COMPETITION TRIBUNAL
REPUBLIC OF SOUTH AFRICA
Case no.: 11/LM/Mar05
In the large merger between:
Medicross Healthcare Group (Pty) Ltd
and
Prime Cure Holdings (Pty) Ltd
_____________________________________________________________________
NonConfidential Reasons for Decision
_____________________________________________________________________
Order
1. The Tribunal issued an order on 15 September 2005 prohibiting this merger.
Our reasons for the order follow.
The Transaction
2. The transaction on which the Tribunal has ruled envisages that Medicross, a
company discussed below, will acquire the entire share capital of, and loan
claims against, Prime Cure, also a company discussed below. firms are
managed healthcare companies providing primary care healthcare services to
medical aid schemes through a network of doctors or “service providers”.
Postmerger, Prime Cure will become a whollyowned subsidiary of
Medicross.
The Parties
3. The primary acquiring firm is Medicross Healthcare Group (Pty) Ltd
(“Medicross”). Medicross is owned as to 80% by Network Healthcare
Holdings Limited (“Netcare”), which is listed on the JSE in the health sector,
with the remaining 20% being held by Netpartner Investments Limited
(“Netpartner”). Netpartner is owned by doctors and other healthcare service
1
providers and holds a strategic shareholding in Netcare as well as having other
functions. One of them is the ownership of 100% of the share capital of
Netcare Direct Managed Care (Pty) Limited ("Netdirect"), a company which
undertakes risktransfer managed care a concept explained at some length
below.
4. Each of these entities is described more fully in the section below dealing with
the structure of the Netcare group,
5. The primary target firm is Prime Cure Holdings (Pty) Ltd (“Prime Cure”). Its
shareholders are:
Brait Private Equity 34.2%
Praxis Private Equity 29.30%
CDC Financial Services Mauritius Ltd 11.3%
Total Support Management (Pty) Ltd 8.22%
Other minority shareholders 16.95%
6. Prime Cure’s subsidiaries include Prime Cure Health (Pty) Ltd, Prime Cure
Management Services (Pty) Ltd and Prime Cure Occupational Wellness (Pty)
Ltd. Prime Cure operates chiefly in that part of the healthcare industry serving
lowincome patients.
7. Prime Cure manages and administers 45 primary healthcare centres or clinics.
They are located in residential townships or industrial areas conveniently
accessible to the homes or workplaces of lowincome earners. Prime Cure
centres accommodate 47 general practitioners (GPs), and nurses play a large
role in providing primary healthcare services at these centres.
8. The healthcare professionals at Prime Cure's centres are grouped into
“incorporated practices” and are charged rental by Prime Cure and fees for
equipment use and staff and administration costs. These practices treat patients
who are members of the medical schemes with which Prime Cure has
managed care contracts, and also other patients who have no connection with
Prime Cure and who may or may not be members of medical schemes.
9. The Prime Cure healthcare centres receive about 800,000 patient visits
annually.
annually.
10. Prime Cure also manages and administers four independent GP practices
which operate outside its healthcare centres.
11. In addition to these operations, Prime Cure has a contractual network of some
2,000 GPs and 800 associated healthcare professionals, such as dentists,
optometrists, and specialists, who provide medical services to members of the
medical schemes which have managed healthcare contracts with Prime Cure.
12. Prime Cure's managed healthcare contracts cover primary, primary plus
2
secondary, or fullrisk (i.e. primary plus secondary plus tertiary) capitation. 1
Capitation is a concept explained below.
13. Prime Cure has managed care contracts with 24 medical schemes, covering
approximately 115,000 "lives" (principal members plus dependants). 2 We
emphasise that, in contrast to Medicross, these contracts are chiefly designed
for health insurance options having lowincome earners as members, although
there is evidence that Prime Cure has made an entry into the "buydown"
market, serving higher or middleincome earners. 3
14. Prime Cure is also involved to some extent in providing occupational health
services under contract to employer organisations. These services comprise
undertaking medical examinations and ongoing health screening, issuing
health certificates, and operating employeeassistance programmes, onsite
clinics and related services. Prime Cure's turnover from this business is not
large.
The Netcare Group Structure
15. Disentangling the complex web of crossholdings and management contracts
in the Netcare group is not a simple matter – in fact, Mr. Pieter Dorfling, an
executive director of Medicross and CEO of Netdirect, and one of the
witnesses at these hearings, often had to explain which hat he was wearing. 4 It
is nevertheless clear that the acquiring firm, Medicross, is a member of a
group of companies that has the Netcare private hospital network at its centre.
16. The Netcare group comprises a number of companies supplying a large range
of healthcare services. Among these are private hospitals and specialised
clinics, pharmacies located within the hospitals, emergency ambulance
services, pathology and dialysis units, and the compilation and dissemination
of healthcare management information. For the purposes of evaluating this
of healthcare management information. For the purposes of evaluating this
transaction – as will become apparent later on the Netcare activities that are
of particular pertinence are, firstly, the network of 64 private hospitals which
is the largest in the country and which we shall refer to as Netcare. Secondly,
Medicross, the clinic network which provides a range of primary care
services and is the primary acquiring party in this transaction. Thirdly,
Netdirect, a managed care company which has been in existence since June
2003, and which targets lowcost medical scheme options. All these entities
are discussed more fully below.
1 Commission's recommendation, File A, page 7 Primary healthcare can be regarded as the service
provided by GPs or other professional service providers with whom patients first engage when seeking
treatment. Secondary healthcare is provided by specialists, generally on referrals from the primary
level. The tertiary level of healthcare is made up of hospitals and specialised clinics.
2 Transcript page 369 the evidence of Mr Hodge, the Commission's expert witness, of Genesis
Analytics. See also the Genesis Report, p 2, and Transcript p 386. Genesis remarks that Prime Cure
increased its GP network from 491 to 2,000 within one year.
3 Transcript pages 368370
4 Dorfling testified that he was accountable to the boards of Medicross, Netpartner and Netdirect.
3
Netpartner
17. Netpartner is owned as to 48% by Netcare and as to 52% by healthcare
professionals who numbered 9,000 at the time of the Commission's
recommendation. Netpartner, with a holding of 16.2% in Netcare (at the time
of the merger notification – it had risen to 17,5% at the time of the hearings) is
the largest single shareholder of Netcare. Netpartner is controlled on the
operational level by Medicross in terms of a management agreement. 5
18. It has several functions, one being to serve as the entity through which doctors
and other medical service providers associated with Netcare hold shares in the
group, and it also acts as an assembly point for these professionals in their
relationships with Netcare.
