by MM.

4. The proposal for a merger between DHS and HB, or parts of their businesses, took up an inordinate amount of time and was ill-considered in the light of the pressing time constraint on the merged entity -- in practice on DHS -- to conclude an integration agreement.

5. While the Commission was content to allow the negotiations to continue far beyond the stipulated 60-day deadline, it was clear to all concerned that the Commission was not abandoning the conditions. The day of reckoning for non-compliance with the conditions, if no integration agreement emerged from the negotiations, was merely being delayed, in DHS’ favour.

6. Another switch entity, MediKredit, applied to DHS for an integration agreement and concluded such an agreement. It did so on terms, which it regarded as highly unfavourable to itself, which it had been forced to accept as it had no alternative.¹⁵The Commission was aware of MediKredit’s position and took account of it in its consideration of the status of the negotiations between HB and DHS.

59. In the light of its conclusions on these matters the Tribunal is not persuaded that the Commission failed to apply its mind to the reasonableness of the negotiations in any significant way, and certainly not in the three stages posited by DHS. It is correct that some passages in the notice of apparent breach suggest that the Commission was somewhat abrupt about its consideration of the issues and its decision finally to issue the notice. The notice must however be read as a whole. The second paragraph of the notice refers explicitly to the fact that it is the failure to conclude the contract and the insistence on back-end access in order to reach an agreement that represents a breach of the conditional approval. Thus the Commission does not confine itself to the failure to reach an agreement as constituting the breach, but couples this failure with the insistence on back-end access. Furthermore there are other passages in other documents from which it is clear that the issues were weighed in a careful and even-handed manner. In their references to individual passages from the papers DHS’ counsel were selective rather than systematic.¹⁶

60. For example, DHS’ counsel made much of a statement in paragraph 41 of the affidavit of Mr Chetty, who testified on behalf of the Commission, that:

“The proof of reasonableness in these negotiations was the fact that no agreement was concluded between a willing seller and a willing buyer.”

Mr Burger’s comment was that:

“That’s obviously reviewable. They haven’t applied their minds to it.”

However, Chetty’s statement, which read in isolation, certainly tends to suggest that the Commission concerned itself more with the outcome than the nature of the negotiations, is immediately preceded by the statement that:

“It has always been the Commission’s position that the end result of these endeavours and negotiations should be a reasonable and mutually acceptable agreement.”

It is clear from this passage that the Commission’s stance at the end of the day was that the nature of the negotiation process and the content of any agreement, which might have been reached were important, not that the only factor it need take into account was the non-conclusion of an agreement.

61. In paragraph 42 of the same affidavit, Chetty moreover makes the statement that:

“It is submitted that what is at issue is the manner in which the negotiations were being conducted.”

Mr Burger’s view on this statement was that it did not explain what the Commission had considered, and from this he proceeded to the conclusion that the underlying issues had not been addressed at all by the Commission. In replying argument, Mr Campbell dismissed this statement as “a throw-away line”.

62. The above-quoted statements of Chetty are of course ex post hoc statements. We did not hear from DHS’s counsel any discussion of paragraphs 39 to 55 of Du Plessis’ founding affidavit and the annexures to it relevant to those paragraphs. In these paragraphs (at pages 13-18 of the record) Du Plessis describes various stages in the negotiations with HB.

63. Du Plessis mentions in paragraph 43 that Chetty telephoned him on the 6^th of March 2002 and revealed that HB had lodged a complaint with the Commission about the lack of an integration agreement. Du Plessis testifies about a number of other topics discussed in that conversation, all being at that time recent developments in the negotiations. Chetty clearly played an active role in this conversation. He was taking an interest in the status of the negotiations and was receiving and conveying information about them. He was obviously giving consideration to the progress being made towards an agreement, and taking account of various submissions of the parties to the negotiation process.

64. In paragraph 45 an account is given of a meeting, which Du Plessis and DHS’ attorney had with three representatives of the Commission, Chetty being one of them. It was clearly a detailed report-back meeting at which DHS’ presentation of the facts did not satisfy the Commission’s team.

65. Further telephone conversations, which Du Plessis had with the Commission about various features of the negotiation process are explained in paragraphs 47 and 55 of Du Plessis’ affidavit. In paragraph 48 Du Plessis describes the receipt of a letter from the Commission dated 9 May 2002 in which the Commission sternly warns DHS of its concern that DHS’ insistence on back-end access to Medscheme and Discovery Health was regarded by the Commission as a failure to comply with the terms of the merger approval.

66. These passages from the papers received no attention in DHS’s submissions at the hearing. They clearly undermine entirely DHS’ protestations that the Commission did not apply its mind to the issues. In any event, the issues as explained in DHS’ submissions were clouded by DHS’ analysis and interpretation of the conditions, which are dealt with more fully in a later part of this decision and which are rejected by the Tribunal.

67. The Commission conceded at the hearing that it might have phrased some of its documents more satisfactorily, but the lapses of which we are aware are minor and are self-evident and do not, in our view, outweigh the overall impression that the matter was approached in a fair and attentive manner by the Commission.

68. It is also artificial to approach the Commission’s discretion in this case in a static manner, as if it heard all the facts on a once off basis and then made a decision. The record shows that the Commission continuously monitored the implementation of the conditions; it received quarterly reports from DHS and had interaction with other players including HB. Its decision is thus formed on the basis of what it was experiencing over a long period of time. Had it issued its notice of apparent breach on the 61^stday after HB had formally requested the conclusion of an integration agreement, DHS’ arguments might have had more force. The fact that it allowed the parties further time to negotiate shows that the Commission was at all times alive to the more nuanced interpretation of the conditions for which DHS contends.

