Life Healthcare Group (Pty) Ltd v Joint Medical Holdings Ltd (74/LM/Sep11) [2012] ZACT 88; [2013] 1 CPLR 227 (CT) (24 October 2012)

78 Reportability
Competition Law

Brief Summary

Competition Law — Merger Control — Approval of merger between Life Healthcare Group and Joint Medical Holdings — Competition Commission recommending prohibition based on potential market dominance — Tribunal finding that merger does not substantially prevent or lessen competition, as Life already exercised de facto control over JMH — Merger approved unconditionally.

Comprehensive Summary

Summary of Judgment


1. Introduction


These proceedings concerned a large merger before the Competition Tribunal of South Africa in which Life Healthcare Group (Pty) Ltd (the acquiring firm) sought to acquire an additional shareholding in Joint Medical Holdings Ltd (the target firm), a Durban-based private hospital group.


The Competition Commission investigated the transaction and recommended prohibition. The Tribunal heard extensive factual and expert evidence over 21–29 May 2012, with final argument concluding on 18 June 2012. The Tribunal issued an order on 24 July 2012 approving the merger unconditionally, and furnished its reasons on 24 October 2012.


The dispute concerned the competitive effects of increased concentration in private hospital services in the greater Durban metropolitan area, with the Commission advancing several theories of harm, most prominently potential price effects. A significant part of the Tribunal’s analysis focused on identifying the correct counterfactual in circumstances where the acquiring firm already held a large minority shareholding and, on the evidence accepted by the Tribunal, exercised de facto control over the target firm prior to the notified transaction.


2. Material Facts


Life Healthcare Group is one of South Africa’s three largest private hospital groups. Joint Medical Holdings owned five hospitals in and around Durban, namely City Hospital, Ascot Park, Maxwell Clinic, Isipingo Hospital, and Durdoc Hospital. Prior to the notified transaction, Life already held 49.4% of JMH’s shares; the remaining 50.6% was held by approximately 306 doctor shareholders, none individually holding more than 6%.


In the notified transaction Life sought to acquire a further 21% of JMH’s shares, increasing its shareholding from approximately 49% to 70%, and thereby acquiring de jure control (in addition to a greater economic interest). The Tribunal accepted that the transaction also had a practical background purpose of resolving an ongoing control impasse, particularly after a rival hospital network (NHN) complained to the Commission that Life was unlawfully negotiating tariffs on behalf of JMH.


By the time of the hearing, there was no dispute on market definition. The relevant market was accepted as the market for the provision of private hospital services in the greater Durban Metro area. On a conventional aggregation of market shares (ignoring any “proportional” adjustment for pre-existing shareholding), the combined post-merger share for Life and JMH would be just over 50% in the greater Durban market, with Netcare at approximately 32%, and a small number of independents making up the remainder. The post-merger HHI was stated as 3566, reflecting an increase of 1072.


A central factual dispute relevant to the counterfactual was the extent and nature of pre-merger control. Life’s evidence (primarily through Mr Wylie) was that Life had exercised sole de facto control over JMH since 1997, despite not holding de jure control under the shareholder agreement structure. On that evidence, major decisions were effectively settled ahead of board meetings between Life executives and doctor-appointed directors, with Life’s position prevailing. The Tribunal considered that board minutes and internal documentation were consistent with this account.


The Tribunal also considered that Life’s current account of de facto control was in tension with historical engagements with the Commission, including the Amahosp merger notification in 2001 (in which JMH appeared as a competitor rather than a controlled firm) and advisory opinion correspondence in 2003 regarding increasing Life’s stake. The Tribunal did not resolve the credibility of Life’s explanations for the historical inconsistency as a separate issue, but it did decide—on the record before it—that the best supported factual position for the purposes of the present proceedings was that Life exercised de facto sole control prior to the merger.


On pricing, the Tribunal accepted evidence that after centralised tariff negotiations were ended by consent arrangements, Life negotiated tariffs on behalf of JMH for most periods relevant to the evidence (with an exception around 2003 when Life held only 25% and JMH was not on Life’s systems). A significant episode occurred in 2011 during valuation disagreements: Life informed GEMS that JMH would no longer participate in Life’s arrangements, after which JMH explored joining the National Hospital Network (NHN) for tariff negotiations; this was later reversed when Life withdrew its exclusion letter and asked for JMH’s reinstatement in Life’s arrangements.


The Tribunal treated this episode as material to assessing the probable pricing counterfactual, particularly whether JMH, if required to bargain independently of Life, would likely have gravitated to the NHN tariff framework.


3. Legal Issues


The Tribunal was required to determine whether the merger was likely to result in a substantial prevention or lessening of competition under section 12A(1) of the Competition Act 89 of 1998, and if so whether any conditions were warranted. This required a predictive assessment, but one grounded in evidence.


A central legal issue was the correct identification of the counterfactual. The dispute involved the application of law to fact: although Life contended that the merger would not materially change market conduct because it already exercised de facto control, the Commission argued that the status quo could not serve as the counterfactual if the status quo depended on unlawful conduct, particularly unlawful coordination on pricing between horizontal competitors.


This led to a further legal question: whether Life and JMH, given their relationship and shareholding structure pre-merger, could lawfully behave as part of a single economic entity (and thereby fall within the section 4(5) intra-group exemption), or whether coordination on pricing would instead be exposed to the restrictive horizontal practice prohibitions in section 4(1) (and presumptions in section 4(2)).


Additional issues included the evaluation of the Commission’s theories of harm on pricing and non-pricing dimensions (specialists, diversion/tunnelling, regional schemes, exclusionary conduct), and a procedural issue regarding the Tribunal’s discretion to admit further evidence after final argument in a merger proceeding.


4. Court’s Reasoning


The Tribunal began by emphasising that merger assessment requires a comparison between the likely market outcome with the merger and the likely outcome without it, and that the “without” scenario is often the status quo but may require adjustment where the status quo is not an appropriate benchmark.


On the evidence, the Tribunal accepted that Life had exercised de facto sole control over JMH since 1997, in a manner consistent with section 12(2)(g) (material influence comparable to ordinary indicia of control), even though Life lacked de jure control at board and shareholder meeting level. The Tribunal reasoned that while the doctor shareholders could theoretically have constrained Life, there was no evidence that they did so in any competitively significant way.


However, the Tribunal rejected the merging parties’ attempt to use that factual control finding to maintain a counterfactual in which JMH and Life continued their pre-merger joint pricing arrangements. The Tribunal treated this as engaging a basic policy principle: firms should not be permitted to rely on unlawful conduct in order to portray a merger as competitively neutral. The Tribunal framed the point as analogous to competitors who collude and then argue that a merger makes no difference because they already set prices jointly.


The Tribunal then analysed the statutory framework governing coordination between firms in a horizontal relationship with partial ownership links. It relied on section 4(2) (presumption of an agreement where one firm has a significant interest in another or there are common directors and they engage in the restrictive practice) and contrasted that with the narrow exemption in section 4(5) for agreements within a wholly owned subsidiary structure or a single economic entity similar to that structure. The Tribunal reasoned that the legislation specifically contemplates that partial shareholdings and board overlap may facilitate unlawful coordination and does not automatically immunise such coordination. It concluded that the pre-merger relationship between Life and JMH—less than 50% shareholding for much of the period, absence of de jure control, and the potential for shareholder challenge—did not place them within the section 4(5) exemption as a single economic entity.


The Tribunal further reasoned that the legality of coordination could not be “cured” by the fact that the original partial interest was lawfully acquired. It referred to its prior approach that changes in control can require notification (citing its own earlier decisions, including Ethos), emphasising that merger analysis proceeds on what is notified and approved, not on later shifts in the nature of control. Against this framework, the Tribunal held that Life’s de facto control did not entitle it to engage in conduct that would otherwise contravene section 4(1), and specifically that joint pricing between horizontally related firms would fall within the prohibition on price fixing in section 4(1)(b)(i).


While the Tribunal expressly stated it was not making a finding of prohibited practice liability (as this was not a prohibited practice hearing), it concluded that the correct merger counterfactual must assume lawful independent pricing between Life and JMH pre-merger. Where such lawful independence did not exist in practice, the counterfactual had to be formulated hypothetically rather than by reference to the parties’ actual (and potentially unlawful) pre-merger behaviour.


Applying this counterfactual to the Commission’s theories of harm, the Tribunal accepted that, for nationally negotiated tariffs with major funders, the merger would not materially increase Life’s bargaining power, and it treated GEMS evidence as supporting that conclusion (including the fact that GEMS had been unaware that JMH was being negotiated for within Life’s arrangements).


The principal remaining pricing question was whether JMH, absent the merger and pricing independently, would have priced lower (for example via an NHN tariff) such that the merger would eliminate that competitive constraint. The Tribunal assessed expert evidence attempting to compare Life and NHN tariffs and overall costs and found it inconclusive due to methodological limits and data constraints. It sought to address the evidentiary gap by appointing an independent expert (Dr Green) to advise on feasibility and methodology; Dr Green explained the complexity and data demands of a robust comparison. The Tribunal decided not to undertake a delayed and resource-intensive further data exercise given uncertainty that it would yield probative results.


On the probabilities, the Tribunal nevertheless found it likely that NHN tariffs to funders were lower than Life’s, drawing support from independent hospital testimony and funder submissions in the record, and from the inference the Tribunal drew from the 2011 GEMS episode. At the same time, the Tribunal accepted that consumables and alternative reimbursement models (ARMs) complicated the comparison between “tariffs” and “overall cost,” and it accepted that Life’s buying power and ARM arrangements could reduce consumables costs to funders. The Tribunal therefore concluded that any likely tariff increase at JMH, measured against the hypothetical lawful counterfactual, would probably be slight, particularly given JMH’s limited size and the likelihood of offsetting consumables effects.


The Tribunal dismissed the Commission’s additional pricing theory regarding uninsured patients as speculative on the record, and it treated non-pricing theories of harm (specialists, diversion/tunnelling, regional scheme bargaining, exclusionary impacts) as insufficiently supported by merger-specific evidence. It noted that some concerns might be real in the private healthcare sector generally, but the Tribunal considered that the Commission’s case lacked specificity showing that this merger would materially change outcomes given Life’s already entrenched de facto influence and the relatively small increment in incentives and footprint.


The Tribunal also addressed the Commission’s attempt to introduce additional evidence after final argument, and the Tribunal’s own request for an independent expert report. It held that merger proceedings are not ordinary adversarial civil trials and that the Tribunal has inquisitorial powers and a duty to reach its own conclusion under section 12A. Relying on the Competition Appeal Court’s guidance in the Wal-mart/Massmart matter, the Tribunal rejected an approach of excessive procedural formalism and admitted the additional evidence and the expert report, while also finding that neither ultimately supported prohibition.


Finally, the Tribunal recorded that the Commission had indicated it was investigating whether joint pricing conduct might contravene section 4(1), but the Tribunal did not decide that issue within the merger reasons. The Tribunal also highlighted potential statutory consequences of inaccurate disclosures to the Commission (including sections 15(1), 16(3), and 73(2)(d)), treating these as matters for investigation rather than as determinative of the merger outcome before it.