19. Netpartner owns all the issued shares of NetDirect.
Netdirect
20. In the sphere of managed healthcare Netpartner operates through its
subsidiary, Netdirect, which is the Netcare group's “entry vehicle” into the
lowcost end of this arena. 6 Netdirect is controlled by Medicross in terms of a
management contract. 7
21. The merging parties supplied remarkably meagre information on Netdirect
when filing their merger notification. No information whatsoever was
proffered on the existence and extent of Netdirect's operations and Netdirect's
focus on lowcost medical scheme options, at least in the version of these
documents which reached the Tribunal. The only evidence available is that
Netdirect has a network of more than 800 doctors. 8 The significance of
Netdirect, as the Netcare group's existing contracting entity for lowcost
medical scheme options, emerged only during the course of the merger
hearings.
22. We deal more with Netdirect below when describing the activities of
Medicross.
Medicross
Medicross.
Medicross
23. Medicross is a primary healthcare entity engaged in four areas of activity:
a. providing primary healthcare through the operation and administration of
medical centres;
5 Transcript page 367 and record page 20
6 The specific business activities of the acquiror and the acquiree will be dealt with below.
7 Transcript page 367. Note that in the parties’ competitive analysis no mention is made of Netdirect.
8 Transcript page 368. Also see parties’ reply at File C page 18 to a query by the Commission.
4
24. Medicross has 53 centres (clinics) around the country, specifically targeted at
higher income patients. GPs, dentists, optometrists, pharmacists and other
healthcare professionals work at these centres. Most of these centres have day
theatres for minor surgical procedures, and other services ancillary to primary
healthcare, such as radiology and pathology, are available.
25. The GPs and other medical professionals working at these centres are not
employed by Medicross and not obliged to work exclusively at Medicross
centres. They are however required to comply with clinical guidelines
specified by Medicross. Medicross centres accommodate 413 GPs and 153
dentists.9
b. practice administration services
26. Medicross provides administration services to 21 independent medical
practices in return for fixed monthly management fees.
c. development of clinical guidelines and disease management programs
27. These are services aimed at maintaining consistent clinical standards at
Medicross centres.
d. Managed care services
28. From Medicross' merger documentation as filed with the Commission, it
emerges that Medicross has managed care contracts with 15 medical schemes,
covering some 35,000 lives, and extending to a multipart formula
arrangement for the remuneration of service providers. Apparently these
activities fall considerably short of capitated primary care.
29. Both Mr Strauss, the witness from Discovery who negotiates directly with the
Netcare group with respect to managed care contracts, as well as counsel for
the Commisson, referred to the confusion which exists regarding the identities
of Medicross and Netdirect in relation to their respective product offerings.
The record shows the difficulties experienced by all concerned in establishing
The record shows the difficulties experienced by all concerned in establishing
the boundaries between Medicross’ activities and those of Netdirect.
30. As indicated above, Medicross manages the operations of Netdirect under a
management contract, and Medicross and Netdirect contract with medical aid
schemes to provide various forms of managed care. Netdirect appears to be at
least the nominal contracting party in relation to lowcost medical scheme
options. Medicross handles the administration of all the arrangements and the
management of relationships with healthcare service providers with whom the
scheme members consult.
9 Record page 240241
5
31. According to Mr Dorfling, there is little overlap between the roles of
Medicross and Netdirect. NetDirect utilises the Medicross managed care
infrastructure for its operational capability. He describes Netdirect as a
“facilitator of risk transfer, network arrangement, network management”
whilst Medicross is a practice management administrator, deriving the bulk of
its income from practice management. 10 He distinguishes Netdirect and
Medicross thus:
“Medicross fulfils the administration function, 100% correct. But again,
Medicross never entered into the kind of total risk taking environment where
that is the core objective of NetDirect. But the management capabilities and
functionalities performed by Medicross in terms of an administrative
agreement, correct.” 11
32. Unfortunately the Tribunal does not have information about the scale of
Netdirect's activities in conjunction with Medicross in terms of number of
lives or turnover.
The Rationale
33. Medicross believes that there is likely to be significant growth in demand for
managed care services for those medical scheme options servicing lowincome
earners. This predicted growth in demand will derive from government’s
efforts to extend medical insurance to its lower income employees and from its
stated commitment to a system of ‘social health insurance’ which is intended
to extend health insurance coverage to uninsured South Africans in the lower
paid part of the labour market. 12
34. In this context, then, Medicross advances two primary reasons for the
transaction. First, Prime Cure brings to the merged entity an established base
of lowincome lives covered. This would enable the merged entity to, in the
words of Mr. Dorfling, ‘hit the ground running’. As we shall demonstrate, a
words of Mr. Dorfling, ‘hit the ground running’. As we shall demonstrate, a
new entrant’s ability to weather the initial period where it builds a base of
lives sufficient to incentivise its doctors to assume risk or, even to offer
discounted fees, is a critical determinant of sustainable entry. Secondly, the
merged entity will be able to take advantage of Prime Cure’s established
relationships with the medical schemes and with institutions, notably trade
unions, which are an important source of access to the medical schemes. In
short, as we shall elaborate below, Medicross believes that the transaction will
enable it to overcome two critical barriers to entry, viz, ‘lives’ and what we
will refer to as ‘social capital’ being a web of relationships, notably with trade
unions, but also, as we shall elaborate, with doctor networks. Note that the
acquiring party makes it absolutely clear that its entry is not dependent upon
10 Dorfling states that only some 5% of Medicross' income has come from its managed care functions.
Transcript page 1104
11 Transcript pages 1103 1105
12 Record page 2456
6
the merger. That is, it can, through Netdirect, enter the market anyway. 13
35. We were informed that the institutional investors who own the significant
majority of Prime Cure’s shares wished to exit from their investment. 14 It
appears that Prime Cure’s shareholders have, over the past three years, been
involved in a number of discussions regarding the possible sale of the
company. These included farreaching discussions with the large hospital
group, Afrox Healthcare (since renamed Life Healthcare). There have also
been several detailed discussions with Care Cross, Prime Cure’s largest
competitor. There were also discussions in September 2004 with
empowerment parties who had expressed an interest in taking control of Prime
Cure.
36. However, as we show below, the stated rationale neglected to outline the
unitary interests between the three chief protagonists in the Netcare Group.
The unitary interests of Medicross, Netpartner and Netdirect
37. The merger notification and competition analysis documents filed by
Medicross, again in the form in which they reached the Tribunal, convey the
impression that Medicross will have to make an entirely fresh entry into the
lowcost area of managed care operations, relying entirely on Prime Cure to
achieve this entry, and that Medicross alone will benefit from the merger. Yet
it is clear to the Tribunal that for the purposes of competition analysis in this
merger there is no distinction between Medicross, Netdirect and Netpartner.