69. It is also worth noting that in terms of the rules of the Commission, the explanation accompanying the notice of apparent breach is intended to be a succinct explanation of the errant conduct rather than a carefully reasoned judgment. ¹⁷

70. HB’s approach to this point in liminewas that issuance by the Commission of the notice of apparent breach was merely an interim step in a decision-making process and was not the kind of action which was reviewable on the grounds that the Commission had failed to apply its mind to the question whether there had been compliance with the conditions. (This approach is consistent with the Tribunal’s own jurisdictional reservations as set out earlier in this decision.)

71. Second, HB contended that it was plain that the Commission issued the notice of apparent breach after fully considering the relevant factors. It could not therefore be accused of not applying its mind to them. Alternatively, HB suggested that the issues raised in this point in liminewent to the heart of the case and should be dealt with only after evidence on it had been led.

72. The Tribunal has, as was stated above, assumed in DHS’ favour, for the purposes for this decision, that it has the review powers contended for by DHS. It considers however that the second point in limineis without merit. Far from the Commission not applying its mind to the issues, it appears to the Tribunal that there is a good showing on the papers that the Commission repeatedly applied its mind to those issues and to ancillary issues pressed upon it by DHS, and perhaps also by HB, and was at all relevant times mindful of the state of the negotiations and reasons for their status at various time. In saying that it considered that there had been a breach of the condition because no agreement had been concluded by 23 May 2002 between a willing buyer and a willing seller, the Commission was merely expressing in a somewhat roundabout way its conclusion that the failure of the negotiations was attributable to lack of reasonableness on the part of DHS.

73. Accordingly, the second point in liminealso fails.

74. The Tribunal issued an order on 22 October 2003 dismissing the points in limineand reserving the costs in regard to them. As had been arranged, argument on the remainder of the case was heard on 21 November 2003.

75. An application was filed by the Commission on 5 November 2002 for leave to introduce the opinions on its behalf of an expert, Ms Lori Baker, and for consent that she be granted access to DHS’ expert reports, for which DHS has claimed confidentiality. The Commission withdrew this application at the hearing on 21 November 2003 after DHS had given notice that it intended to oppose the application. A decision on the issue of the costs of the withdrawn application, which were claimed by DHS, was reserved.

76. The matter proceeded on 21 November 2003, with Mr Burger presenting DHS’s case on the merits of its application.

77. Immediately after the mid-morning break Mr Burger announced that DHS was applying for the recusal of Mr Phatudi Maponya from the panel of the Tribunal which had heard the matter up to then. The grounds for this application were stated to be that DHS’s attorney had during the break become aware for the first time that Mr Maponya had some relationship with the Pretoria law firm, Maponya Inc., which practises in a form of association with the law firm Webber Wentzel Bowens, which has acted throughout the matter for HB. Mr Burger stated that this had led to a perception on the part of DHS that Mr Maponya might be subject to a conflict of interest. He stressed that there was no allegation that DHS had been prejudiced by Mr Maponya’s participation in the panel thus far. Mr Burger undertook on behalf of DHS that it would not attempt to invalidate any of the preceding proceedings or decisions of the Tribunal in the matter on the ground that Mr Maponya had participated in them.

78. Mr Maponya explained that he was no longer a partner in the law firm Maponya & Partners and was merely a consultant to it on various matters. However, in view of the fact that the issue had been formally raised, he recused himself from the panel with immediate effect.

79. Another member of the panel, Mr Lawrence Reyburn, pointed out that he had at one time been a partner in Webber Wentzel Bowens and later a consultant to it, but had resigned and no longer had a formal connection with that firm. Mr Burger announced that there was no objection by DHS to Mr Reyburn’s continued participation in the panel.

80. A direction from the chairperson of the Tribunal was then sought under s. 31(3)(a) of the Act as to whether the matter should proceed to finality before the presiding member, Mr Norman Manoim, and Mr Reyburn, as a two-man panel. The direction was duly given and the hearing continued on that basis.

81. Mr Burger contended that the review application in respect of prayers 1 and 2, on the alternative basis set out in the amended notice of motion (that is, under s. 27 of the Act read with rule 47 of the rules of the Tribunal), was still to be decided despite the Tribunal’s earlier decision on the second point in limine.

82. The Tribunal does not accept this contention. It considers that its decision on the second point in liminedisposed of all the review issues except those dealt with in the remaining part of this decision.

L Reyburn

Concurring: N Manoim

Did the applicant substantially comply with the conditions?

83. In this section of the decision, we ask whether DHS has substantially complied with the conditions for the approval of the merger: If the answer to that question is yes, then the Rule 39 notice must be set aside. If the answer is no, then the Commission is entitled to proceed with the further procedural steps that its rules contemplate, consequent on a determination that there has been a breach of merger conditions.¹⁸

84. It is common cause in this application that –

• HB is a healthcare switch entity as contemplated in paragraph 1 of the conditions;

• HB made a proposal for its switch technology to be integrated with a PMA package controlled by DHS, and this proposal was refused;

• DHS made a counter-offer to HB which the latter refused;

• the parties had protracted but unsuccessful negotiations concerning access;

• at the time of the issuing of the notice of apparent breach, and indeed to date, no agreement has been reached between DHS and HB to grant it access to the former’s PMA software.