5. Outcome and Relief


The Tribunal held that the merger was not likely to substantially prevent or lessen competition under section 12A(1), and that there were no public interest considerations under section 12A(3) requiring intervention.


The merger was approved unconditionally by order issued on 24 July 2012, with reasons issued on 24 October 2012. The reasons do not record a costs order.


Cases Cited


Mondi Limited and Kohler Cores and Tubes (Competition Appeal Court Case Number 20/CAC/Jun02).


Afrox Healthcare Limited and Amalgamated Hospitals Limited (Competition Tribunal Case Number 53/LM/Sep01).


ISCOR Limited and Saldanha Steel (Pty) Ltd (Competition Tribunal Case Number 67/LM/Dec01).


Ethos Private Equity Fund IV and Tsebo Outsourcing Group (Pty) Ltd (Competition Tribunal Case Number 30/LM/Jun03).


Copperweld Corp. v Independence Tube Corp. 467 U.S. 752 (1984).


Viho v Commission (Case T-102/92) [1995] ECR II-17.


Viho v Commission (Case C-73/95P) [1996] ECR I-5457.


Imperial Chemical Industries Limited v Commission (Case 48/69) [1972] ECR 619.


Ahmed Saeed Flugreisen and Others v Zentrale zur Bekämpfung Unlauteren Wettbewerbs (Case 66/86) [1989] ECR 803.


Minister of Economic Development and Others v Competition Tribunal and Others (Competition Appeal Court Case No: 110/CAC/Jul11 and 110/CAC/Jun11).


Legislation Cited


Competition Act 89 of 1998, including sections 1, 4(1), 4(2), 4(5), 12(2)(g), 12A(1), 12A(3), 13(1), 15(1), 16(3), and 73(2)(d).


Rules of Court Cited


Rules of the Competition Commission governing information required to be supplied with a merger filing (referred to generally, without specific rule numbers in the reasons).


Held


The Tribunal found, on the evidence, that Life exercised de facto sole control over JMH since 1997 (without de jure control), but that this did not entitle the merging parties to rely on pre-merger joint pricing as the merger counterfactual, because the counterfactual must be lawful competitive conduct, which required independent pricing between Life and JMH pre-merger.


On the merger assessment, the Tribunal held that the Commission’s theories of harm—pricing and non-pricing—were not established to a degree sufficient to show a likely substantial prevention or lessening of competition. The Tribunal considered that any incremental competitive effect on pricing (measured against the lawful counterfactual) was likely slight, particularly given JMH’s limited size and the complexity of overall cost comparisons.


The Tribunal admitted additional evidence after final argument (including an independent expert report and further emails tendered by the Commission), holding that merger proceedings are inquisitorial in nature and should not be constrained by excessive procedural formalism. The additional evidence did not change the competitive assessment.


LEGAL PRINCIPLES


The Tribunal applied the principle that merger analysis requires a comparison between the likely world with and without the merger, and that the counterfactual must reflect a legally permissible benchmark rather than the continuation of potentially unlawful market conduct.


It applied the control principle in section 12(2)(g) that de facto control can arise through the ability to materially influence a firm’s policy in a manner comparable to ordinary indicia of control, even where there is no de jure majority at board or shareholder level.


It applied the restrictive practices framework in sections 4(1), 4(2), and 4(5) to hold that partial ownership or influence over a horizontal competitor does not automatically create immunity from horizontal collusion rules. The intra-group exemption is confined to wholly owned subsidiary structures and comparable “single economic entity” arrangements contemplated by section 4(5), and does not extend to the type of partial ownership and contested governance structure found on the facts.


It applied a procedural principle that merger proceedings are not ordinary adversarial civil litigation and that the Tribunal has inquisitorial powers and a duty to reach its own determination under section 12A, including the ability to admit further evidence where necessary, consistent with the Competition Appeal Court’s guidance that merger hearings should not be stultified by excessive procedural formalism.

1




COMPETITION TRIBUNAL OF SOUTH AFRICA

Case No: 74/LM/Sep11
013235

In the large merger between:

Life Healthcare Group (Pty) Ltd Acquiring Firm

and

Joint Medical Holdings Ltd Target Firm



Panel : Norman Manoim (Presiding Member)
Lawrence Reyburn (Tribunal Member)

Medi Mokoena (Tribunal Member)
Heard on : 21-29 May 2012
18 June 2012
Order issued on : 24 July 2012
Reasons issued on : 24 October 2012


Non-Confidential Decision


INTRODUCTION
[1] This is a merger of two hospital groups which i mpacts specifically on the
Durban Central area. The Competition Commission (“ Commission”) had
recommended a prohibition of the proposed merger.

[2] The factual and expert evidence in this merger was heard from 21-29 May
2012 and the last day of closing arguments was 18 J une 2012. The Tribunal
issued its order on 24 July 2012 unconditionally ap proving the merger. Our
reasons for the approval are set out below.

2

[3] During the hearing of this matter the Commissio n called the following factual
witnesses:
• Mr Niresh Bechan - hospital manager of the Ethekwi ni Hospital and Heart
Centre
• Mr Glen Passmore - board director of the Hillcres t Private Hospital
• Dr Stan Moloabi – executive, healthcare managemen t for Government
Employees Medical Scheme ( GEMS)
• Mr Ebrahim Asmal - hospital manager of the Lenmed Shifa Hospital

[4] Professor Alex Van Den Heever, a specialist in the South African healthcare
industry, and Mr Simon Pilsbury, an economist from Oxera, were called as the
Commission’s expert witnesses.
[5] The merging parties’ factual witnesses were:
• Mr Matthew Prior - funder manager at Life Healthc are Group
• Mr Kurt Wylie - board member of Life Healthcare Group
• Mr Jonathan Lowick - Group Strategy and Developme nt Executive at
Life Healthcare Group

[6] Dr. Nicola Theron an economist from Econex, was called as the merging
parties’ expert witness.

THE PARTIES TO THE TRANSACTION

[7] The primary acquiring firm is the Life Healthca re Group (“Life” or “LHG”),
1 a
wholly owned subsidiary of Life Healthcare Group Ho ldings. Life Healthcare
Group is a public company listed on the Johannesbur g Stock Exchange and
incorporated in terms of the laws of the Republic of South Africa. 2
[8] Life is not controlled by any single entity. So me of its main shareholders are
Old Mutual Assurance Company South Africa (13.25%), GEPF Equity

1 http://www.healthcare.co.za/Default.aspx
2 Incorporated in the Republic of South Africa under Registration number: 2003/002733/06.

3

(7.75%), CBNY: International Finance Corporation of South Africa (5.12%),
Industrial Development Corporation of South Africa (5.02%) and Mvelaphanda
Strategic Investments (Pty) Ltd (3.25%). 3
[9] Life is currently one of the three largest priv ate hospital groups in South Africa
and has facilities which include hospitals, rehabil itation units, occupational
health clinics and facilities which care for chronically ill patients. 4
[10] Joint Medical Holdings (“JMH”), the primary t arget firm in this merger, owns
five hospitals in and around the Durban area, namel y; City Hospital, Ascot
Park, Maxwell Clinic, Isipingo Hospital and Durdoc Hospital.
[11] Life currently holds a 49.4% shareholding in JMH. The remaining
shareholders of JMH are individual medical practiti oners or their family trusts,
deceased estates or companies controlled by medical practitioners
(collectively referred to as ‘doctor shareholders’) who in aggregate hold 50.6%
of the issued share capital of JMH. JMH has approxi mately 306 doctor
shareholders. None of these doctor shareholders ind ividually holds more than
6% of the shares. 5

THE TRANSACTION AND THE BACKGROUND

[12] In terms of the present transaction, Life seek s to acquire a further 21% of
the shares in JMH, thus increasing its shareholding in JMH from 49% to
70%. 6 Post merger, Life will not only assume a greater e conomic interest in
JMH, but de jure control as well.







3 Commission’s Report dated 18 January 2012, page 7.
4 Ibid .
5 Commission’s Core Bundle A File 1 page 43-44.
6 Jonathan Lowick explained during the hearing that the reason Life did not make an offer to go up to
100% was that it believed in having its doctors as shareholders. Transcript 24 May 2012 page 104.

4

THE RATIONALE FOR THE TRANSACTION

[13] Jonathan Lowick, Life’s Group strategy direc tor, stated that the proposed
transaction was in line with Life’s long-term strat egy since it had a long-
standing desire to own a majority of the shares in JMH. 7

[14] For JMH’s doctor shareholders the rationale is for them to realise all or part
of their investment in JMH.

[15] The more probable, though not expressly articu lated reason, is that the
merger resolves an impasse in the control situation of the hospital group as
we discuss more fully later. 8 Lowick admitted during the hearing that a
complaint from a rival hospital network, the Nation al Health Network
(“NHN”), to the Competition Commission, alleging th at Life was unlawfully
negotiating tariffs on behalf of JMH, had exacerbat ed the need to resolve
the control situation. 9 Life, the de facto controller of JMH, will now, post
merger, become the de jure controller as well, whilst its economic interest
increases commensurately. This issue is discussed more fully later.

RELEVANT MARKET

[16] Unusually for a competition case, by the time it came to the hearing there
was no dispute about the relevant market. Despite s ome differences both
the Commission and the merging parties have accepte d the market
definition as being the market for the provision of private hospital services in
the greater Durban Metro area. 10 If one ignores the pre-merger
shareholding that Life has in JMH, then the accreti on in market share that
the merger brings about would be as follows:

7 Transcript 24 May 2012, page 72.
8 This is implied by the evidence of Lowick who desc ribed how Life had been struggling to acquire a
majority stake since he had knowledge of the issue. This was from 2009.
9 See transcript page 102. The complaint was made in October 2010.
10 The area in which four of the JMH hospitals are lo cated in Durban central also has two Life

hospitals, two Netcare hospitals and an independent in close proximity. See Exhibit A, a map of the
Greater Durban area, indicating the location of sev eral hospitals belonging to JMH, Life and other
groups. Asmal, of Lenmed Nushifa, one of the Commis sion’s witnesses, remarked in an email to a
funder that hospitals in Durban were no more than a 5 kilometre radius from one another. See Exhibit
G1.

5

Market shares in Greater Durban market

Firm Pre -merger market share Post -merger market share
Life 34.59 50.08
JMH 15.49 0
Netcare 32.38 32.38
Ethekwini 3.08 3.08
Hilcrest 7.97 7.97
Nu Shifa 6.48 6.48
Total 100.00 100.00


The post merger HHI would be 3566 representing an increase of 1072.
11

[17] On this basis the merger brings about an incre ase in Life’s market share of
15.5% in the region; from 34.5% to just over 50%. L ife of course has a pre-
existing holding of 49% in JMH and a degree of pre- existing control over it.
Life argued that this pre-existing shareholding of 49% needs to be factored
into the market share calculation on a proportional basis and if allowance is
made for this and the fact that it only intends to acquire up to 70% of the
equity in the present transaction then its market s hare accretion in the
Greater Durban market would be only 4.65% and its t otal shareholding
would be 45.4%.