They all operate in unison to enhance the commercial interests of the Netcare
group. Further, Netdirect is already a competitor in the lowcost part of
managed healthcare. 15
38. The true picture emerges only from such documents as the Netpartner
prospectus, dated August 2003, in which the overall strategy of entry into low
prospectus, dated August 2003, in which the overall strategy of entry into low
cost managed care by group entities is announced, 16 and from the annual
report of Netcare for the year to September 2004, where Netpartner's
"substantial progress in developing its business model" is described in the
operational review of the group. 17 This progress is said to have included the
formation of Netdirect, a company which "has facilitated the assembly of a
national network promoting and providing managed care products ". Netdirect
is recorded as having concluded contracts with Discovery, Liberty,
Momentum, and Eclipse medical schemes, all of which were to become
13 The merging parties state, via their expert, that it will take 1836 months to establish a viable
presence in this market. Later Dorfling on behalf of Netdirect, states it could take 34 years.
14 Transcript page 855 and page 1184
15 The evidence of Dr Stillman, the merging parties’ expert witness, coming some time after the filing
of the parties' merger documents, is however posited on the seamless identity of interests and conduct
of these companies in relation to the merger, and the established participation of Netdirect in this arena.
Transcript pages 751753
16 File C of the record, at pp 181 et seq
17 Starting at p. 103 of File A of the record.
7
operational in January 2005.
39. In response to a letter from the Commission, Medicross' attorneys produced
the standard agreement which apparently prevails between Netdirect and the
healthcare service provider groups with which Netdirect contracts. This is a
comprehensive document clearly indicating a bias towards the interests of
Netpartner.18 The names of what appear to be Netdirect's customers or
putative customers, and certain details of transactions or impending
transactions with them, are also revealed in this document. 19 These entities
include Day 1, Ingwe, MomentumPulz, Transmed, Nimas, Pathfinder,
Spectramed, Protea, Siswe, Xpresmed, Bestmed, Discovery, Medicross itself,
Liberty, and Eclipse (various options). It is not clear to us that agreements
with all these entities were concluded, but clearly an ambitious programme of
action was under way, growing from the initial contractual base described in
the annual report for the year to September 2004.
40. In the light of this information the statements of Medicross in its merger
notification documents about the rationale for the merger and of the motives
and competitive position of Medicross, described below, must be treated with
extreme caution.
41. As an example of the contrast which pertains between the merger notification
and less guarded expressions of the role players, we have taken note of a due
diligence report on Prime Cure prepared by a committee on which various
Netcare group entities including Medicross, Netpartner and Netdirect were
represented.20 Mr Dorfling, the CEO of Netdirect and an executive director of
Medicross, was one of the committee members, and he was crossexamined
about some of the information in this report. One of the pages of this report is
headed "1. Purpose", and under its first bullet point it contains the statement
that:
headed "1. Purpose", and under its first bullet point it contains the statement
that:
"the purpose of the limited due diligence performed by [Medicross] was to gain
an understanding of the business processes of [Prime Cure] and to determine to
which extent Primecure [sic] could add value to the current Medicross service
offering with a specific view to enhance the Managed Care capabilities of
[Netdirect] as envisaged in the Netpartner venture ". (our emphasis)
42. It is also significant that the original sale agreement for the merger was
entered into between NetPartner and the shareholders of Prime Cure. That
agreement enabled NetPartner to nominate another party to undertake the
transaction and that party subsequently came to be Medicross. 21 Medicross
was apparently chosen at the last moment, for no other reason than that it had
18 See record, File B, page 221, clause 2.5.2, which lays down that providers who are not shareholders
of Netpartner will be penalised by a monthly deduction of R180 from the remuneration due to them
from Netdirect.
19 Record File A pages 229243
20 Record File B pages 276352
21 Transcript page 870
8
the available cash. Again, the importance of the identity of the acquiring party
is revealed to the Netcare group.
43. The triangular symbiosis between these companies in regard to the merger is
thus clearly revealed.
44. In the parties' market and competitive analysis report, forming part of the
merger notification, the rationale for the merger is stated in paragraph 2. 22
After mentioning possible backoffice savings, this document states that the
principal reason for Medicross' interest in acquiring Prime Cure is Medicross'
belief, shared by Prime Cure, that " administration of managed care is an
important growth area " and Medicross' belief that " its expansion in this area
will be more costeffective if it can build on the relationships that Prime Cure
has already established with medical schemes (as opposed to having to build
these relationships from the ground up) ". Some other or related purposes are
also given, in anodyne terms, none revealing that Medicross is already
extensively immersed in risktransfer managed care in the lowincome end of
the market through its triangular relationship with Netdirect and Netpartner.
45. On the merging parties' own evidence, Netdirect is already providing risk
transfer managed care to lowcost options. It has stated its intention to offer
both primary and secondary managed healthcare for lowincome earners in a
tender to the Government Employees' Medical Scheme (GEMS). 23 Dorfling
states that Netdirect’s product offering is more comprehensive than that of
Medicross in that it involves a primary through to a tertiary product offering,
whilst Medicross has only a primary care offering. 24 Whatever these
differences, the purpose of the merger is clearly not as described in the parties'
merger notification documents but rather to extend the combined strength of
merger notification documents but rather to extend the combined strength of
Medicross and Netdirect (and hence also Netpartner) in generating business in
lowcost managed care. The contradiction extends beyond the stated purpose
of the merger to the assertions made on behalf of the merging parties at the
hearing that Medicross operated in a different relevant market from Prime
Cure in that Medicross was not active in the lowcost end of the managed care
market.25
The Hearing
46. The Commission recommended that this merger be prohibited. The
Commission’s recommendation was filed with the Tribunal on 30 June 2005.
A prehearing meeting was held on the 13 July 2005. We note, for the record,
that at this prehearing meeting the merging parties asked that we order the
22 Record File A pages 245246.
23 File D at page 1667
24 Transcript page 1112. Netdirect has entered into a contract with Discovery
for an option known as the Netcare Plus plan, commencing in 2005, by which
the primary, secondary and tertiary care risk was assumed by Netdirect.
25 Transcript pages 6
9
Commission to hand over all the notes of the Commission pertaining to its
interaction with third parties. The parties were told at the prehearing that this
request – which embodied potentially farreaching consequences for the
exercise by the Commission of its investigatory powers – could only be
determined on proper application before a duly constituted panel of the
Tribunal. An application to this effect was then filed and was heard by the
Tribunal on 29 July 2005. In the course of the hearing, the applicants
withdrew their application.