85. On all other issues the consensus breaks down. First there is disagreement on how the conditions should be interpreted. This essentially boils down to a dispute over the extent to which HB was under an obligation to make a reasonable proposal for access - with DHS seeking to elevate the extent of this obligation, and the Commission and HB seeking to diminish it. Secondly, and more fundamentally, the parties disagree over what an agreement containing “reasonable terms” means.¹⁹ This second disagreement can be distilled again into whether “reasonable terms” are to be assessed from their effect on the commercial interests of DHS i.e. would a reasonable, prudent business person in the same position as DHS have entered into such an agreement? If this is what “reasonable terms” means, then DHS’ offer of June 2002 to HB, which the latter refused because it alleged that it exceeded what the market would consider “reasonable”’, may nevertheless from this subjective vantage-point, viz. the protection of the commercial interests of the firm granting access, be reasonable. So too would DHS’ insistence that an access agreement must be predicated on reciprocal back-end access.

86. On the other hand, if HB and the Commission are correct, the conditions must not be approached from the perspective of DHS’ commercial interest, but from what are reasonable terms for a rival to be granted access, and for reasonable terms read in “market-related” terms. They further argue that the insistence on back-end access is an extraneous issue and that it evidences, if anything, DHS’ unreasonableness.

87. In its replying heads, DHS contends that it is common cause that the Commission’s conditions are unambiguous. This is a remarkable statement given the length of the record on this point. At best it can be said that each side has shown confidence in its own reading - the fact that these readings yield such different results suggests that the interpretation of the order is no simple matter.

88. The source of the dispute created by the language of the conditions is the frequent use in the first two paragraphs of the word “reasonable”’ to qualify various obligations. The first two paragraphs are replete with the word, which the drafters seem to think needed to be inserted in every crevice where it could sensibly fit, limited only by the constraints of syntactical coherence. Its usage has largely led to the dispute we have today. It was not however the Commission’s initial choice of language. Indeed the first draft proposed by the Commission that appears in the record omits the word.²⁰ It was only when the lawyers representing the merging firms responded to that proposed draft that this word was inserted with such abandon, at their suggestion.²¹

89. As we indicated earlier, the first dispute arising from the interpretation of the conditions is what the extent was of the obligation on the party making what the order terms “a reasonable written request”. DHS argues for a conjunctive reading of the two paragraphs, so that the reasonable request is also one subject to terms that are reasonable commercially, financially and technically. (Mercifully in this case it is also common cause that both offers were technically reasonable, so this issue is off the table.) The request need not contain the terms of a full-blown agreement, but it seems what DHS has in mind are the core terms, perhaps sans the “boiler plate” clauses that are standard fare in most written contracts of this nature. Until a request of this nature has been made, so DHS argues, it is under no obligation to respond. Put another way, it is a necessary precondition before it incurs any obligation in terms of clause 2. On this approach it argues that since the HB offer contained terms that were not commercially reasonable its request failed to trigger any concomitant obligation from DHS.

90. In contrast HB argues that the two paragraphs must be read disjunctively. All that is required is that the request be reasonable. In argument Mr Unterhalter suggested that this meant no more than:

“… that a party with clear commercial intent, wanting to enter into a contract, sends a request for the kind of interconnection that is contemplated in the condition, then the obligation is triggered”

and that the phrase “commercially, financially and technically” does not qualify the request.²² Indeed this is why this language is not found in the first paragraph, which is the one that deals with the request, but only in the second paragraph, which is the one that deals with the merged firm’s obligations.

91. Interesting though this debate has been, we need not decide it. Even if we accept DHS’ contention, although we have doubts about it, we find, for reasons that we explore more fully below, that HB’s offer contained commercially, financially and technically reasonable terms. Given that finding, the debate over what obligation HB had in terms of the conditions is rendered academic.

92. What we must decide is the second part of the debate and that is the content to be given to the word “reasonable”. Again this debate can be distilled further into the meaning to be attributed to the term “commercially reasonable”.

93. Mr Burger made much of this term for reasons that become clear later. Several aspects of his argument are in our view uncontentious. One is that since the term “commercially reasonable” is used in the condition, as something distinct from “financially reasonable”, it must be given a distinct meaning, which would suggest it refers to terms other than purely financial ones such as price.

94. The next is that when the conditions refer to a “reasonable agreement” it must contemplate a range of possible agreements that could be described as reasonable. It is perfectly clear that agreements of this kind are subject to variation and complexity and that more than a single form would meet the test. Thus within the realm of reasonable agreements that meet the test, there may be one preferred by DHS, but not by HB, and vice versa. Provided DHS can prove it offered one that was in the range of what may be termed reasonable, albeit HB may have rejected the offer, it has substantially complied with the conditions. DHS might even be compliant if it rejected a reasonable offer from HB provided of course it responded with a reasonable offer from its own side.

95. However it is the next leg of Mr Burger’s argument, which is the real source of contention in this application. He builds on the term “‘commercially reasonable” to found a case for rejecting HB’s offer, defending DHS’ offer of June 2001, which HB rejected, and asserting the demand for back-end access. Essentially what he argues is that DHS is not obliged by the conditions to enter into an agreement that is commercially harmful to it. In this respect he argues that harm means harm to the DHS entity as a whole, not just Med-e-mass, the entity that would grant access. ²³

96. What DHS argues is that it would never in the ordinary course give access to a competitor in the position of HB, as opposed to any other competitor which is less threatening to it, unless it got something substantial back. Thus if it can prove it would never enter into a particular transaction in the ordinary course of business because it did not benefit the group to do such a deal, then it would, it follows, be an unreasonable deal – unreasonable being judged from the perspective of what would a rational firm, similarly circumstanced, do. If this proxy firm would not enter into an agreement on those terms because they would be irrational then such an agreement is commercially irrational and it would not be reasonable to expect DHS to enter into such an agreement.