[18] The Commission did not accept this approach to determining market share
increments.
12 We do not need to decide this issue in the present case. The
proper consideration is what is the relevant counte rfactual because once
this has been identified the proper competition con sequences of the merger
can be more clearly identified.


11 The source for these figures is Table 3.2 in the O xera report prepared by Simon Pilsbury, the
Commission’s expert.
12 See comment of Pilsbury in the Oxera report, page 22, where he says he considers this an unusual
and inappropriate approach to determining market shares.

6

APPROACH

[19] We will first decide the proper counterfactual. Having decided that we then
go on to consider the various theories of harm advanced by the Commission
and the merging parties’ response.


THE COUNTERFACTUAL

[20] In merger cases the assessment of the relevant counterfactual is an
essential part of the analysis. Essentially this in volves a comparison of
market outcomes; the market that would prevail with out the merger, usually
taken as the status quo, compared with the scenario that is likely to prevail
post-merger. The difference between the two scenari os informs the threshold
question raised by section 12A(1) of the Act viz. – whether the merger would
lead to a substantial prevention or lessening of completion. Usually the status
quo serves as the proxy for what the market would be like absent the merger,
while the post- merger future requires a predictive analysis.
13

[21] This has been the approach of the merging part ies, who argued that Life
had, since its acquisition of a 25% stake in JMH, i n 1997, exercised de facto,
albeit not de jure, control over JMH. Since de facto control suffices as a form
of control in terms of section 12(2)(g) of the Comp etition Act, Act 89 of 1998
(“the Act”), the only difference the merger makes is the establishment of de
jure control. 14 Since the merging parties argue that the move from de facto to
de jure will have minimal impact on JMH’s behaviour for co mpetition
purposes, the merger essentially retains the pre-me rger competitive status
quo.



13 See Mondi Limited and Kohler Cores and Tubes (Competition Appeal Court (CAC) Case Number
20/ CAC/ June02) at paragraph 38, where the CAC hel d that the section whilst not permitting
speculation lacking an evidential basis nevertheles s still, “... enjoins the Tribunal to make a predictive
judgement based on the evidence which has been placed before it.”

judgement based on the evidence which has been placed before it.”
14 We set out the terms of this section in paragraph 30 below.“

7

[22] The Commission’s approach is that the facts of this case constitute an
exception to the normal approach that the status qu o serves as the proxy for
the market without the merger. The reason it does s o is that whilst Life may
control JMH presently in the same way as it might p ost merger, the question
is whether it does so lawfully. If it does not do s o lawfully then the status quo
should not serve as a proxy for the market without the merger. 15

[23] In order to decide this we need to review the history of events leading up to
this merger as well as evidence on this aspect give n in the course of the
hearing.

[24] In 1997 Life’s predecessor acquired a 25% inte rest in JMH. 16 The
shareholders agreement concluded then is still oper ative. 17 As far as legal
form goes, the agreement gives Life rights to appoi nt 25% of the directors
and to exercise a veto in relation to a number of key operational decisions in
JMH. It does not give Life the right to control the board of directors or the
majority of votes at an annual general meeting. Alt hough Life acquired a
further stake in JMH in 2004, taking its holding to the present one of 49.4%,
this did not lead to an alteration of the original shareholders agreement nor
to the control situation despite the fact that Life had almost doubled its
equity in the company. Whilst it was allowed in pr actice to appoint another
director to the board, Life still did not bring its board representation in line
with its shareholding percentage. We are told that immediately prior to the
merger Life appointed three directors to the JMH bo ard of eight. 18 As we
discuss below this further acquisition was not the subject of a merger filing
under the Act.


15 In the European Commission’s merger guidelines all owance is also made for situations where the
pre-merger market counter factual is not the statu s quo, but also might require a predictive analysis ,

See Economics for Competition Lawyers , Gunnar Niels et al, Oxford University Press 2011, page
338-339.
16 The predecessor was a company called Presmed which later became Afrox and then, after a
restructuring of the shareholding in Afrox in 2005, became Life. Nothing turns on this change in name
as the company remained the same throughout and for that reason, to avoid confusion, we will simply
refer to Life throughout although this appellation is not historically accurate for the period prior t o
2005.
17 Shareholders agreement, record File A pages 386-410.
18 See witness statement of Kurt Wylie paragraph 3.2.

8

[25] Thus to summarise – immediately prior to the present merger Life did not
control the majority of votes at either board or ge neral meeting level
although the shareholders agreement gave it rights of veto in certain
specified instances.

[26] In the present merger, Life contended that des pite the absence of de jure
control, it has de facto controlled JMH since 1997.

[27] The witness put up by the merging parties to testify on this aspect was
Kurt Wylie who wears two hats – he is both an execu tive of the Life
Healthcare Group and also served on the JMH board f rom 2006 to June
2011 as one of Life’s nominees. 19 He could not testify about the position
that pertained earlier than this, but he could stat e that he was told that the
manner in which things worked when he assumed his p osition on the JMH
board was no different from what had prevailed sinc e 1997. Hence, he
contended, it was reasonable to assume that the man ner in which Life
related to JMH from 2006 onwards was no different t o the position in the
period from 1997 until 2006.

[28] Wylie’s evidence was that Life controlled all the major decisions that JMH
made. 20 Prior to board meetings agreements on the agenda i tems were
reached between the doctor-appointed directors and Life’s executives. Life’s
views always prevailed. JMH has discovered all its board minutes for the
period. Nothing has emerged from a perusal of these that was inconsistent
with Wylie’s version. 21 This is not to say that there were not differences in
strategy between the directors nominated by the doc tor shareholders and
those of Life. Indeed one internal strategy documen t authored by Life

19 Wylie’s witness statement page 2.
20 See Wylie witness statement paragraph 3. Similar e vidence was also given by Matthew Prior, Life’s
funding manager. See Prior’s witness statement at p aragraph 2.7 where he stated, “LHG is by far the

largest single shareholder in JMH and, in practice, controlled JMH from before the time that
negotiations at a hospital and scheme/ administrator level commenced.”
21 The Commission did identify minutes where some mat ters of disagreement arose from time to time
between doctor appointed directors and those nomina ted by Life. But nothing material arose from this
indicating that Life could not prevail over the oth er shareholders when the issue was important
enough to it.

9

executives highlights these differences very clearl y. 22 But what the record
shows is that to the extent that differences existe d, the views of Life always
prevailed.

[29] The next event of importance that occurred was Life’s 2001 acquisition of
Amalgamated Hospitals Limited (“Amahosp”), a firm t hat managed and
owned four hospitals in Kwa Zulu Natal. This merger , as required, was
notified as a large merger in terms of the Act in J uly 2001. In terms of the
notification requirements an acquiring firm, in thi s case Life, is obliged to list
all the firms it controls. If Wylie’s evidence is a ccepted then Life controlled
JMH in 2001. But the Amahosp merger notification makes no mention of this
fact. 23 In another part of the form, the notification desc ribes JMH as a
competitor. 24 Market shares of the firms in the greater Durban ma rket are
given and there again JMH is listed as a competitor . The only clue of any
link between the two hospital groups is the annual financial report of Afrox
which, as required by the notification form, is ann exed to the filing. In the
report there is a list of associated companies. Amo ngst those listed is JMH,
together with a statement that the firm (then Afrox ) holds 25% in what is
described as an associate. 25

[30] In terms of section 12(2)(g) of the Act a pe rson controls a firm if that
person- “ ... (g) has the ability to materially influence the pol icy of a firm in a
manner comparable to a person who, in ordinary comm ercial practice, can
exercise an element of control referred to in paragraphs (a) to (f). ”

[31] On Wylie’s evidence JMH was controlled by Life by virtue of section
12(2)(g) since 1997. This meant that JMH should hav e been reflected in the
Amahosp filing as an ‘ acquiring firm’ , because in terms of the definition
section of the Act, an acquiring firm includes all firms directly or indirectly

22 See Afrox position paper April 2004, Record File B1 1656.

22 See Afrox position paper April 2004, Record File B1 1656.
23 The record of the merger notification, dated 23 July 2001, is contained in Exhibit T1 submitted by
the Commission.
24 Ibid, page 7 paragraph 16.1. See also the competitiveness report which lists competitors and their
market shares. In this table JMH is listed as a competitor. See Annexure 2 to the merging parties
Competitiveness Report.
25 See Ibid at Annual Report for Afrox Healthcare Group for year 2000 – page 57.

10

controlled by another acquiring firm. 26 Since Afrox then was an acquiring
firm and controlled JMH at the time, it should have been reflected as such
on the form CC4, which is the merger filing form. J MH was not reflected as
such and instead, as we noted, the contrary impress ion was created in the
filing that JMH was a competitor of the acquiring a nd target firms. The
Amahosp merger was subsequently approved by the Tri bunal
unconditionally. Mention is made in the reasons for the approval that
amongst the competitors of the merged firm in the m arket analysis was
JMH.27

[32] In 2003 Life’s attorneys wrote to the Commissi on requesting an advisory
opinion. The attorneys wrote that Life sought to ac quire a 49% stake in
JMH. Their factual submission was that Life would n ot be able to control
JMH at board or general meeting level and asked if for that reason Life was
still required to notify the transaction. In the at torneys’ view this was not
required by the Act.28 The shareholders agreement entered into in 1997 was
not, it appears, submitted to the Commission. The s ubmission also does not
make mention of the fact that Life had had a 25% st ake in JMH since 1997.
But the most significant omission, if Wylie’s evide nce in the present case is
correct, is the failure to mention that at that tim e Life already de facto
controlled JMH. The reader of the letter would have reasonably assumed
that Life had no pre-existing stake in JMH and was preparing to acquire, ex
nihilo, 49%. What is more curious is that the same firm of attorneys had
notified the 25% acquisition by Presmed (Life’s Pre decessor) in 1997 to the
Commission’s predecessor, the Competition Board.29

[33] Despite this submission the Commission took th e view that the stake could
lead to de facto control and advised Life to notify it. 30 It did not do so, nor did
it acquire the further 24% stake. Instead, almost a year later, a different firm

it acquire the further 24% stake. Instead, almost a year later, a different firm
of attorneys requested an advisory opinion on the s ubject from the

26 Section 1(1)(i).
27 Tribunal Case Afrox Healthcare Limited and Amalgamated Hospitals Limited Case Number:
53/LM/Sep01 paragraph 16 page 4.
28 See record Bundle A File 1 page 411.
29 See record Bundle A File 1 page 422.
30 See record Bundle A File 1 page 417 .

11

Commission. This time the 25% stake was revealed as well as the existence
of the shareholders agreement. The argument now adv anced was that the
pre-existing stake had conferred a degree of contro l by way of the 1997
shareholders agreement, but since that agreement wo uld remain in place
the acquisition of the additional 24% would not change the quality of control.
It was contended that Life exercised joint not sole, control over JMH.