47. The hearing took place on 11, 12, 15, 16, 17, 19, 31 August and 1 September
2005.
48. The following witnesses were called by the Commission:
Dr Reinder Nauta, of Carecross
Mr David Strauss, of Discovery
Mr Brian Davidson, of Life Healthcare Group
Mr James Hodge, of Genesis Analytics, the Commission’s expert
49. The following witnesses were called by the merging parties:
Dr Laubscher Walters, of Medscheme
Dr Robert Stillman, of Charles River Associates (CRA), the parties’ expert
Mr Sean Patterson, of Brait Private Equity and a director of Prime Cure
Mr Bisnard , Managing Director of Sizwe
Mr Pieter Dorfling, Executive Director of Medicross and CEO of Netdirect
50. The Tribunal called the Council for Medical Schemes which was represented
by Mr. Stephen Harrison and Mr. Alex van den Heever.
Competition Analysis
The Healthcare Environment
51. Section 12A(2)(e) of the Act provides that when determining whether or not a
merger is likely to substantially prevent or lessen competition we should take
account of ‘the dynamic characteristics of the market, including growth,
innovation and product differentiation.’ We will indeed do this when we turn
to a detailed examination of the impact on competition of the transaction
before us. However, because there are several broader ‘dynamic
before us. However, because there are several broader ‘dynamic
characteristics’ that impinge significantly on the markets implicated in this
decision, we thought it appropriate to outline aspects relevant to the general
environment in which healthcare is provided before proceeding to a detailed
consideration of the transaction itself. Pertinent to our consideration are the
general state of healthcare provisioning in South Africa, the policy objectives
of the South African government in the realm of healthcare provision, the
mechanisms whereby government intends achieving those objectives, and the
place and role of the private sector, including the merging parties and many
10
others who participated in these hearings, in this wider context.
52. The provision of adequate health care to all the citizens of the country is
clearly an important plank in government’s efforts to tackle poverty and
inequality. High and middle income South Africans (and this would include a
significant proportion of those in employment) receive healthcare through
South Africa’s sophisticated private healthcare system comprising the full
gamut of general practitioners, specialists, hospitals and pharmacies. Private
healthcare is funded by an array of medical schemes serviced by the
administration companies, data processing companies and managed care
companies that are an integral part of South Africa’s sophisticated ‘first world’
private healthcare system.
53. However the majority of the population – and this includes a significant
number of those in the lower reaches of formal employment – rely on the
public health system for meeting its needs. The reality – and possibly the only
agreed certainty in the fraught debate surrounding the provision of healthcare
in South Africa – is that the private healthcare system, and notably, although
not exclusively, the private hospital network, is characterised by significant
excess capacity, while the public healthcare system is simultaneously
resourceconstrained and increasingly unable to cope with the demands made
of it. A major thrust of government’s efforts to improve healthcare
provisioning is thus to utilise the excess capacity in the private healthcare
system, the better to reduce the demands on the public system, to, in other
words, move a strata of those presently reliant on public healthcare over to the
private healthcare system.
54. The constraint in effecting this movement of people from public to private
54. The constraint in effecting this movement of people from public to private
healthcare is finance. Put simply, the vast majority of those who are presently
reliant upon the public healthcare system cannot afford to fund private
healthcare. They cannot, in other words, afford the monthly premiums charged
by any of the variety of healthcare insurance schemes available on the market.
From a public policy perspective the upshot of this funding constraint is that
the public sector is increasingly incapable of delivering quality healthcare to
those who rely upon it while the private sector remains, as it were, structurally
overcapacitated. From the perspective of the private sector, it is unable,
despite its excess capacity, to service a potential market of millions of South
African whose healthcare needs are not adequately catered for by the public
sector. Hence medicals schemes and the array of services that cluster around
them, have reached the limits of their market (the number of principal
members of medical schemes has remained stagnant at approximately 7 000
000 for some nine years26) and some important elements of the private system
– notably the hospitals – are characterised by significant excess capacity.
55. The challenge then is to devise funding arrangements affordable to a large part
of that portion of the population that presently utilises the public healthcare
26 Commission’s Recommendation page 8, citing the Council for Medical Schemes annual report
20034
11
system. These would be those thousands employed in the lower reaches of the
public and private sectors as well as the selfemployed in both the formal and
informal sectors – that is to say, the target group for private healthcare
provisioning is not the indigent, but it is certainly the poor, euphemistically
dubbed the ‘lowincome’ sector of the population.
56. The firms involved in this merger are at the centre of this challenge precisely
because they are in the business of designing and implementing affordable
models of adequate healthcare provision, models that will enable the medical
schemes to charge affordable monthly premiums and guarantee a level of
service that the consumer is willing to purchase, that the providers are willing
to supply, and that the relevant regulatory authorities are willing to sanction.
57. Despite millions of potential customers, the existing market in lowincome
healthcare provision is extremely small. Certainly, low income earners rely
almost entirely on the public hospitals for secondary and tertiary care and
although much of the primary care received by this part of the population is
from private sector providers this is, in the vast majority of instances, directly
funded by the consumer. For the most part, low income earners are not
members of medical schemes and hence they are not insured for medical
events.
58. A relatively insignificant proportion of the lowincome population have
bought into the few funds whose design attempts to accommodate them. The
simple reason for this is that they are, by and large, still not priced at levels
that lowincome earners can afford. Enter then, private companies dedicated
to providing a service that is aimed at decreasing the cost of healthcare
insurance. This they do through attempting to address the two drivers that
insurance. This they do through attempting to address the two drivers that
underpin high health insurance costs, namely the cost of healthcare products
and services, and, secondly, the cost of risk.
59. Enter too government, driven by public policy imperatives and capacitated,
first, by its huge purchasing power in the shape of many hundreds of
thousands of uninsured public sector employees, second, by its ability to
subsidise the provision of affordable healthcare both to those within its own
employment net and to other lowincome consumers of healthcare services,
and thirdly by its ability to regulate healthcare costs and medical insurance.