97. For this DHS relies on the testimony of its expert, Mr Magennis. We go on to consider, in the next section, the reports of both Mr Magennis and the experts for HB.

The expert reports

98. We received expert reports from both DHS and HB. As was mentioned above, although the Commission applied to tender its own expert report, it abandoned this application in the face of opposition from DHS.

99. Initially at the pre-hearing on 15 August 2003 we were advised by counsel for both HB and DHS that each side would call one or more experts to deliver oral testimony. When we heard argument on the points in limine on 20 October 2003, counsel advised us that the parties had agreed that there was no need to call the experts for oral testimony and that, provided each side’s expert could have an opportunity to file a report in response to the other, this would suffice. This procedure was followed. Although we also received a report from Deloitte and Touche on behalf of DHS and on behalf of HB a report from Louis van Deventer of HAS Software Pty Ltd, these reports have not been considered separately except to the extent that they have been relied on by what we will refer to as the principal experts.²⁴

100. The principal expert for DHS was Mr Reg Magennis from Elixir Health Consulting, and for HB it was G:enesis. Both filed an initial report and subsequent to that a second report responding to the other’s initial report.

Magennis’ initial report

101. In his initial report Magennis concludes that:

“DHS will lose its switching business over time if Healthbridge is granted (or is allowed to retain) “front-end” access without granting DHS “back-end” access to Discovery and Medscheme systems to facilitate real time electronic trading.” ²⁵

102. Based on this assumption, he concludes that if DHS were to develop a charge for front-end access, under circumstances where it is not granted back-end access, it would be appropriate to compute such a charge based on the losses incurred by DHS Switch, and thus the DHS Group, if its switching business is partly lost to HB.²⁶ He works through what he assumes these costs to be and then concludes, going further and assuming DHS’ switch business is lost entirely, that:

“..therefore a case can be made for charging HB up to R436,50 per Med-e-Mass serviced site in a scenario where DHS is forced to compete in an environment where the playing fields between DHS and HB are not level.” ²⁷

(This charge of R436,50 is made up of R117,50 to compensate for Med-e-Mass’s costs plus an amount of R319 to compensate for the loss of business to DH Switch).

103. He thus concludes:

“ in the author’s opinion the proposed monthly charge of R130 per site for Med-e-Mass integration is sufficient to place Med-e-Mass in a position to sustain its profitability without the revenue and costs benefits that are derived from its linkage with DH Switch. In fact R117,50 would be adequate for this purpose. However a further amount of up to R319 per site would be required to compensate DHS if its DHSwitch business is lost. The DHSwitch business is likely to be lost if it does not enjoy the same ‘back-end privileges’ at Medscheme and Discovery that are enjoyed by HB.

Therefore the price of R130 (with a minimum annual charge of R100 000) for all sites) proposed by the DHS subsidiary Med-e-Mass for integrated access to its front end software is not unreasonable.” ²⁸

G:enesis initial report

104. G:enesis commences by defining what it understands reasonable to mean. It observes that the word ‘reasonable’ is not a term of art in economics or in competition analysis but that since the language is used in the context of competition policy and more specifically merger approval, this is the context to “shade and inform its practical meaning”. Given this, G:enesis proposes an interpretation formulated as follows:

”DHS is obliged to accept an offer from a switch that would reasonably reward it for costs incurred in providing access (i.e. be “financially and commercially reasonable’) and that would be technologically feasible and stable (“ technically reasonable”).²⁹

105. With this interpretation as its point of departure G:enesis then conducted two exercises to test the reasonableness of the HB offer to DHS. The first exercise tested whether the HB offer was reasonable in that it accorded with market norms in the sector for granting of access. The second exercise was to test whether the HB offer conformed to “reasonableness” by providing a reasonable reward to DHS for the costs of access. The distinction between the two is that in the latter exercise the costs of the firm granting access may be unreasonable, because of its own possible inefficiencies, and hence the need to test these costs to see if they accord with market norms.

106. As part of the first exercise (the accordance with market norms test) G:enesis examined 32 contracts between various switches and PMA’s. ³⁰ After reviewing the figures which are set out in its report, G:enesis comes to the conclusion that not only did the HB offer fall comfortably within market practice but that the transaction fees it offered are at high end of the market.³¹

107. G:enesis then goes on to assess the impact of the offer on Med-e-Mass.³² Significantly, as this is the major point of departure between it and Magennis, G:enesis states that it does not take into account the impact of the offer on DHS’ switching business as this:

“.. would defeat the purpose of the exercise if HB had to fairly compensate DHS for loss of business,” ³³

108. G:enesis then concludes that the HB offer was lucrative for DHS, even using conservative estimates. Med-e-Mass, it says, would have profited significantly from the offer.³⁴

109. After having completed its two exercises and finding that the HB offer was reasonable on both tests, G:enesis then went on to test whether the DHS offer was reasonable, of course based on the same assumption of what “reasonable” meant as we referred to earlier. G:enesis examines the terms of the DHS offer. It identifies four main characteristics of this offer:

1. HB has to pay DHS a R120 per month as minimum value per site. This fee would escalate by 10% each year;

2. The fee referred to above was subject to a minimum monthly fee of R120 000 (escalating by 10% per year) even if the roll-out of sites did not reach 1000.