[34] After an initial objection the Commission was persuaded that this view was
correct. 31 In their letter the attorneys relied purely on an interpretation of the
terms of the shareholders agreement. Whilst this wa s a fair reading of the
shareholders agreement, no mention was made that th e de facto situation
was that Life controlled JMH despite only holding 2 5% -- the view that Wylie
has propounded in this merger. After receiving the Commission’s opinion
that notification was not necessary Life proceeded to increase its
shareholding in JMH from 25% to 49%.

[35] But there were further intervening events. In April 2004
the Commission
concluded a consent order with the Hospital Associa tion of South Africa
(”HASA”) and its members, amongst which was Life. I n terms of that
agreement HASA agreed to end centralised industry n egotiations regarding
the setting of private hospital tariffs with the as sociation that represented
private healthcare funders. 32 From then on each hospital or hospital group
had to negotiate tariffs separately with each of the funders.

[36] This meant that whilst JMH and Life had charge d the same tariff up till
then, being that of the industry as the outcome of centralised negotiations,
they could not do so thereafter.

[37] The evidence in this case is that in the period for which the witnesses could
provide information Life has negotiated tariffs on behalf of JMH at all times

31 See record Bundle A File 1 page 416 for the initial objection, and page 474 for the subsequent
acceptance.
32 See paragraph 7 of the HASA consent agreement 24/CR/Apr04. A similar consent agreement was
entered into with Board of Health Care Funders South Africa which negotiated on behalf of funders.
See 07/CR/Feb05.

12

except for the year 2003. 33 The reason given for the exception in 2003 was
that at that time Life held only 25% of JMH and JMH was not on Life’s
financial systems. Certainly after Life had acquire d the shareholding which
moved its interest in JMH to 49%, Life negotiated t ariffs on behalf of JMH.
According to Lowick this was permissible notwithsta nding the earlier
consent order undertakings because Life had interro gated its ownership of
JMH with the Commission and from this process Life had received comfort
that it could treat JMH as part of its network for the purposes of establishing
tariffs.34

[38] Asked specifically if the issue of negotiating tariffs on behalf of JMH had
been revealed to the Commission Lowick admitted it had not, but he was
not able to explain the reason for the omission as he had not been in his
current position at that time. 35

[39] The Commission argued that this was collusive. Life as a competitor of
JMH was not entitled to negotiate tariffs on its be half. We deal with this
contention later.

[40] As we noted earlier, in the current merger Lif e argues that it has had
control of JMH since 1997, including control over t he setting of its tariffs,
and hence the merger will have no affect on competition.

[41] This position, as we have shown, is inconsiste nt with the stance that Life
took towards the Commission in its Amahosp notifica tion and in its
correspondence when seeking an advisory opinion on two occasions in
2003 regarding its contemplated acquisition of a further stake in JMH.



33 See evidence of Lowick at transcript 24 May 2012 page 84.
34 Transcript 24 May 2012 page 85.
35 Transcript 24 May 2012 page 86. In a later answer he says that the advisory opinion from the
Commission gave Life comfort that it could include JMH in its negotiations. See transcript page 91.

13

[42] This begs the question: which version of the f acts regarding control is
correct? We queried this with the merging parties during the hearing. 36 No
witness gave evidence which dispelled the apparent contradictions.37

[43] In correspondence with the Tribunal the mergi ng parties’ attorneys offered
an explanation. 38 They properly concede that evidence previously sub mitted
by the merging parties or by Life and relied on by the Commission and
Tribunal to the effect that JMH was a competitor of Life at the time, was not
correct. They state that those responsible for the Amahosp notification are
no longer employed by Life and that therefore they have to speculate as to
why the Amahosp merger was presented on the basis t hat it was. They go
on to surmise that at the time of the Amahosp notif ication jurisprudence
around control was in its infancy and hence the ext ent of the concept was
not as fully appreciated as it is now. 39

[44] For the purpose of this case we do not need to decide whether this
explanation is credible. What we do have to decide for the purpose of the
counterfactual is which version of control is corre ct: the Amahosp filing
version in 2001, the version accompanying the secon d 2003 request for an
advisory opinion, or the version advanced in this m erger. The choice is
between no control of Life over JMH, (as implied by the Amahosp filing), an
attenuated form of joint control (second advisory o pinion letter in 2003) or
unvarying sole de facto control (Wylie’s evidence in the present hearing).

[45] On the evidence before us it would appear that the Wylie version is correct.
The minutes trail, the document serving before the Life board to approve the
deal, are all consistent on this point. 40 Wylie was thoroughly cross-
examined on the control issue and in our view answe red satisfactorily to
issues within his personal knowledge.

36 Transcript 21 May 2012 page 3.

36 Transcript 21 May 2012 page 3.
37 Lowick when asked this says, “I think the focus is on the shareholder’s agreement and I don’t know
why.” Transcript 24 May 2012 page 92.
38 The present attorneys representing Life in this m erger were not involved in either the Amahosp
notification or the two applications for an advisory opinion referred to above.
39 See letter from merging parties attorneys to the Tribunal dated 12 June 2012.
40 See Record Bundle 1 File A page 360 Life Board pape r July 2011 where it is stated, “Life has no
written management contract with JMH, but decided o n certain matters relating to the strategy of
JMH, particularly its business plan, budget, revenue, capital expenditure and operational costs.”

14


[46] We therefore find that Life or its predecessor s have de facto had sole
control of JMH since 1997. Whilst legally the docto r shareholders or their
nominated directors, acting jointly, may have been able to constrain that
control, we have no evidence that they ever did so, certainly in any manner
that might be competitively significant. The merger therefore serves to bring
the de jure situation in line with the de facto situation, as Life will now control
a majority of the votes at board level and at general meetings.

[47] Having made this finding we must answer the ne xt key question which
affects an important counterfactual question. If Li fe has controlled JMH
since 1997 will the merger make any difference to p ricing behaviour post
merger since Life has controlled JMH’s pricing sinc e 2003/4 when central
bargaining over hospital tariffs was outlawed?

[48] Behind this question is an important principle of law.

[49] Let us consider the position from first princi ples. It is a trite proposition that
if two competitors had colluded on pricing and then sought to merge they
could not rely on that prior collusion to argue that the merger would make no
difference to pricing post merger because the count erfactual is a market in
which they did not compete. This would be contrary to a precept of public
policy that firms cannot benefit from their unlawful conduct.

[50] The merging parties argue that their situation is not analogous to the one
outlined above, as they had lawfully acquired joint control in 1997, and joint
control must pre-suppose joint pricing, otherwise c ontrol is stripped of its
essential meaning. However this argument ignores a key provision in our
legislation which provides specifically for such si tuations and in a manner
contrary to that contended for by the merging parties.


[51] The Act provides for a regime that discourages horizontal interests in
competitors. Whilst not making such holdings unlawf ul the Act creates a

15

presumption, when consideration is given to horizon tal restrictive practices
in which a combination of firms is involved, that a n agreement exists
between the firms where one of those firms holds a substantial interest in
the other or they have directors in common. 41

[52] The intention of the Act is to facilitate the prosecution of firms in such a
situation and thus to discourage such partial forms of control. But at the
same time the Act exempts from the provisions of se ction 4(1) agreements
between firms and their wholly owned subsidiaries. 42 This exemption is also
extended to firms that constitute part of a ‘single economic entity’. 43 The
term ‘single economic’ entity is not given precise definition. But the Act does
give some guidance as it states that the exemption applies to the
constituent firms of a “ single economic entity ” that is “...similar in structure...”
to entities in the wholly owned subsidiary –parent relationship.

[53] The use of the word “ similar” assumes that the subordinate or controlled
firm need not be wholly owned by the parent firm (o therwise subsection 5(b)
would be redundant), but still seems to suggest tha t the structure is not far
removed from it. What is contemplated by the exempt ion is not merely firms
in a single economic entity. Its meaning is more limited than this since there
is an additional requirement that the relationship between the relevant firms
must be “similar in structure ” to the relationship between a parent company
and one or more wholly owned subsidiary or sub-subsidiary.

[54] In U.S. law in terms of the so-called ‘Copperweld’ doctrine a firm is deemed
not to be able to collude with itself. 44 The boundaries of ‘self’ when located

41 Section 4(2) which states “ An agreement to engage in a restrictive horizonta l practice referred to
sub-section 1(b) is presumed to exist between two or more firms if –

sub-section 1(b) is presumed to exist between two or more firms if –
(a) any one of those firms owns a significant inter est in the other, or they have at least one
director or substantial shareholder in common; and
(b) any combination of those firms engages in that restrictive horizontal practice.
42 Section 4(5)(a).
43 Section 4(5)(b).
44 Copperweld Corp. v Independence Tube Corp. 467 US 752, 104 Sct ( 1984).In Copperweld a
firm and its wholly owned subsidiary were deemed in capable of conspiring with one another for
purposes of section 1 of the Sherman Act because th ey did not represent separate economic
interests. The application of Copperweld by the district courts in these latter cases has be en
inconsistent. Consequently there are contradictory judgments, for example in Novatel v Cellular Tel.

16

in separate corporate entities become more tenuous, the less the interest of
the holding company in the subordinate company. Nei ther US law nor
European law would extend the notion of ‘self’ or a s in Europe, a single
economic entity, to the present one between Life and JMH pre-merger.

[55] If the legislature had intended a partial cont roller or joint controller of a
company to be immune from liability for colluding w ith it by ordinary
operation of law as a consequence of partial or joi nt control, then it is hard
to see why this provision was inserted in the Act. Quite clearly the
legislature had no such intention. JMH and Life are therefore not, pre-
merger, where on the facts the holding firm does no t even enjoy de jure
control of the subordinate firm, and holds less tha n 50% of its shares,
constituent firms within the same single economic entity.

[56] The only time when a firm can be certain of im munity from the
consequences of a section 4(1) prosecution is where it and the subsidiary
form part of a wholly owned subsidiary-parent relat ionship or the type of
single economic entity contemplated by section 4(5) (b). Where the
relationship between the controlling and the controlled firm falls short of this,
such collusion will not be exempt from the conseque nces of sections 4(1)
and 4(2).

[57] Nor does the fact that the controlling firm ac quired the substantial interest
as part of a previously notified merger entitle it to immunity from section

Supply (1986) 51% voting control was deemed to be covered by Copperweld whereas in Aspen Title
& Escrow v Jeld-Wen (1987) 75% ownership of the subsidiary was held to be insufficient to qualify for
Copperweld protection. There is thus no clear approach in the US for cases with partly-owned
subsidiaries.

subsidiaries.