60. The state’s principal initial instrument is the Government Employees Medical
Scheme (“GEMS”) a medical scheme comprising a bouquet of high and low
income options which has recently been registered and which government
intends rolling out from the beginning of 2006. GEMS is intended to cover
those of its employees who are currently insured with other schemes – it is
estimated that there are upwards of 60 medical schemes that provide coverage
for government employees as well as those, predominantly lowincome,
government employees who are presently uninsured and, as such, are
dependent on the public healthcare system. There are currently about 1.1
million lives in the public sector. It is estimated that some 400 000
12
government employees are not covered by any medical scheme. It appears
that the intention is to phase in GEMS, focusing first on migrating already
insured government employees from their current schemes to GEMS. The
second phase will focus on persuading those government employees who are
presently uninsured to take up one of GEMS’ lowcost options. 27 We were
informed that government intends membership of GEMS to be voluntary –
however a subsidy will not be made available to employees who choose to
belong to a scheme other than GEMS.
61. Government has recently published the GEMS tender document. It calls for
bids for the administration of five options, two of which are directed at low
income employees. These latter specify that the bidder is required to provide
capitated care. The significance of this will become apparent later but, suffice
for the present to note, that this is a scheme where risk is transferred from the
scheme to the managed care provider or, ultimately, the providers of medical
care, namely the primary care providers, the specialists and the hospitals. The
GEMS tender calls for businesses to bid for eight different types of contracts
an administrator; a clearing house; two providers of primary healthcare
services ( for the “Sapphire” and “Topaz” options , which are described later
on28); an HIV/AIDS management company; a hospital service provider;
managed care services and information technology services.
62. This complex background impacts powerfully upon the competition analysis
of this merger. All merger regulation is, by its very nature, speculative, or, in
the wellknown description of Judge Richard Posner, ‘predictive’. 29 Anti
trust decision makers have attempted to lend as much science and certainty as
possible to the process of merger regulation by utilising evidence of past and
current market behaviour in tandem with economic theories and tools, which,
current market behaviour in tandem with economic theories and tools, which,
in combination, permit of intelligent and relatively reliable prediction.
63. However this merger is characterised by particularly severe analytical
problems. We are, in essence, dealing with a new market. As already
outlined, the poor have been effectively excluded from the market for private
healthcare services by the inability of the health insurance sector to devise
affordable funding options and by the inability of the healthcare service
providers to reduce costs to the sort of levels that enable the supply of a
marketable package of healthcare services.
64. This is not to say that cost has not been a consideration in the supply and
funding of private healthcare in the upper and middleincome markets. Of
course it has been and cost considerations confronted in these markets
underpin many of the debates – as well as the institutions and instruments –
that are pertinent to the current efforts to extend these products to the poor.
27 See transcript page 321
28 Both of these seem to be primary care options which allow for transfer of risk to the MCO. The
differences between them are set out at transcript pages 113940
29 See Judge Posner’s remarks in Hospital Corporation of America v. Federal Trade Commission 807 F.2D
1381 (1986).
13
These are all instruments of ‘managed care’. Hence efforts to discourage the
ultimate consumers of healthcare products from ‘overutilisation’ have been
led by the introduction by Discovery Health – South Africa’s largest medical
schemes administrator – of the ‘savings account’ concept. A variety of
managed care concepts aimed at disincentivising ‘overprovision’ by the
service providers – these ranging from preauthorisation for hospital services
through to the identification of designated (and, therefore, often discounted)
service providers that rely on the utilisation of networks of GPs, specialists
and hospitals – have, in relatively recent years, become ubiquitous features of
current healthcare provision.
65. However, none of these interventions have brought private healthcare within
the range of lowincome consumers. Indeed the record shows that they have
not been particularly successful in holding down the costs of private
healthcare although it is, of course, extremely difficult to construct the
counterfactual. What is reasonably clear, however, is that in order to extend
private healthcare to the poor, new approaches and products are going to have
to be devised. The parties to this merger – and particularly Prime Cure, the
target firm are at the forefront of this thinking and this is why they are
counted amongst the small number of firms that have made inroads, slight
though they may be, into the untapped market for the provision of private
healthcare to the poor.
66. The analytical complexities of this merger are massively compounded by the
important role played by regulation in the private healthcare market, not to
mention by the state’s role as a critically important direct provider of
healthcare services. As already noted, the very notion of extending private
healthcare services. As already noted, the very notion of extending private
healthcare to the poor is catalysed by the state’s decision to adopt this model
in its efforts to meet its mandate to provide affordable healthcare to all of its
citizens.
67. The poor, of course, have always been with us although this, on its own, has
not inspired entry into this market. It is the advent of a state dedicated to
providing healthcare to the poor that has created this market. The merging
parties have claimed that the expressed desire of key private players – notably
the large medical scheme administrators – to enter this market is inspired by
the downward pressure on their prices and margins. The Commission has cast
some doubt on whether prices and margins in this part of the broader
healthcare market have been subject to downward pressure, but even if this is
so, then it is fair to say that this has, in significant part, resulted precisely from
state intervention particularly from the imposition of a prescribed set of
minimum benefits as well as a range of other regulatory interventions. Nor
has the state been content to focus its attention on the healthcare funders. The
producers and retailers of pharmaceutical products – another important driver
of healthcare costs – have also come in for considerable and controversial
attention from the state. These interventions have, in turn, impacted on other
providers of healthcare services who are more directly implicated in this
transaction, notably doctors and private hospitals, the latter a particularly
14
important component of healthcare costs, who have derived large margins
from the sale of medicines.
68. Nor is the regulatory framework settled. Far from it. The state’s intervention
in the area of pharmaceutical pricing and provision has been subject to
swirling public controversy and wideranging litigation that is, as yet, not yet
fully resolved. In the area of healthcare funding and management, the state
has, predictably, moved most decisively in the provision of healthcare
insurance to its own employees. As already elaborated, it has done this by
registering a new medical insurance fund – GEMS – and within this
framework it has, it appears, prioritised the movement of those of its
employees already covered by medical insurance into GEMS and, then, the
extension of coverage to those of its lowest paid employees who are, as yet,
uninsured. It has indicated that once this has been achieved – itself, as we
shall indicate, no mean task – it will seek to extend this coverage beyond the
public sector, thus constructing what is generally referred to as a system of
‘social health insurance’.
69. However, while the state’s overall objectives are reasonably clear, the precise
tools that will be deployed to achieve these objectives are by no means finally
resolved. While it appears, as will be elaborated at length, that GEMS has
opted for capitation as the preferred mode of providing costeffective options
for its lowincome employees, many commentators, including some who have
actually entered bids for the tender, still insist that alternative modes of
healthcare management may produce preferable outcomes. There is a strong
body of opinion, eloquently articulated by witnesses in these hearings, that
insists that the provision of affordable healthcare to the poor is an oxymoron
insists that the provision of affordable healthcare to the poor is an oxymoron
in the context of the present system of prescribed minimum benefits.