3. In terms of the proposed contract, however, DHS was only obliged to roll-out 420 sites within seven months. In addition HB could not select the sites at which the installation would occur. HB would only discover after installation at which sites installation occurred.

4. HB pays an integration fee of R380 per hour and an installation fee of R300 per hour.

110. G:enesis then analyses the implications of each of these terms and concludes that not only was the DHS offer not in line with market norms, but that it was in fact even more unfavourable than those contained in the agreement with MediKredit, an agreement which Genesis says is not a market-related agreement, but contains terms that exemplify DHS’s ability to exercise post-merger market power over MediKredit. (Recall that the MediKredit agreement is the only one that had been concluded by DHS with another switch as part of compliance with the terms of the Commission’s order. In a letter to the Commission MediKredit complains about these terms but states that it had no choice but to accept them.)³⁵

111. Genesis concludes that the DHS offer was not in line with market norms nor indeed was the MediKredit agreement, and would result in a massive increase in the price charged for access. Genesis concludes that this amounted to an attempt by DHS to raise its rival’s costs and hence prevent effective competition in the switching market.³⁶

112. G:enesis goes into an analysis of this aspect in greater detail later, when responding to Magennis ‘s report, as we discuss below.

Response to G:enesis by Magennis

• In response to G:enesis, Magennis first deals with the analysis of the HB offer to DHS. He does not contest G:enesis’ thesis of the reasonableness of the offer vis a vis Med-e-Mass. Rather his point of departure is that any request for integration needs to be assessed from the perspective of the DHS group as a whole. He states in his response:

“ it would be inconceivable that the group shareholders would permit a subsidiary to take decisions that benefit that subsidiary in the short term but that threaten the long term viability of the group as whole.” ³⁷

113. Thus Magennis appears to concede that the HB offer is beneficial at least to Med-e-Mass. He goes on to argue that front-end access would create an asymmetry in the market that would lead to foreclosure effects, unless DHS gets fair and unfettered back-end access. ³⁸ He then details how HB could successfully implement a foreclosure strategy against DHS.

114. Magennis then examines the G:enesis critique of the reasonableness of the DHS offer. Magennis observes that G:enesis attacks the DHS offer on three bases: 1) that the MediKredit agreement is out of line with market norms 2) that the offer to HB is even worse than contained in the agreement with Medikredit, and 3 ) that the fixed R120 monthly fee proposed by DHS to HB translates into 45% of switch revenues per site and is thus three times the pre-merger market norm of 15%.³⁹

115. Magennis does not deny that the MediKredit agreement is more favourable than the offer to HB and hence is discriminatory. However he defends the discrimination saying that as Medikredit has no back-end connectivity:

“..the agreement has fewer business implications for DHS than a contract with HB would have”.⁴⁰

116. He then proceeds to defend the fact that the MediKredit agreement is not in line with what G:enesis contends is the market norm. His basis for doing so is that MediKcredit is already a switch and accordingly should pay more for access than a company which is not a switch such as HAS.⁴¹ He argues further that DHS cross-subsidises Med-e-Mass. The risk of accommodating HB is even greater than that of MediKredit because of HB’s ability to foreclose once it gains front-end access without granting back-end access.⁴²

117. He argues that Med-e-Mass should charge a premium to HB that equates in its price to the risk such a contract would represent to the DHS group as a whole:

“In our opinion it would only be logical to reduce this price premium if and once the risk premium associated with the back end access is convincingly and comparably reduced.” ⁴³

118. He goes on to refer to his initial report and points out that his computations were based on actual cost and revenues pertaining to DHS. As he put it :

“ they do not rely on a comparison with ‘ market norms’ as presented in the G:enesis report.”⁴⁴

119. Magennis does not dispute the 45% contention about the integration fee but again relies on the fact that if HB achieves greater roll-out the fee as a percentage of site transactions would reduce.⁴⁵

120. He goes on to criticise G:enesis’s methodology for not including in its calculations the fees that HB is likely to charge DHS for back-end access. He says:

“The value of these (unknown) fees should be set against the fees charged by Med-e-Mass for front end access.” ⁴⁶

121. On the R120 000, he says that if it is true that HB will effectively secure 100% of the switching market once it gains unfettered front-end access (in the absence of a symmetrical switching agreement), then the R120 000 minimum fee will have no economic relevance. He then engages in some ad hoc defence of the R120 000, by saying that the proposal allows for a management committee, which would oversee the performance of the agreement. He says that both parties would have an interest in rolling out at least 1000 sites over the 7-month period designated in the agreement, including Med-e-Mass, because it would spread costs of additional staff needed for the roll-out. He says that it is his understanding that it is the responsibility of HB to approach doctors to convert to real-time connectivity.

122. Magennis at first appears to justify the minimum R120 000 and R120 per site fee as Med-e-Mass’ way of protecting itself against the cost of roll-out. He argues that roll-out is risky and hence Med-e-Mass is entitled to protect itself against this risk and pass it on to HB. He says that given that Med-e-Mass has to retain a national footprint, it is reasonable for it to engage in strategy that ensures that there is a fixed price if it cannot be sure how quickly roll-out will take place. ⁴⁷ Yet even here the cost issue is always linked to back-end access. After referring to the fact that the R120 is similar to MediKredit’s R117,50 he states:

“ This calculation is relevant in a scenario where HB is placed in a position to eliminate DHSwitch.” ⁴⁸

123. Having said all this, however, he comes back with a devastating conclusion that if matters relating to back-end could be resolved:

“it is reasonable to assume that the DHS group would be under less financial and commercial risk and therefore could afford to be more flexible in its pricing structure.” ⁴⁹

G:enesis in reply

124. G:enesis commences its response by looking at the issue of back-end access since this is a central plank of DHS’s case. It points out that although Magennis relies heavily on this as part of his critique of HB’s approach he does not consider the various offers that HB and Discovery made to DHS in respect of back-end access.