The EU approach is based on the principle that if t wo or more entities form an economic unit within
which the subsidiaries have no real freedom to dete rmine their conduct in the market, the entities
essentially form a single undertaking and cannot be expected to compete with each other, thus
relations between them are not classified as collus ive arrangements ( Viho v Commission (1995),
Case T-102/92, (1995) ECR II-17; upheld on appeal C ase C-73/95P (1996) ECR I-5457 at [16-18] ICI
v Commission (1972), Case 48/69 ICI v Commission [1972] ECR 619 at [134 Ahmed Saeed
Flugreisen and Others v Zentrale zur Bekampfung Unl auteren Wettbewerbs (1989) Case 66/86,
[1989] ECR 803. The governing test is therefore whe ther the subsidiaries enjoy the autonomy to
determine their own conduct in the market.

17

4(1). Where the notified merger contemplates only p artial control the
acquirer is not entitled to the status of a sole co ntroller unless it notifies the
fact of sole control to the authorities. This is be cause merger analysis is
performed on the basis of a merger notification as lodged and not on what
might later become the prevailing form of control. For this reason we have
held previously that a change from joint to sole co ntrol triggers a further
notification as does the crossing of a bright line. This was laid down in our
Ethos decision. 45

[58] This does not mean that the controller, not co ntemplated in section 4(5),
cannot lawfully exercise some measure of control ov er the subordinate firm.
What it does mean is that control, where firms are in a horizontal
relationship, cannot extend to conduct that is unla wful in terms of section
4(1). Where the controlling firm and subordinate fi rm so act, they do so
unlawfully. The relevance of this to the present co unterfactual is that Life
was not entitled to set prices in conjunction with JMH. To the extent that it
did so, this conduct contravened section 4(1)(b)(i) of the Act, which prohibits
competitors from jointly fixing the price for their services.

[59] It is not our purpose to make a finding that L ife and JMH have in fact
colluded on pricing because we are not engaged in a prohibited practice
hearing. The only issue relevant to the present pro ceedings is the merging
parties cannot rely on a past history of joint pric ing to create a
counterfactual that the merger would make no differ ence to pricing post
merger, because they had priced jointly before it.

[60] The correct approach to a counterfactual must be to compare what
behaviour by firms would have been lawful competiti on between them pre-
merger, with the post merger scenario. It is not pe rmissible to use unlawful
competition as the yardstick of measurement. Where such a lawful

competition as the yardstick of measurement. Where such a lawful
counterfactual does not exist in practice, pre-merg er, as in this case, the
lawful counterfactual must then be the subject of h ypothesis. Merging


45 ISCOR Limited and Saldanha Steel (Pty) Ltd Case Number 67/LM/Dec01 and Ethos Private Equity
Fund IV and Tsebo Outsourcing Group (Pty) Ltd Case Number 30/LM/Jun03 at paragraphs 18 – 47.

18

parties cannot benefit from conduct that was, pre-m erger, unlawful, to make
the comparison with the post merger world seem benign.

[61] We conclude on the counterfactual issue as follows:
1. Life exercised de facto control over JMH, pre-merger, but not de
jure control;
2. The de facto control was lawfully acquired in 1997 and existed
even before Life increased its stake in 2003 from 25% to 49%;
3. The de facto control did not amount to unfettered sole control as it
was always subject to challenge at board or general meeting by a
voting majority of doctor shareholders or their rep resentatives; for
this reason, inter alia, JMH and Life did not cons titute part of a
single economic entity;46
4. Life’s de facto control did not entitle it to collude with JMH in
respect of conduct that would otherwise contravene section 4(1) of
the Act; and
5. The correct counterfactual assumes that Life and JMH would, pre-
merger, have priced their services independently of one another.

Commission’s theories of harm

[62] The Commission has advanced several theories of harm to support its
contention that the merger be prohibited and we con sider each one
separately.

Pricing


[63] There are two issue raised in respect of prici ng. First is the effect of the
merger on pricing for Life, the second on pricing i n respect of JMH. In
respect of the first, both Commission and the mergi ng parties are agreed
that the merger will not enhance Life’s ability to increase its pricing power in

46 There may be additional reasons why they do not constitute a single economic entity but we
consider it unnecessary to detail them.

19

respect of insured patients who form part of nation al medical schemes. 47
This is because pricing is negotiated nationally wi th individual funders and
the addition of the JMH beds will have no meaningfu l effect on Life’s
negotiating strength. This evidence was confirmed b y the representative of
the only funder to testify at the hearing, Dr Moloa bi, who is an executive
director of Gems. 48 The best proof of this contention was the fact tha t,
unbeknown to Gems, Life had negotiated on behalf of JMH when
negotiating tariffs with Gems. 49 (We deal later with the two residual issues
raised by the Commission in respect of the merger o n Life’s pricing, namely
the effect of the merger on private patients of Lif e, i.e. patients who are not
members of medical schemes, and those who are membe rs of a regional
medical scheme.)

[64] The question whether JMH’s pricing post merger would increase became
an issue during the hearing. Here of course the iss ue of the proper
counterfactual is pertinent. On the merging parties ’ version this issue was a
simple one. They had been negotiating on behalf of JMH pre-merger and
so, since JMH was already charging funders Life tar iffs, the merger would
make no make difference to pricing – the status quo would continue.
However, as we have found that this is not the corr ect counterfactual, we
cannot accept this argument.

[65] What then would the proper counterfactual on pricing be, given that there is
no factual evidence of the two groups pricing indep endently of one another?
JMH was either pricing as part of the industry’s ce ntral bargaining together
with all other hospital groups that belonged to HAS A, including Life, or after
2003, negotiating as part of Life. Fortunately, dur ing the hearing we
received evidence of an event that assists us in ar riving at the most
probable hypothesis.



47 See Oxera report paragraph 5.5 and Econex Report page 15 .

47 See Oxera report paragraph 5.5 and Econex Report page 15 .
48 He stated “... we do not consider that the addition of a few hospitals to the existing hospitals of a
group would affect negotiations substantially .” See Moloabi witness statement, paragraph 19.
49 Transcript 22 May 2012, page 36. Moloabi conceded that he could not be influenced by something
he did not know.

20

[66] During 2011, whilst Life was negotiating with the doctor shareholders over
pricing for their shares as part of the present tra nsaction, a disagreement
arose on valuation. Matthew Prior, acting under ins truction from chief
executive Mike Fleming, wrote a letter to Moloabi of Gems to inform him that
as from 1 July that year “...Joint Management Holdings will no longer
participate in the arrangements between the Life He alth Care Group and
GEMS” .50 A similar message was given to JMH’s management. J MH
proceeded to consider joining an independent hospit al grouping known as
NHN (the National Hospital Network) which is entitl ed to negotiate a single
tariff with funders on behalf of its members. 51 As it happened the period of
JMH’s exclusion from the Life network was short-liv ed. On 28 July 2011
Gems was informed by Life that: “There has been a change of
circumstances regarding the JMH Hospitals and we ac cordingly withdraw
our letter dated 20 th June 2011. In the circumstances please reinstate JMH’s
inclusion in the Life Healthcare arrangements.” 52 What this meant was that
Life would again be negotiating on JMH’s behalf. It would appear that the
negotiating tactic had worked.

[67] There is other evidence apart from this episod e that supports the
probability that were it to have negotiated indepen dently of Life, JMH would
have joined the NHN camp.
53 In 2011, according to a letter written by
NHN‘s Otto Wypkema to an advisor, JMH had approache d NHN to join it. 54
Since the merging parties accept this fact as well, this proposition is not
contentious. What is contentious however is what th is would have meant in
terms of pricing to funders. Would JMH’s pricing o n an NHN tariff be lower
to funders and if so, would the difference between the latter tariff and that of
Life have been significant?



50 Letter from Life to GEMS dated 20 June 2011. See E xhibit D2.See evidence of Lowick on this.

Transcript 24 May 2012, pages 96-98.
51 We are informed that the NHN has received an exemption from the Commission entitling it to do so.
52 See letter from Life to GEMS dated 28 July 2011. Exhibit D3.
53 The costs of having the necessary personnel capabl e of negotiating with funders is not insignificant;
Prior estimated that the costs could amount to R3-R 4 million rand annually. See Prior witness
statement paragraph 2.8
54 See record pages 1740 to 1741, emails from Bhoola of JMH to Wypkema, and then from Wykema
to Wouter Meyer. See also transcript 24 May 2012 page 99.

21

[68] To answer this we need to appreciate how hospi tal pricing operates. When
hospitals bill funders for services there are three categories of expense.
First, there is a tariff for hospital services; for example ward fees and
operating theatre fees. Second, the cost of medicin es used by the patient.
The third category comprises the materials the hosp ital requires to perform
its services for the patient.


[69] All hospitals pass on the same costs for medicines as these prices are now
regulated by what is termed single exit pricing. Whilst this does not preclude
a hospital or doctor from using a generic equivalen t this is an area in which
price variances between hospitals is constrained by regulation. 55 How
hospitals or doctors may manage the use of drugs is a different matter. It is
therefore only the other two categories, namely the fee for services and the
charge for consumables that offer a basis for compa ring overall hospital
costs. Life emphasises, correctly that the issue is not which tariff is lower to
funders, but which hospital has lower overall costs i.e. tariffs plus
consumables.


[70] This issue was not resolved in the volume of p apers that were exchanged
prior to the hearing. During the hearing, factual w itnesses who were based
at independent hospitals testified on behalf of the Commission that in their
experience NHN hospitals had lower tariffs to funde rs than those of Life. 56
However the examples of tariff differences that wer e given were anecdotal
and subject to criticism for this reason by the merging parties.57

[71] For this reason during the course of hearing we asked that the Commission
and merging parties’ experts to perform a more rigo rous comparison. The

55 See Prior testimony transcript 23 May page 38.
56 See for example Ebrahim Asmal witness statement pa ragraph 20 where he alleges that the

56 See for example Ebrahim Asmal witness statement pa ragraph 20 where he alleges that the
difference is some 15-20% higher at hospital groups , and Niresh Bechan witness statement
paragraph 13 who gives the same figure for the difference in charges.
57 As an example Asmal alleged that a normal delivery at his hospital would cost R 2500 less than at
a Life Hospital. But when challenged whether he kne w this for certain he admitted it was just an
assumption. See Transcript 23 May 2012 page 22-23.

22

Commission during its investigation had access to s ome data from the NHN
whilst Life had access to its own data.

[72] This exercise was carried out, but its results proved inconclusive. Simon
Pilsbury, the Commission’s expert, performed a limi ted comparative
exercise.58 This exercise relied on the oral testimony of a fa ctual witness
(Asmal) to decide which category of procedures to c ompare. Based on this
he selected 10 procedures to compare and then added three more following
discussions with the merging parties’ expert Dr The ron. Due to availability
problems he limited his data to that of one particu lar fund offered by one
funder - Discovery. His conclusion was that the dif ference in tariffs was
marked and varied from [...] to [...] depending on the weighting given to the
respective categories and whether minor surgery was included. 59

[73] Dr Theron considered the sample too narrow, bo th as to funder and types
of treatment. Instead, she performed what she said was a wider exercise,
using other schemes as well as Discovery. Her research revealed that whilst
Life was more expensive for Discovery [...] it was cheaper for others (for
Metropolitan, [...] cheaper); it was, if one appropriately weighted f or the
relevant factors and the schemes in terms of benefi ciaries, and if one
excludes minor surgery, overall very similar to tar iffs for NHN. 60
Unfortunately, Theron did not perform this exercise in conjunction with
Pilsbury, but gave him her figures at the moment he was to commence his
oral testimony. (Pilsbury testified before her). Wh en he testified he said he
had not had time to verify the figures and so could not comment on them.
He did not subsequently do so. As a result we were deprived of the benefit
of a joint document from the experts on this issue.