However the state has not, as far as we are aware, indicated that it is even
reconsidering the imposition of prescribed minimum benefits which are
generally identified as a cornerstone of a regulatory system that is directed at
producing adequate healthcare to all of its citizens. We should also add that
past experience, compounded by the huge stakes involved in the GEMS
tender, suggests that the award of the tenders will be the subject of intense
contestation including timeconsuming litigation.
70. How does all of this uncertainty and fluidity impact on our consideration of
the transaction that is presently before us? It means, in short, that the past is a
particularly unreliable guide to the future. As we shall demonstrate, extreme
uncertainty bedevils an analysis of the impact of this transaction even at the
most fundamental and elementary level of merger regulation. Hence, as we
shall elaborate, we are not able, with any confidence, to predict the response of
consumers to price movements in the products offered by the merging parties.
71. It is our view, then, that this extremely fluid context, the absence of an
established and stable regulatory framework for this embryonic market as well
as for some related and longstanding markets (for example, pharmaceuticals),
demands that we adopt a particularly cautious and circumspect approach to
15
private interventions, such as this merger, that will inevitably impact on the
development of the market under consideration. Public interest considerations
impinging on the outcome of interventions in this area – be they interventions
by the state, by regulators or by private market participants – are, for
unimpeachably good reason, unusually intense and this also predisposes us to
particular circumspection.
72. We are, to state the obvious, dealing with a transaction in a market that is
central to the interests of the state, to the private sector and to ordinary
consumers. It may well be that in a year’s time, or, more likely, in five years’
time, the regulatory framework and the parameters of the markets implicated
in this transaction will be more certain and that the consideration of an
identical or similar transaction will produce a different outcome. However, it
is in the nature of merger analysis that changing eras and contexts produce
different outcomes. There is no single answer that stands for all time.
The Relevant Markets
73. The papers filed in this merger display an unusual degree of consensus
between the merging parties and the Commission over the definition of the
relevant market. This is particularly unusual in the context of a merger that
the Commission recommends be prohibited. One would have expected a
deepseated divergence on the boundaries of the relevant market. However,
despite their initial closeness to the Commission’s view, the parties have
ultimately argued for a relevant market significantly broader than that
contended for in their formal merger filings.
74. It is common cause that the merging parties’ activities overlap in respect of:
i. primary healthcare (through the operation and administration of primary
healthcare centres) and
healthcare centres) and
ii. the administration of capitated managed care options. 30
75. There is thus agreement on the fact that there are two relevant markets. The
first can be dealt with relatively easily. This is the market for the provision of
primary healthcare services. This consists of the operation of medical centres
through which doctors, dentists and other healthcare professionals provide
primary healthcare services.
76. The commission examined the geographical locations of the Medicross and
Prime Cure medical centres and concluded that the merging parties had
overlapping facilities in five centres, namely, Bloemfontein, Bluff (Durban),
East London, Kimberley and Port Elizabeth.
77. However, it appears that the Medicross and Prime Cure facilities are directed
30 See paragraphs 16.1 in the forms CC(2) filed by each of the merging parties. This is a verbatim
rendition of their description of the areas of overlap.
16
at markedly different income groups. Prime Cure centres are located in close
proximity to lowincome communities, such as masshousing townships and
industrial areas, generally within walking distance for potential patients. By
contrast, Medicross centres are located so as to target middleincome market
earners, within easy driving distance of their residential suburbs.
78. It is clear that the clinics managed by the merging parties represent but a small
portion of available primary care in the areas in which they are located –
generally urban and metropolitan. Although this availability varies widely as
between different geographical locales there is no evidence to suggest that the
clinics of the merged entity will acquire market power for the provision of
primary healthcare in any single locale.
79. It is common cause between the parties and the Commission that this market
presents no competition problems, and we concur with this assessment. This
market will not be discussed further.
80. The second relevant market for which the parties, in their initial filings at
least, contended is that for the provision of capitated managed care options. 31
Although the Commission’s market definition is not particularly lucid, it too
holds most consistently that the market is that for capitated managed care
options.32 However, on several occasions it does refer to ‘the market for the
provision of managed care services with a national network of service
providers’.
81. This is the problematic market in this merger and it is precisely here that the
merging parties have sought, at the stage of the proceedings before the
Tribunal, to widen the boundaries of the relevant market for which they
originally contended. Their stance at the hearings and in their closing
argument was that the relevant market was the market for primary managed
argument was that the relevant market was the market for primary managed
care services for lowcost medical scheme options. On this view the provision
of any form of managed care at the primary level, and not necessarily
capitation, makes the provider a competitor in the market. Throughout the
hearings the Commission consistently took the position that the market is that
for capitated managed care, provided on a national basis.
82. It is clear that the Commission derived its list of competitors in the relevant
market from information provided by the parties in their original filings, where
capitation was specified as a feature of this market.
83. As we shall elaborate below, in their initial filings both merging parties
informed the Commission that the participants in the capitated managed care
market are Care Cross, Prime Cure, Medicross, Faranani, Metropolitan
(through, it later transpired, an entity called Qualsa), and ‘other IPAs’, that is,
31 The precise meaning of the term ‘capitation’ is dealt with below. Suffice for the present to note that
we follow the definition provided by Mr. Dorfling, the witness representing Medicross, in Exhibit 10
page 3
32 Transcript page 3, Hodge evidence Exhibit 3 page 3 and transcript page 365
17
Independent Practitioners’ Associations. As we shall see, the data supplied by
the parties indicate that Faranani, Metropolitan and the IPAs are fringe
players, with Care Cross, Prime Cure and Medicross accounting for the lion’s
share of the market. The Commission’s investigators certainly accepted that
the only players in the capitated managed care market were those identified by
the parties. Moreover, we note – and elaborate below that the views of the
parties as contained in their initial findings were confirmed in the
Commission’s interaction with other parties in the broader health care market.
84. It is clear that whenever, in their initial filings, the parties specifically applied
themselves to identifying the relevant market they explicitly and unanimously
opted for the provision of capitated managed care as that market. Moreover,
they explicitly and repeatedly identified Carecross, Primecure and Medicross
as the only significant participants in that market with very much smaller
players, including Faranani, Metropolitan and selected regional Independent
Practitioners Associations (IPAs), competing on the fringes of that market. In
Item 16.4 of its ‘Statement of Merger Information’ (Commission Form
CC4(2)) Medicross states quite unequivocally that
‘the merging parties, are in the broadest sense (our emphasis), competitors for
the administration of capitated managed care options (emphasis in the
original).