“This omission undermines the authority and persuasiveness of the Elixir report” ⁵⁰

125. G:enesis then goes on to argue that real-time back-end access, the supposed competitive advantage that DHS asserts HB has over it, is not all that its made out to be. After considering the history of successive claim-lodging technologies, of which real time is only the latest, G:enesis argues that real time has not proven particularly advantageous to many doctors and is not a prerequisite to competing in the market. Thus although the real-time advantage is premised on the fact that the doctor is accessing the network as and when patients come in (i.e. many times a day) and secondly that it aids cash flow since the doctor is paid within 9 days (Discovery offers payment within 24 hours if a real-time end-user is prepared to pay an additional 0.9% of the value of the claim for the expedited payment) and not 10 days later, as is the case with batch switches like DHS, G:enesis’ research of Medscheme’s records shows that 95% of end users claim once a day or less. On payment frequency it argues that for smaller practices the alleged 9-day advantage is negligible (For a practice with a R 60 000 per month income this would amount to R 54 per month.) ⁵¹

126. G:enesis then goes into much greater detail in arguing why the DHS offer was not reasonable. It argues 1) that the DHS offer contained a large post-merger increase in price, 2) that when compared to market norm it would foreclose a significant part of the market to HB and 3) that it would render the strategy of HB of universal roll-out as ‘not achievable’.

127. Some prefatory remarks are needed to understand this roll-out argument and the statistics G:enesis seeks to rely on. It is common cause that not all practices are lucrative for switches. This is because certain practices generate such a low level of claims per month that they fail to cover the switches’ fixed costs for the site. Thus the switches compete not for the whole market, but for the lucrative end of the market i.e. those practices that will generate the most claims and hence the most revenue for the switches, as their revenue is based on the number of claims made on their networks.

128. From the perspective of a PMA supplier or at least based on the model in the offer of DHS to HB, the quality of revenue of the practice is irrelevant and as its cost is a fixed one, it will get the same fee irrespective of how many claims a practice lodges. Thus Med-e-Mass on the DHS offer is protected in the roll-out from the ‘claims quality’ of the practice.

129. From the perspective of HB’s two shareholders, Medscheme and Discovery, roll-out even to poor-quality practices is in their interests as the more practices that are on-line the less the administrators have to have their own staff and an infrastructure geared to dealing with the more expensive system of manual claims.

130. HB and DH Switch operate on different revenue models for their switching businesses. The essence of the difference is that HB receives payment, partly from Medscheme or Discovery for each claim it sends to them, whilst the balance comes from the practice that generated the claim. DHS derives its full fee from the practice, which has put it at a slight price disadvantage when compared to HB. (This has fuelled DHS’ notion that HB is subsidised by its two medical administrator shareholders.) Thus for switches the issue is not merely about roll-out, it is, to quote G:enesis, the difference between ‘cherry picking’ and ‘lemon picking’.

131. G:enesis then goes on to perform an analysis of the implications of the DHS offer on HB’s potential for roll-out, by applying the fees set out in the DHS offer to a sample of Medscheme practices to demonstrate how this would affect HB’s costs. Its calculations show that if the roll-out was just to the top 50% of the Medscheme practices (i.e. those that are the most profitable in terms of claims generated) HB’s costs would increase by 39%. If the roll-out were extended to all the Medscheme practices, HB would operate at loss.⁵²

132. G:enesis then argues that the cost implications of the DHS offer would be a possible increase in price by HB. G:enesis then does an exercise that compares the market norm for access (that it had referred to in its initial report) with the DHS offer. Under the market norm offer, 3400 sites of Medscheme (52%) would be profitable, whilst under DHS’ offer only 1400 (21%) would be profitable. HB has acknowledged to G:enesis that it does roll-out to sites that are not profitable for it on a fixed cost basis. Approximately 23% are negative in this respect. Although out of this 23%, 16% lose up to R20 per month, none have a pre-fixed cost loss of more than R20. Conversely, it argues that on the DHS scenario, 4900 sites would have a pre-fixed cost loss of more than R20, as opposed to none presently.

133. G:enesis also maintains that the DHS offer is an insuperable barrier to entry to switches that want to roll out to smaller practices. It argues that if roll-out started from the biggest sites it would be profitable up to 1400^th site, but thereafter becomes less profitable until it runs at a loss for HB from the 4600^th site onwards, assuming a total roll-out would be up to 6500 sites. Servicing the entire market would mean that HB would lose R230 000 per month. Applying the market norm, however, rolling out to all 6500 sites would mean that HB could still generate a profit, albeit lower than otherwise.⁵³

134. G:enesis goes on to analyse the fixed payment model vs. its own and suggests that it offers no incentive for Med-e-Mass to roll-out to high value sites. ⁵⁴ It points out that since DHS owns its own switch DHS is incentivised to roll out to lower as opposed to high revenue sites. Thus over a 36-month period HB could be left with only loss-making sites.