[74] We were not satisfied with this outcome and so we decided after hearing
final argument that this was sufficiently important an issue for us to attempt

final argument that this was sufficiently important an issue for us to attempt

58 See Exhibit R.
59 See Exhibit R.
60 See Exhibit S. Theron also excluded minor surgery as she said this contributed only [...] to Life’s
revenue and noted that Pilsbury’s use of Discovery had not taken into account DRG’s and the
influence of Keycare one of the lower cost Discovery options. See Exhibit G page 4.

23

to get better evidence. Pricing appeared to us the fundamental competition
issue raised, given the counterfactual we had deter mined. We therefore
decided at that stage to explore the feasibility of commissioning evidence on
this aspect from an independent expert of our choos ing. Given that the
experts for the Commission and merging parties had not only come to
different conclusions but also adopted different me thodologies we sought
guidance from the independent expert on the following issues, viz. whether:
(1) either expert had offered a satisfactory compar ison of the
respective costs to funder;
(2) if not, whether this comparative exercise coul d be done on a more
reliable basis than suggested by these experts whic h avoided the
criticisms made out above or was the information to complex to
analyse?
(3) if the answer to (2) was yes, whether he would able to do this
exercise or whether he would require more informat ion and if so, what
evidence and from whom?

[75] The expert we appointed, Dr David Green of Gre en West Knowledge
Consultants, provided a very valuable methodology f or determining the
comparison. In fairness to the two other experts, Green’s approach points to
the complexity of this exercise and the need for da ta to perform it from
willing funders.

[76] In brief, Green’s opinion was that the exercis e of comparison is highly
complex. He illustrated the number of factors that would have to be
considered to make the comparison robust. For insta nce he says that
hospital practices that relate to the timing of adm ission and discharge of
patients or their movement within a hospital from w ards with different levels
of care will have consequences for differences in b illing between hospitals.
He also highlights a further difficulty, which is t hat cost comparisons may be
bedevilled by the fact that some hospitals attract more complex and hence
more expensive cases than others. For this reason h e stated that a simple

more expensive cases than others. For this reason h e stated that a simple
comparison of cost per episode of hospitalisation m ay not provide a clear
indication whether one hospital or group is more ex pensive than the other.

24

Further complicating the picture are the attitudes of treating doctors.
Hospitals could legitimately argue that they have n o control over these. But
he also makes the point that hospitals also create the incentives for doctors
to make use of technology in which a hospital has i nvested even where it is
not necessary on a clinical basis to use the technology.

[77] His view nevertheless was that the comparative exercise was still possible
but it would require obtaining data from at least o ne open scheme and one
restricted scheme, done over the period of one year and should exclude the
treatment of the very young and the elderly, as the latter two groups require
higher than average costs and may distort the sample. 61

[78] We decided that the answer to this question wa s insufficiently probative to
the merger to be worth performing in order to meet the rigorous standard
that Green suggested. In short, the complexities of the exercise might still
have yielded a result that would be subject to meth odological quibble, the
data set may in any event not have been readily ava ilable in the form
required, and data may have had to be obtained from less willing sources,
thus adding to delay. We also had to have regard to the fact that the
exercise, if it was to be performed, was coming at a late stage in what had
been a prolonged procedure for the merging parties. Because there was still
no certainty that we could reach a robust conclusio n and because of the
uncertainty over the practicalities of the undertak ing, we concluded that the
prejudice to all concerned of increased cost and de lay outweighed any
possible but uncertain benefit.

[79] For this reason we must rely on what evidence we have in the record to
come to a conclusion based on the reasonable probabilities.

[80] It is probable that the NHN tariff to most fun ders is, overall, lower than that
of Life. The evidence for this comes from the anecd otal testimony of the

of Life. The evidence for this comes from the anecd otal testimony of the
several Commission witnesses who are employed by in dependent private
hospitals. The merging parties, apart from the expe rt testimony of Dr


61 Greens’ full report has been added to the record as Exhibit U.

25

Theron, did not bring their own evidence to contrad ict this. But most
importantly the Gems episode, described earlier, be ars testimony to the fact
that both merging parties understand the tariff iss ue similarly. Life
executives understood and JMH managers did too: if the latter were
threatened by the former with going to NHN tariffs, they would be worse off.
This was the reason why Life adopted the negotiatin g tactic it did and why
after five weeks it succeeded. Life argued that the pressure point was not
the threat of lower tariffs but the inconvenience o f changing to another
negotiator, but we find this unconvincing.

[81] In oral evidence Moloabi was not prepared to c ommit himself to an answer
on this because of confidentiality agreements with hospitals. 62 However, in a
written submission provided to the Commission as pa rt of its investigation
before the hearing, Gems had stated that its tariffs for independent hospitals
were lower than those for Life. 63 In another submission to the Commission,
another funder, Metropolitan expressed the view tha t privately owned
independent hospitals tend to charge lower fees tha n fees charged at a
hospital belonging to one of the three main hospital groups. 64


[82] Moreover, given that Life is larger than NHN a nd that bargaining takes
place nationally, it is likely that this greater si ze would allow it to negotiate
for better tariffs from funders than would NHN.

[83] We find on what evidence we have that it is p robable therefore that NHN
tariffs are likely to be lower to funders than the Life tariff. This does not
answer the question of whether they are less expens ive to funders in terms
of total cost of hospital services. Tariffs are onl y part of the cost to funders,
as we noted earlier. According to the testimony one of the Life witnesses,
Matthew Prior, consumables compromise [...] of the cost to funders of
private hospital costs.
65 He testified that because its hospitals operate

private hospital costs.
65 He testified that because its hospitals operate

62 See transcript 22 May 2012 page 33-34.
63 See record file A 3 page 1106 paragraph 8.
64 See Record page Metropolitan Medical Scheme, p. 1073.
65 Transcript 23 May 2012 page 39.

26

several agreements with funders, in particular Disc overy, where the cost of
consumables is part of the remuneration package, Li fe’s cost to funders is,
when consumables are included, lower in practice th an that of NHN. This is
because of an incentive offered by the funder to pa ss on to the hospital
group some part of any saving achieved by the hospi tal group on
consumables. The funder has its own assessment of w hat the cost of
consumables is and where consumables can be procure d at a cost below
the funder’s expectation, the hospital and funder s hare in the saving
realised, according to a contractually agreed formu la. 66 Prior stated that Life
was a pioneer in adopting these alternative reimbur sement models (‘ARM’s)
in the private hospital sector. His evidence was th at [...] of Life’s income is
based on an ARM not a fee for service basis (‘FFS’). 67

[84] We were told that, on the other hand, independ ent hospitals, and this
would include the NHN, did not operate on such a fo rmula but charge on a
FFS basis and then pass on the cost consumables to the funder at the net
acquisition price (NAP). This being so, they have n o incentive to lower the
cost of consumables to funders. Green also accepts the difficulty of
comparing tariff prices in FFS models with those in an ARM model as the
ARM model includes a wider range of services than t hose contained in tariff
prices. 68

[85] This would have been an issue where evidence f rom funders themselves
would have proved invaluable. Van den Heever, the C ommission’s expert,
for instance challenges the notion that ARMs are a result of competition

66 There are limits to which hospitals may accept the risk in the cost of treatment.Green points out tha t
many ARM agreements have ‘stop loss’ clauses, where if the utilisation of services exceeds an
agreed level, the hospital can bill the patient on a normal fees for service (FFS) basis. See Green

report, exhibit U, paragraph 5.

An example of such an agreement was contained in a contract between Discovery Health and Life
where a formula exists based on what is termed the ‘DRG rate’. See clauses 5.3 to 5.5 of agreement,
record B1 page 1754. .DRG stands for diagnosis rela ted group. DRGs are used to group all charges
for hospital inpatient services into a single 'bund le' for payment purposes. (See McGraw-Hill Concise
Dictionary of Modern Medicine.) By way of example v arious forms of a typical treatment .e.g. a
caesarean might be classified as having levels of s everity and the hospital plan would then set prices
for the activity that vary depending on its level. Discovery makes use of DRG’s for its Keycare option.

67 See Prior’s witness statement paragraph 4.1 Also transcript May 23 page 44.
68 Green report, Exhibit U, paragraph 5.

27

between hospitals and suggests that they exist to s mooth risks over a
period of time for medical schemes. 69 But he does not have the data to give
a view on the comparative issue. Regrettably it was not forthcoming. We are
cautious about suggesting that there is significant reluctance amongst
funders, who are the best informed consumers of pri vate hospital services,
to give evidence, but the experience of several hos pital mergers to date
would suggest that they are loath to divulge eviden ce on what may be for
them competitively highly sensitive information. Mo loabi as we noted earlier
testified that confidentiality agreements prevented him from revealing
certain information. Not only do the funders want t o play off hospital groups
against one another but, as importantly, they do not wish rival administrators
to know what they have won or for that matter lost at the negotiating table.
This reticence came to light in the neutered inform ation gleaned by the
Commission in its investigation, and the hesitancy of funders to be more
forthcoming does not help the Commission perform its function.

[86] We therefore only have the evidence of Life to rely on when considering
whether its costs of consumables are lower to funde rs than those of NHN
hospitals. The Commission’s witnesses in general te rms accepted this
proposition and it also accords with economic likel ihood. Life, being a large
group with buying power, is probably able to secure lower input costs for
itself than NHN.

[87] We are therefore faced with the following conc lusions. If JMH was part of
NHN its tariffs would be lower to funders but not its consumables. We do not
know the extent of the lower tariffs or of the high er costs of consumables.
We do know that pre-merger JMH did not procure its consumables through
Life but on its own. Thus some saving in this respe ct, post merger, can be
expected. If JMH were a larger group and there wer e evidence of a greater

expected. If JMH were a larger group and there wer e evidence of a greater
differential between the tariffs and the costs of c onsumables this might have
been a cause for concern, suggesting a substantial lessening of competition
post merger. But the small number of beds which Lif e will gain from this

69 Transcript 25 May 2012 page 30.

28

merger, the fact that the differential in tariffs w ould be offset by a decrease
in the cost to funders of consumables, the move to alternative funding
models such as designated service providers (‘DSP’s ’), suggest that the
effect of a possible increase in tariffs at JMH wou ld, post merger, as
compared to our pre-merger hypothetical, be slight. (Note that in practice
the merger will not lead to higher prices to funder s as Life has already been
setting JMH’s prices to funders: we are in this ana lysis simply pursuing the
logic of hypothetically independent pre-merger pricing by JMH.)