85. In the same paragraph Medicross identifies itself, together with Prime Cure,
Carecross, Faranani, Metropolitan and ‘other IPAs’ as the competitors in the
market. Medicross further estimates that it, Prime Cure and Carecross account
for 87,7% of the ‘market share (lives covered by managed care’), with
Faranani, Metropolitan and the ‘other IPAs’ accounting for the remainder of
Faranani, Metropolitan and the ‘other IPAs’ accounting for the remainder of
‘lives covered by managed care’, that is, for 12.3% of those lives covered by
capitated managed care. These views are then precisely echoed in Paragraph
16.4 of Prime Cure’s equivalent submission.
86. Even though on its face this appeared, on this market definition, to be a
potentially problematic transaction from a competition perspective, the
merging parties clearly took comfort in their view that they were not
competitors because of the different market niches that they targeted, with
Medicross targeting middle and highincome consumers, and Prime Cure the
lowincome consumers. On this basis they held that the merger presented no
horizontal problems. Again in Paragraph 16.4 of their respective form CC4 (2)
submissions, both parties, having, as indicated above, identified themselves as,
‘in the broadest sense, competitors for the administration of capitated managed
care options’, continue:
“However, at the narrower level as discussed in the Analysis, they are not
viewed as competitors due to the fact that the parties target different income
groups. (our emphasis ).”33
33 We note in passing that this assertion ignores – indeed suppresses – the Januslike links between
Medicross and Netdirect.
18
87. This conclusion was cogently countered by the argument advanced in the
Commission’s recommendation. This acknowledged the distinct market niches
which had existed in the past, but nevertheless, concluded that the transaction
would give rise to a substantial lessening of competition, relying on the
doctrine of ‘potential competition’. Effectively, the Commission argued that
Medicross, through or in combination with Netdirect, was poised to enter the
lowincome segment of the market and that its competitor in the lowincome
segment, notably Care Cross, through its associated company, One Care, had
already entered the upperincome segment.
88. Clearly the parties recognised in the course of preparing for these hearings that
their defence of the transaction rested on thin ice and so have sought to amend
their case by expanding the contours of the relevant market.
89. Counsel for the merging parties argued that his clients should not be held to
their original definition of the relevant market. He insisted that because these
are not adversarial proceedings but rather proceedings in a truthseeking
enquiry, the parties’ filings cannot be given the status of pleadings but are
rather their initial contentions in an unfolding enquiry in which their ideas and
opinions will evolve as evidence and argument are submitted to the Tribunal.
He has sought to characterise the Commission’s defence of its view of the
relevant market as unduly dogged and inflexible. 34
90. While, in general, there may be some broad validity in these contentions, they
are ultimately not persuasive. It is one thing to argue that the Commission
should be prepared to confront, with a relatively flexible mind, evidence and
argument submitted to the Tribunal to the effect that the Commission had
opted for too narrow a market definition. However, it is quite another matter
opted for too narrow a market definition. However, it is quite another matter
to insist that the parties should be at liberty to broaden or abandon the
boundaries of the relevant market for which they initially contended. The
view of the relevant market that is contained in their initial filings reflects the
merging parties’ understanding of the world in which they conduct their
business and this view from the coalface appropriately guides the
Commission’s investigation. The parties’ effective definition of the relevant
market is in fact derived from their identification of their competitors. 35 We
can understand why business people may misinterpret a request to identify a
‘relevant market’ – this is a term of art in competition law and economics that
may well not be easily understood by one not versed in antitrust theory.
However they are not asked to do this. They are rather asked to list their
competitors, and this is the first, and most important, building block in the
Commission’s definition of the relevant market. There can surely be few
business people worthy of the name who would not understand a request to
34 Transcript page 1218
35 Item 16 of the Statement of Merger Information requires that ‘for each identified product and
service, (the parties) identify, and provide contact details for, the five producers or providers in each
identified geographic area with the largest estimated share of total turnover during the last full 12
months.’
19
identify their own competitors.
91. It is wholly conceivable – even likely that a merging party which is familiar,
or is familiarised, with the nature of a competition enquiry may take an unduly
expansive view of its competitors and the reason why the Commission then
interrogates these submissions further is to establish whether the views they
contain are sustainable or not. However the Commission cannot reasonably be
expected to believe that the merging parties have inadvertently omitted to
mention a host of significant competitors. In the hearings, the witnesses for
the merging parties identified companies or divisions of companies such as
Solutio, Qualsa and Yarona as competitors in the market, and yet we are asked
to believe that they somehow neglected to refer to them as competitors in their
initial filings. For example, the legal representatives of the merging parties
castigated the Commission at the hearings for failing to enquire of
Medscheme, a large medical schemes administrator, whether its managed care
division, Solutio, considered itself to be a competitor of the merging parties.
Contrary to the merging parties’ initial submissions, in which Solutio warrants
not a single mention, they now contend that it is a particularly significant
presence in the market. The merging parties having not seen fit to identify
Solutio as a competitor, what reason would the Commission have had to
pursue this line of enquiry?
92. The Commission in fact approached Medscheme and many of the other
medical scheme administrators whom the merging parties now insist are their
competitors, because they were identified by the merging parties as their most
important customers.36 The Commission’s interrogation of these identified
customers is entirely appropriate to that relationship. It approached each of
customers is entirely appropriate to that relationship. It approached each of
the significant competitors identified by the merging parties – these being
Faranani and Carecross – who confirmed the parties’ view that they were
indeed competitors and who were then interrogated on that basis. It is
noteworthy that in the course of the Commission’s investigations Faranani and
Carecross both confirmed the parties’ initial submissions regarding the identity
of participants in the relevant market. 37 Dr. Nauta was crossexamined at
length on the identity of his competitors. While he acknowledged that entities
like Qualsa (Metropolitan), Solutio (Medscheme), Sizwe and the IPAs
accounted for a smattering of lives on capitated options, he did not view any
of them as a significant competitive presence.
93. As indicated above, the parties’ initial view of the participants in the market
was generally borne out in the Commission’s interviews with other
participants in the broader health care market. Hence Transmed, one of the
smaller medical schemes, identified Prime Cure and Medicross as providers of
capitated options. Although it had an agreement with Qualsa (Metropolitan),
36 See Paragraph 18 of the forms CC4(2) filed by the merging parties
37 For Faranani’s view see Record File D, p133. For Care Cross see Transcript pages 20,64 and 69.
Note that Dr. Nauta, the witness from Carecross, also identified Netpartner as a potential competitor
(transcript page 120)
20
it did not consider Qualsa to be a competitor of the merging parties. 38
Spectramed, another small scheme, also listed Prime Cure, Medicross,
Carecross and Faranani as well as a firm called Healthcare Alliances. 39 A
third small scheme, Ingwe, also identified Prime Cure, Medicross, Faranani
and Carecross as national providers of national primary care solutions. Ingwe
was of the view that the IPAs’ regional character placed them outside the
market.40
94. In this regard we view the evidence of Mr. Strauss of Discovery as particularly
revealing. Here was a witness from South Africa’s largest medical scheme
administrator with responsibility for exploring and concluding agreements
with service providers. His is surely a particularly privileged vantagepoint
from which to identify participants in the market for the provision of capitated
managed care options. Schemes administered by his company are, of course,
significant users of these services.