135. Figure 5 of the report illustrates the “long tail” of unprofitable sites. G:enesis goes so far as to suggest that DHS was “astute” in its offer as a no-lose situation for DHS. If the offer were rejected then HB would be denied access to the front end; if the offer were accepted it would raise HB’s costs relative to those of its rivals.⁵⁵

136. G:enesis also deals with who will give effect to the roll-out: the PMA supplier or the switch. Recall that Magennis had said that it was for HB to persuade practices of the advantages of real time. G:enesis says that it may be that HB has to persuade doctors to ask for conversion, but DHS is still responsible for its implementation. G:enesis points out that in terms of the contract DHS undertakes to rollout 420 sites within 7 months.

137. The upshot of this analysis is that Genesis concludes that the DHS offer raises HB’s costs and for this reason is unreasonable.

138. G:enesis also goes on to defend the reasonableness of HB’s and Discovery’s back-end offer to DHS. As will appear later in this decision this is not an issue that we have to consider.

Analysis of the expert testimony

139. There is no serious dispute of fact among the experts. What the experts disagree upon is how the concept of reasonableness should be analysed.

140. Magennis assesses the notion of reasonableness, as we have seen, from the affect it has on the business of DHS as a whole in the market in relation to HB and concludes that if DHS gives up front-end access without securing back-end access, the business of DHS will be adversely affected as it will lose all or part of its switching business to HB, because of the latter’s real-time advantage.

141. Let us assume for the moment that we accept the evidence of Mr Magennis in this respect. Could this be what the Commission’s order means when it refers to a reasonable access agreement and in particular commercially reasonable terms?

142. Against this we have the interpretation advanced by HB and the Commission. According to them DHS enjoys, post merger, market power over the switching segment of the market. It has achieved this through the vertical integration of the dominant PMA firm with a significant switch player, which post-merger would allow it to foreclose its rivals in the switching market. On the evidence before us the only competitor of significance is HB. For HB and the Commission the real but unstated basis of the Magennis argument is that it is unreasonable for DHS to contract on terms with its rival that do not compensate it for loss of its market power (euphemistically referred to as its competitive advantage). Thus they argue that what DHS wants to be compensated for under the mantle of reasonableness is really compensation for loss of its market power. This, they argue, cannot be a proper interpretation of the conditions.

143. We are persuaded by this analysis of the Commission and HB. The purpose of the access remedy, as the Commission explains in its merger decision, is to ensure that the switching market remains competitive after the merger. In other words, in return for approving a merger, which would lessen competition, the Commission seeks to retain pre-merger levels of competition by ensuring access to the PMAs of DHS. Implicit in this formulation is the assumption that by permitting access, DHS will be forced to face competition and accordingly be unable to exercise post-merger market power over the switching market.

144. It follows, once one appreciates the economic consequences of what access means (when posited, as the Commission explains in its report, as the antidote to the exercise of market power) that the firm that acquired market power and was now to be inhibited from exercising it by granting access, would be worse off – at least to the extent that its market power premium would be forgone. It makes no sense for the Commission to impose a condition that would allow access only on condition that the merged firm is compensated for its loss of market power. Yet that is precisely the implication of the case that DHS has made. Whilst such an interpretation of the conditions may have been plausible were this an arbitration assessing damages arising from a breach of a commercial contract, it is nonsensical in the context that we are dealing with here – an order by a body charged with protecting competition, which as part of this function can regulate the approval of mergers inter alia to prevent the exercise post merger of market power by the merged firm.

145. When the peculiar language of the conditions is assessed in this statutory context, and when they are read with the Commission’s reasons for its approval of the merger, the concept of reasonableness contended for by DHS becomes untenable.

146. In adopting this approach to the interpretation of the conditions we are following an approach that is well settled in case law and which has been followed by us previously. That is one set out in decisions such as Administrator, Cape, and Another v Ntswaqela and Others (1) 1990 (SA) 705 at 715 E-I, and Firestone South Africa (Pty) Ltd v Genticuro 1977 (4) SA 298 (A) and The ANC v The UDM and Others Constitutional Court of South Africa, Case No: CCT 43/02.⁵⁶

147. If we interpret the conditions in a manner consistent with the Commission’s reasons for its decision then it is perfectly clear what it intended. Firstly the sentence that serves as a preamble to the conditions states:

“However, the Commission is of the view that the conditions set out below would significantly reduce those concerns. The merger is, therefore, approved, subject to the following conditions…”

148. Thus the Commission makes it clear that the conditions serve to remedy the anticompetitive effects that it finds will come about as result of the merger. Given this intent, it seems that the Commission saw the grant of access as vital to remedy this problem. It is unlikely that the Commission would easily have contemplated a situation where access did not occur, since access was the cure or partial cure to the ill it saw. If access did not occur, the merger would proceed as if approved unconditionally.

149. However, the Commission could also foresee that a firm required to give access is not obliged to do so without some compensation. The Commission seemed to recognise one could cure the competition concern without on the other hand making the merged firm vulnerable to some competitor who required access at say sub-market terms. This would go beyond the Commission’s intention of ensuring competition in the switching market and lead to the opportunity for exploitation of the merged firm.

150. In this regard we agree with HB that this is the meaning to be given to the reasonableness terms i.e. the merged firm was not obliged to enter into an agreement for access that was not in accordance with market norms for such access or did not adequately compensate an efficient firm for the costs of granting such access. It does not follow that this allowed the merged firm to seek compensation at a supra-market price or to impose terms that are unrelated to the access agreement and which may render it less likely to lose its market power. The exploitation of the merged firm that the Commission sought to protect by the caveat of reasonableness in its order did not extend to compensation for losses in the switching market that access by a competitor might occasion. This would amount to interpreting the condition as if the Commission desired only the appearance of competition, but not its reality.