Other pricing issues

[88] Pilsbury suggested that the merger would possibly lead to increased prices
for non-insured patients. This evidence was of an e ntirely theoretical nature
and we do not consider that the Tribunal is justifi ed in concluding on that
tentative hypothesis alone that this concern is war ranted. It is common
cause that uninsured patients pay higher fees as th ey do not benefit from
the lower tariffs negotiated by funders for their b eneficiaries. There was
however no empirical evidence as to why this class of patient, thought to
presently constitute about [...] of the Life patient base, would be worse off
post merger. 70 It appears that at Life pricing to private patient s is subject to
a discretion exercised by individual hospital manag ers. Managers have the
authority to offer discounts up to a ceiling of R 5 0 000. 71 Wylie’s evidence
was that post merger JMH would move on to the same policy. It is not clear
what the current position is at JMH although Wylie suggests it is similar to
that of Life. 72 There is no concrete evidence that the merger will change this
behaviour. 73 Whilst uninsured patients face the prospect of ver y high prices
from private hospitals there is no evidence to sugg est that this situation will
be made worse by the merger. Expressed differently, there is no evidence

be made worse by the merger. Expressed differently, there is no evidence
that JMH offers uninsured patients better terms than does Life and, even if it
does, that Life has an incentive to alter these ter ms. Uninsured patients

70 Wylie witness statement paragraph 4.6.
71 Transcript 24 May 2012 page 7.
72 Transcript 24 May 2012 page 7.
73 Theron states that there is no reason why the move from 49% to 70% would lead to a change in
incentive to charge private patients more and that if it did, they, being more price sensitive, would
move elsewhere. Econex report page 35.

29

appear to be offered rates that depend on hospital- based considerations
rather than group-based considerations, local capacity at a hospital appears
to drive or at least influence the extent of discounting.

[89] Pilsbury also suggested that Life would not of fer discounts to regionally
based schemes. Whilst he dealt with this as a prici ng issue, we deal with all
the competition issues raised by effects on regional schemes together in the
discussion of non-pricing issues that follows.

Non-pricing issues

[90] The remaining competition concerns raised by t he Commission all
constituted theories of harm that might plausibly a rise as a result of a
hospital merger. The problem was a lack of detail i n making the theories
relevant to this particular merger and taking them outside the realm of broad
competition concerns in private healthcare that are not merger-specific. We
therefore do not consider it necessary to deal with them in any detail.

[91] The Commission alleged that that the merger wo uld lead to decreased
competition for specialists. It is true that privat e hospitals compete for the
presence of medical specialists in practices locate d on the hospital
premises and these specialists, depending on the fu nding model, can drive
demand to a specific hospital or group of hospitals. It follows that if there are
fewer hospitals to choose from this competition wou ld be constrained.
However, as JMH has not competed seriously with its major shareholder
since 1997, it is a matter of speculation what this competition might have
amounted to.

[92] We have evidence that independent hospitals en tered this market since
1997, including one, Ethekwini, whose entry was dri ven by specialists who
left a Life hospital because they were unhappy ther e.
74 A group of

74 Hilcrest with 137 beds was formed in 2011: Passmor e’s witness statement paragraph 4. Ethekwini

has 250 beds. See Bechan’s witness statement paragraphs, 7 and 22.

30

cardiologists had decided to form their own hospita l in July 2008 and have
been successful in doing so.

[93] Subsequent to the hearing, on 21 June 2012, th e Commission received an
email from Mr Asmal in which he attached correspond ence between a
doctor at his hospital, Nu Shifa, which is an indep endent, and an
administrator at Entambeni, a Life hospital, during June 2012. 75 The Nu
Shifa based doctor, a gynaecologist, had asked for admission privileges at
Entambeni as that was what his patients had request ed. The doctor does
not state this, but it appears from other emails th at the patients were
covered by a designated preferred provider scheme w hich limited the
patients to either a JMH hospital, namely City Hos pital, or Entambeni. 76 A
reply came back from the hospital manager at Entamb eni to say that due to
“... operational challenges ” he was unable to offer the doctor admission
rights “ ...at this time ”.

[94] This incident is again anecdotal. If it is mea nt to suggest a more
generalised exclusionary strategy on behalf of Life to deny facilities to
practitioners at other hospitals, the Tribunal woul d require much more
evidence than this single incident gives us. Nor do es the incident establish
merger specificity. For instance, did JMH routinely give outside practitioners
access to its theatres, but possibly, under its new controller Life, be
incentivised not to do so? We have no evidence of s uch a policy existing at
either group. Life in any event denied there was su ch a policy and alleged
that at the time of the request there was no capaci ty to accommodate
another gynaecologist. 77 This new evidence does not suffice to support a
theory that the merger will lead to additional harm to competition in the
quest for specialists.


75 This correspondence forms part of an application t o lead additional evidence brought by the

Commission on 10 July 2012. We explain the context of this application more fully below. This
application is Exhibit V in the record.
76 Ibid page 46,
77 Letter from merging parties to the Tribunal respon ding to the Commission application to lead
additional evidence, dated 17 July 2012, paragraph 16.2.This letter is Exhibit W.

31

[95] Another theory of harm advanced by the Commiss ion’s expert, Pilsbury,
related to the partial ownership of Life in JMH tha t would eventuate, post
merger. The expert testified that post merger Life would have the ability to
channel patients and doctors away from JMH to nearb y Life hospitals where
Life holds a greater economic interest. The problem with this theory is that
Life already has this ability pre-merger given its de facto control. Post
merger its economic interest increases from 49% to 70%. The merger if
anything is likely to make ‘tunnelling,’ as Pilsbur y termed this behaviour,
less likely than it was pre-merger because Life’s e conomic interest is
greater.

[96] Finally, we consider the effect of the merger on regional medical schemes.
The Commission argued that given the number of hosp itals and
concentration that Life had in the greater Durban a rea post merger, regional
medical schemes would be forced to include Life hos pitals if they were to
construct a network of hospitals to which their mem bers could be directed.
The addition of five hospitals to Life’s portfolio would make this increasingly
difficult. The adverse effect, as we understand the Commission’s case, is
twofold. The increased footprint available to Life would be exclusionary for
other hospitals and would furthermore raise costs f or regional schemes
since there would be fewer hospitals to bargain with.

[97] Life contended that there were no truly region al schemes in KZN. 78 It
provided statistics to show that national schemes w ith a high proportion of
members in the KZN area constituted only a minute p ortion of Life’s total
income. When administrators negotiate they do so na tionally and national
rates are set except in the case of hospital arrang ements that are based on
the principle of DSPs. Moreover, the medical scheme administrators who
undertake these national negotiations typically rep resent a number of

undertake these national negotiations typically rep resent a number of
schemes and their outlook and objectives are inclin ed to be national rather
than regional.



78 See Prior witness statement paragraph 2.19.

32


[98] The possibility of harm to regional schemes wa s advanced as a theory by
the Commission and was supported by some of the wit nesses representing
private independent hospital witnesses, but no fund er provided such
evidence. Whilst such a theory of harm may be corre ct it was insufficiently
advanced on the evidence before us. Much of the ora l testimony relied on
the assumption that Life had negotiated exclusive d eals with funders to the
exclusion of other independent hospitals. However t he evidence from Life
was that these are not exclusive agreements but rat her that discounts are
offered if certain targets are achieved. It is not clear from the evidence what
volume of business in JMH is subject to such scheme s. But given JMH’s
limited size - its five hospitals collectively have fewer beds than Netcare’s
single hospital in the area - it is unlikely to hav e impacted on the negotiating
power of Life. Again here we have the evidence of G ems as a natural
experiment. Gems’ Moloabi was unaware that his fund had negotiated with
Life on behalf of the JMH hospitals until he learne d otherwise during his
testimony at the hearing. 79 If JMH had been a significant bargaining chip he
would surely have been aware of this.

[99] Life does not negotiate with all funders. Negotiati ons are time-consuming
so the fund would have to be large to make it worth while. 80 Similarly large
funders do not negotiate with smaller hospitals. La rge funders and large
hospital groups like Life have their own tariffs th at they apply to all those
with whom they have not negotiated a different tari ff. Given that the
probable counterfactual in this case is a negotiati on with JMH as part of the
NHN network, it is not clear from the evidence befo re us whether the latter,
with the former added to it, would have been more l ikely to negotiate with
individual medical schemes to establish its network . It seems again
probable that membership by JMH would not have infl uenced this
possibility.

probable that membership by JMH would not have infl uenced this
possibility.

79 See transcript, 22 May 2012 page 36. See also Exhibit H Life Health Care Billing doc ument for
2011 page 95 where the JMH hospitals are listed in a schedule as associates.
80 Lowick in his evidence referred to the fact that w hen individual negotiations started in 2003 there
were 150 odd schemes although some were represented by common administrators. Transcript 24
May 2012 page 84.

33



[100] We do not know whether, as alleged by some of the Commission’s
witnesses, funders are excluding their hospitals fr om the funders’ DSP
networks because of the superior bargaining power o f the large hospital
groups which negotiate DSPs in such a way as to ens ure the exclusion of
independent hospitals or whether the reason for the ir exclusion is that they
are not price-competitive, as suggested by the merg ing parties. Whilst
independents appear to struggle to get on to these networks except where
they fill a gap in the footprint that a group-align ed hospital cannot fill, we do
not have evidence of exclusionary bargaining by Lif e. The relevant
agreements have been examined by the Commission and whilst they
provide for the equivalent of volume-based discounts they do not provide for
exclusivity – at least insofar as these agreements have been brought to our
attention. The more recent email forwarded as part of the Commission’s
application to admit new evidence does not take thi s aspect any
further. 81 Whatever the ills of the negotiations that may take place between
the hospital groups and the funders to create DSPs it seems unlikely that
the merger will make much of an impact on this.

[101] There is thus insufficient evidence before us that the merger is likely to
have an effect on bargaining with regional schemes. What may have been
more fruitful to explore is why there are no signif icant regional schemes and
if national bargaining is a constraint to their formation. That is a much bigger
question than this merger.

[102] It is clear from the evidence in this merger that the most competitively
significant events since the abolition of centralis ed bargaining are the
annual negotiations that take place between the three hospital groups which
between them represent more than 80 % of the private hospital market in

81 See Exhibit V. Here the email from Robyn Ambler of the Nimas unit of Metropolitan Health to
Wypekema of the NHN, is relied on by the Commission as it purports to show that Metropolitan had
excluded an independent hospital from its network i n favour of a Life hospital. It does not seem to sa y
anything more than that the hospital was not consid ered as the scheme already had an existing
agreement with Life. This does not take issues at t he hearing any further. It does not for instance
even show that Life had insisted on exclusivity when this agreement was made apparently in 2009.

34

the Greater Durban area, and the three largest priv ate administrators which
represent approximately 77% of the administration m arket. 82 According to
Prior, Life derives 80% of its income from five schemes or administrators.

[103] Price formation and other benefits for the la rge number of insured private
patients are thus a result of these negotiations. T he manner in which the
negotiations are conducted is clearly of vital public interest and this case did
not offer an opportunity for that topic to be investigated at any depth.

Increased concentration

[104] The Commission also led Professor Van den He ever to advance the
thesis that increasing concentration in the private hospital market causes a
correlated increase in the real costs of hospital c are.83 The Commission did
not persist with this issue in final argument so we do not deal with it at any
length although it was the subject of extensive cro ss-examination from the
merging parties’ counsel. The problem with the prob ative value of the
evidence was that it is based on an incomplete data set. Theron relying on
later data concluded that the market had become les s concentrated yet the
costs were still increasing. Thus on her evidence t here are reasons other
than concentration for the above-inflationary incre ases. Although Van den
Heever rejected most of her suggestions, he conceded that omissions in the
data available to him were material. 84 We consider that we cannot rely on
this aspect of the evidence to come to the conclusi on that the merger will
lead to an anticompetitive outcome.85


82 For the hospital figures, see Van den Heever’s rep ort, paragraph 9; for the medical scheme figures
see the witness statement of Matthew Prior, paragraph 2.11-2.13.
83 According to Van den Heever, “... indications are that for every 1% change in ma rket concentration
there is a 0.8% real adjustment in hospital costs o ver and above GDP growth. Based on this

there is a 0.8% real adjustment in hospital costs o ver and above GDP growth. Based on this
reasoning a 3.1% increase in national overall priva te hospital costs could result from this merger due
to the 3,8% change in national concentration .” See Van den Heever report, paragraph 9.15.
84 Van den Heever’s data set omits information for th e years 2007, 2008 and 2009. Transcript 25 May
2012, page 9. He conceded that if Theron’s data is correct then concentration should have declined
(transcript 26) but he was not in a position to per suade us why his data set was more reliable than
hers.
85 See 25 May 2012 transcript page 701.

35

General comment on the theories

[105] The theories of harm advanced by the Commissi on are not necessarily
misplaced. However they lacked specificity to the facts of this merger. In the
final outcome it seems unlikely that the pre- and p ost merger scenarios will
differ materially. This is partly attributable to t he limited size of the JMH
hospitals relative to the rest of the Life group. Further, even if the merger
were prohibited, the fact that Life holds nearly 50 % of JHM and that there is
no other major individual shareholder, makes it unl ikely that JMH would,
without the merger, have constituted a competitive threat to Life even if
there had been independent behaviour. Regardless of Life’s status as a joint
controller, third parties dealing with the JMH Grou p, whether customers in
the form of funders or specialists seeking a home f or their practices, are
likely to view a group halfway in Life’s camp as being a Life entity and would
interact with it accordingly, thus dulling its comp etitive potential as opposed
to a hospital group that was wholly independent.

Conclusion on theories of harm

[106] For this reason the post merger counterfactua l is a more nuanced one
than contemplated in argument by either the Commiss ion or the merging
parties. Permitting the merger is more competitivel y significant than
suggested by the merging parties who have to wish a way the history of a
collusive arrangement on pricing. Prohibition of th e merger on the other
hand would be unlikely to lead to greater competiti on and would more likely
lead to a stalemate since Life is not being diveste d of its 49% stake, which
would remain Life’s even if the merger were prohibi ted. Rather, prohibition
would lead to a situation of paralysis; a de facto controller not having
enough of a stake to provide the incentive to inves t further in JMH in order
to improve its service offering, and a collection o f disparate minorities not

to improve its service offering, and a collection o f disparate minorities not
having the economic interest or the degree of influence to provide a decisive
independent force.

36

Admission of evidence after the close of final argument

[107] On the 18 th June 2012 the Tribunal heard final argument from t he
Commission and the merging parties. Ordinarily that would have marked the
end of the hearing. Two events altered that typical scenario. As we noted
earlier the Tribunal decided after the hearing to a ppoint an expert in the
form of Dr Green to provide a report on a particula r issue, and the
Commission sought to introduce a further item of ev idence. 86 The
introduction of both Dr Green’s report and the furt her item of evidence from
the Commission was opposed by the merging parties a nd hence we had to
rule on the issue. We have decided to admit both it ems into the evidence.
They now constitute Exhibits U and V in the record.

[108] The argument from the merging parties was tha t as the hearing had
terminated after the hearing of final argument, the Tribunal had no discretion
to admit further evidence; alternatively that it wa s unfair and prejudicial for
evidence to be admitted at that late stage of proce edings given that the
merger had been pending for a considerable time since notification.

[109] As it happens this was a storm in a teacup. N either item bolstered the
case for prohibition. One, Dr Green’s report, clari fied for the panel the
complexities of a comparative exercise between the pricing of Life and
NHN. The second, the additional information provid ed by the Commission,
was a submission of email exchanges, open to differ ent interpretation, and
insufficiently probative to change our conclusions on the two issues it
concerned viz. the effect on competition for specia lists and the alleged
exclusion by funders of independent hospitals at the behest of Life .87


86 The Tribunal wrote to all the parties to advise th em of this on 27 June 2012. On 29 June 2012 the
Commission advised it would be bringing an applicat ion to introduce new evidence but first wanted to

engage the merging parties. The application was fil ed on 10 July 2012. The merging parties were
given leave by the Tribunal to file their answer by way of a letter instead of an affidavit. This was
received on 17 July 2012.

37

[110] The Commission’s application was brought not as its own desire but at
the request of two of the witnesses to these proceedings, Messrs Asmal and
Bechan. Both forwarded this correspondence to the Commission after the
final argument in the hearing. 88 The substance of these two issues has
already been dealt with above and does not require further elaboration. We
now deal only with the procedural challenge raised by the merging parties
as to the Tribunal’s entitlement to admit such evid ence after final argument
had been delivered at the hearing.

[111] The argument advanced by the merging parties is that once closing
arguments have been concluded the proceeding has co me to an end and
no new evidence may be admitted. For this legal pro position several cases
involving civil litigation were invoked. That analo gy is the problem. This is a
merger proceeding, not adversarial civil litigation . Secondly the Tribunal
does not sit as neutral referee between sparring li tigants, but exercises, as
the Act and numerous past decisions have recognised , inquisitorial powers
that distinguish it from the function of the passiv e adjudicator in adversarial
civil proceedings. Thus understood the notion that the conclusion of final
argument constitutes some termination of proceeding s by way of a final
curtain that cannot be lifted thereafter to admit h ighly relevant evidence is
unsound.

[112] Indeed, as the Competition Appeal Court recen tly emphasised in its first
decision in the Wal-mart/Massmart merger, the Tribunal is not only entitled,
but is duty bound, if it considers it does not have sufficient evidence in the
record before it, to use its inquisitorial powers t o glean the evidence. As the
Court there observed “ Merger hearings, the object of which is to determin e
whether a merger can be approved, should not be stu ltified by an excess of
formalism or of procedures best suited to a trial .”
89

89


88 The hearing concluded on 18 May 2012. The Commissi on received the emails on 21 and 22 June
2012 respectively.
89 Minister of Economic Development and others v Compe tition Tribunal and others ( Wal-mart )
Competition Appeal Court Case No: 110/CAC/Jul11 and 110/CAC/Jun11 page 50 paragraph 85.

38

[113] This is because in merger cases particularly, as the Court explained, the
Tribunal is bound to come to its own conclusion abo ut the merger’s
consequences. Section 12A, which sets out the subst antive issues the
Tribunal takes into account in considering mergers, requires the Tribunal to
“determine ” the matter.


[114] This of course does not detract from merging parties’ rights to make
submissions about their concerns for expeditious co nclusion and to express
a view as to whether a particular endeavour propose d by a panel to secure
more evidence may be misguided. This is precisely h ow we attempted to
manage the use of our independent expert before the merging parties
objected to the exercise. Once it was clear that the merging parties objected
to this endeavour we proposed a discussion at a pre -hearing on the issue of
admission of Dr Green’s report and of the new evide nce which the
Commission sought to place on record. The merging p arties then recanted,
reversing their original insistence that we rule on the application as a prior
step. They chose instead to make submissions by wa y of a letter to the
Tribunal.

[115] The posture of the merging parties proved mos t counter-productive to
their own cause. A matter that could have been reso lved expeditiously by
way of a pre-hearing and a possible meeting of the respective experts, as
the Tribunal proposed, was prolonged and elicited a number of lengthy legal
submissions on issues of procedure rather than the merits. Ultimately all
that was achieved was a delay in admission of this further evidence.

Contravention of section 4(1) of the Competition Act

[116] The Commission announced at the commencement of the hearing that it
was investigating whether the conduct of the mergin g parties in engaging in
joint pricing constituted a contravention of sectio n 4(1) of the Act.
90 We do
not in this decision need to comment further on that issue.

90 See transcript 21 May 2012 page 23.

39

Lack of proper disclosure to the Commission

[117] In this merger, as noted earlier, we have acc epted the version offered by
Life that it has de facto controlled JMH since its acquisition of its 25%
interest in 1997 although this did not amount to de jure control. This was the
evidence of Mr Wylie and we assume it represented t he considered view of
Life. This version however is at variance with pri or information given to the
Commission by Life’s predecessor, Afrox, in interac tions with the
Commission on at least two prior occasions, the not ing of the Amahosp
merger and the requests for an advisory opinion in respect of the purchase
of the further 24% stake by Life in JMH in 2003. Th ese events have been
described above in this decision.

[118] These are matters that the Commission needs t o investigate. The
following provisions of the Act are relevant in thi s respect. Firstly section
15(1), which provides that the Competition Commissi on may revoke its own
decision to approve or conditionally approve a smal l or intermediate merger
if the decision was based on incorrect information for which a party to the
merger is responsible, or the approval was obtained by deceit. This
provision must be read with section 16(3) of the Ac t, which makes it
applicable, with the changes required by context, t o a large merger.
Secondly, section 73(2)(d) which states that a pers on commits an offence if
he or she knowingly provide false information to the Commission.

[119] In addition merging parties are required to c omply with section 13(1) of
the Act, read with the rules of the Commission, whi ch prescribe the
information to be provided with a merger filing. T hat information includes a
list of firms controlled by the merging parties.





PUBLIC INTEREST

40

[120] There are no public interest considerations i n terms of section 12A(3) of
the Act arising from this merger that would alter our decision to approve the
merger.

CONCLUSION
[121] The merger will not bring about a substantial prevention or lessening of
competition for the reasons advanced. There were no substantial public
interest considerations raised in this merger. The merger was therefore
approved without conditions on 24 July 2012.



____________________ 24 October 2012

N Manoim Date

Lawrence Reyburn and Medi Mokoena concurring.


Tribunal Researcher: Londiwe Senona / Songezo Ralarala

For the merging parties: D.N. Unterhalter S.C. and G.D. Marriott instructed by
Webber Wentzel

For the Commission: D.I. Berger S.C. and N.J. Jele instructed by the State
Attorney