151. The experts for HB are correct when they state that the entire purpose of the remedy was to create a competitive situation in the switching market. It is consistent with these conditions, and indeed access conditions in general, that in consequence the firm granting access may find its market position eroded - that is unless it reacts by competing more vigorously. What DHS regards as a perverse outcome of the reading of the order by HB is in fact the situation that the order contemplates. This is the raison d’ être for the condition.

152. We find that DHS’ interpretation of the conditions is incorrect, and accept the version of the Commission and HB, namely that ‘reasonableness’ meant that the access terms should not be exploitative of the merged firm, in the sense we have explained above. Indeed, there is an equally plausible third reading of ‘reasonableness’ which is not inconsistent with this one above, and that is that if the merged firm gave access on terms that perpetuated its market power - the suggestion made by MediKredit in relation to its agreement with DHS - those terms would be unreasonable. In other words, reasonableness, which every one has contended had to be read to benefit the merged firm, might have been meant to protect the competitive process from abuse. Nevertheless, we need not find that this latter interpretation is the correct one or the only correct one. It suffices for us to reject the interpretation of DHS and to accept that of the Commission and HB as the more plausible.

153. Having come to this conclusion on the interpretation, we find on the facts that HB has established that its original offer was reasonable. Firstly, in terms of its consistency with market norms, even being conservative in this regard. Secondly, and this appears to be uncontested, it was reasonable in the sense that it adequately compensated Med-e-Mass for the costs of access. It was not necessary for the offer to compensate DHS for its loss of its switching business, assuming, accepting Magennis, that this might have been a consequence of the agreement.

154. It is of course not sufficient for us to find that HB’s offer was reasonable. We must also find DHS’ responding offer unreasonable. In this respect we find that HB’s experts have made out a case for this offer not only being less favourable to HB than the market norm, but also likely to raise its costs of doing business. Again, this evidence is not seriously contradicted, and DHS‘ riposte, that compensation for the loss of market power justifies these terms, has already been rejected by us as a proper approach to interpreting the conditions. Since that was the only basis on which DHS sought to defend the reasonableness of its offer, it too must fail, and the offer is considered unreasonable.

155. It follows then that DHS’ insistence on back-end access was an issue extraneous to the proper interpretation of the order and a device for DHS to compensate itself not for the reasonable costs of granting access to the PMA market, but a loss of market power or the potential to exercise that power in the switching market.

The non-conclusion of an agreement

156. DHS says in par 12.8 of its heads of argument, dated 13 September 2003, that it is clear from the Commission’s notice that the non-conclusion of the agreement triggered the CC19 notice and not whether there was substantial compliance with the merger obligations. This failure by the Commission to apply its mind should by itself lead to a review in terms of rule 39(2)(b), as well as in terms of s 6(2)(e)(ii) and s 6(2)(f)(ii) of the Promotion of Administration Justice Act 3 of 2000.⁵⁷

157. However, we need not address this issue since this argument was addressed in great detail as an in limine point in the first part of our decision where we found it to be without merit. ⁵⁸

Hacking

158. DHS argued as an independent and self-standing reason for the denial of access, that HB had allegedly sought to achieve access to its PMA software by hacking.

159. Again we do not need to go into the record on this issue, which is lengthy, acrimonious and disputed by HB. Indeed, in fairness to HB, we must record that it denies the allegations strenuously and points out that a High Court application for an interdict was withdrawn after HB filed its answering affidavit. To add another twist to the tale it appears that the proceeding is to be renewed by DHS in another form.

160. We are certainly in no position to determine if hacking occurred. What we must ask is, if DHS bona fide believed that hacking took place, did this justify it not entering into an agreement with HB? In our view it did not. Mr Burger correctly conceded in argument that the hacking issue did not relate to whether HB was a flawed person and thus an undesirable party to contract with. Rather he pinned his colours to hacking being a reason for not reaching a timely agreement. This argument does not make sense on the record. It is quite clear from the correspondence that after the allegations of hacking were first aired in August 2001 the parties continued to negotiate.⁵⁹ Indeed, the existence of DHS’ April 2002 offer premised on back-end access indicates that it was this issue, not hacking, that was foremost in DHS’ consideration about negotiating access. Hacking appears to have become a post hoc rationale to justify DHS’ failure to grant access and to serve as another arrow in its quiver of defences should the others fail.

Conclusion

161. In our view the applicant has failed to establish that it has substantially complied with the obligations contained in the conditional approval of the merger.

162. In the first place it refused to enter into an agreement proposed by HB on reasonable terms.

163. Secondly, the terms on which it was prepared to grant access were unreasonable both in relation to its offer of June 2001 and its subsequent demand to make access conditional on back-end access. Thirdly, the linking of the refusal to the alleged hacking was on the facts before us an extraneous issue that did not reasonably constitute a ground for not agreeing to access.

164. The application therefore fails.

Costs

165. HB is awarded the costs of the application on a party and party basis including the costs of three legal representatives. No costs are awarded to the Commission as we have not yet heard any argument to persuade us against our present understanding of our powers which is that we can neither award costs for or against the Commission.⁶⁰ Had the situation been otherwise the Commission would have been entitled to costs.

166. For the same reason no costs are awarded against the Competition Commission in respect of the withdrawal of the application to file a late expert report.

4 February 2004

N. Manoim Date

Concurring: L Reyburn

---