95. Strauss’ view is that Medicross, Prime Cure and Carecross compete in this
lowincome market, with Carecross being “the major player”. 41 He
acknowledged that Discovery had been approached by many entities claiming
to be primary care providers, but added that none of them had proved to be
costeffective or “of substance enough for Discovery to contract with
them.”42 Regarding Solutio, Strauss’ view was that it was effectively a data
management entity. 43 Discovery’s view of Qualsa was that it managed in
hospital expenses rather than primary care. Furthermore, having a national
footprint is imperative to Discovery’s selection of a primary care provider and
Qualsa did not, in Discovery’s view, have such a national footprint nor did it
attract enough lives to make it robust. Strauss remarked that Discovery would
not use it. 44
attract enough lives to make it robust. Strauss remarked that Discovery would
not use it. 44
96. As for Yarona, Strauss testified that this entity had initially spun off as a
division of a medical scheme and had previously made an offer to Discovery,
but Discovery did not consider it to be a player. Specifically on Yarona
38 Record File D, pages 912
39 Record File D, pages 978. Healthcare Alliances is included in the parties’ expanded list of
competitors in the market although none of the witnesses was able to provide any significant
information regarding this firm’s activities.
40 Record File D, pages 1045
41 Transcript pages 140, 182
42 Transcript page 141
43 Note the following exchange between the Commission’s counsel and Discovery’s witness, Mr.
Strauss:
Adv Berger : Have you heard of an organisation called Solutio?
Mr. Strauss: Solutio, part of Medscheme?
Adv Berger : Correct.
Mr. Strauss: Yes, I’ve heard of them.
Adv Berger : Would you consider them as a player in the capitated managed care market?
Mr . Strauss: I wasn’t even aware that they would offer those services. My understanding of
Solutio is that they were effectively a data management group.
44 Transcript pages 141 – 144. See also transcript at pages 174, 289.
21
Strauss stated:
“Yarona’s proposal relied on a network of contracted providers, but they had
to bring in a third party by the name of Calabash to manage the network and to
take risk within their proposal. So when we are talking specifically about
Yarona, Yarona themselves, as I understand it, have a list of contracts at a
particular rate per consultation, but they as an organisation do not take risks
and they as an organisation, as I understand it, do not manage the doctor’s
utilisation patterns”. 45
97. Because the merging parties have made so much of Solutio’s alleged presence
in this market, it is as well to spell out the precise extent of its involvement.
98. Dr Walters testified as to Solutio’s involvement with lowcost insurance
options. He referred firstly to the Bonitas medical scheme, an open scheme
administered by Medscheme which has a lowcost capitated option
(“Boncap”). This is managed by Prime Cure and Faranani. This option
consists of 400 lives, with 4 000 being added in January 2006, making a total
of 5 400 lives. 46 […….CONFIDENTIAL……]
99. Secondly, in respect of the Liberty medical scheme, also an open scheme
administered by Medscheme, there is presently a capitated lowcost option,
with Faranani as the primary service provider. Similarly, Liberty has a
mediumcost option, which is a ‘virtual’ capitated model, with Medicross as
the service provider. […….CONFIDENTIAL……]. Walters testified that the
number of lives with Faranani was of the order of a few hundred, while
Medicross administered some 600 lives.
100.Thirdly, Walters referred to Protecta, a closedscheme also administered by
Medscheme, with the lowcost capitated option currently managed by Prime
Cure. […….CONFIDENTIAL……] He could not specify how many lives
were entailed.
Cure. […….CONFIDENTIAL……] He could not specify how many lives
were entailed.
101.Walters also made mention of Sasolmed, a closedscheme which also has a
lowcost option for which Solutio is presently bidding. It is not a capitated
product but a managed feeforservice option, that is, for the most part,
confined to Secunda, Sasolburg and Pretoria. It has about 5000 lives in the
lowincome option. He also made mention of the AECI scheme where it
appears that manages the lowcost option. It appears that Solutio’s role is
limited to oversight of the Carecross contract on behalf of the scheme in order
to ensure that service delivery is adequate.
102. Ultimately, it seems that the only primary care capitated current business that
Solutio itself provides is in respect of the Daimler Chrysler scheme. This is a
closed regional scheme where, acknowledges Walters, the capitation fee is
exceptionally generous.
45 Transcript page 287
46 Transcript page 5578
22
103.We concur with the Commission that Solutio’s share of the capitated
managed care market, when measured by number of lives, is so small that it
cannot be considered a competitive constraint on the major participants in the
market. It is limited to schemes administered by Medscheme where, for the
most part, it continues to rely on one of the major providers of capitated
options. We should also note that most of the lowcost schemes in which it
plays a role are small, closed and regionallybased.
104. It is, of course, eminently possible that the exercise of market power on the
part of the participants in a relevant market may induce new entry into the
market. This will be an important part of our enquiry, particularly when the
question of entry barriers is examined. It is also possible that the conduct of
existing players in the market may be constrained by potential rivals who are
easily able to utilise existing assets and knowhow deployed in a related
market to enter the market in question socalled supplyside substitution. 47
This too will be examined when we consider entry barriers. However we are,
on the basis of their own contentions, satisfied to conclude that the parties
confidently identified the relevant market as that for the administration of
capitated managed care and, more revealing of their view as to the boundaries
of their market, they identified the participants in that market their
competitors in other words as Care Cross, Prime Cure, Medicross, Faranani,
Metropolitan and certain of the IPAs. It would be difficult to deny that the
data in the parties’ own filing reveal that Faranani, Metropolitan and the IPAs
are nothing more than fringe players. We will show that the evidence
demonstrates that they are destined to remain on the fringes of the market.
demonstrates that they are destined to remain on the fringes of the market.
105.We find that the relevant product market is that for the provision of
capitated primary managed healthcare products.
106.There are various forms of capitated products. There is the form of capitation
where the managed care organisation effectively assumes the risk from the
scheme. The most advanced form of capitation is where the managed care
organisation then transfers the risk to the service provider. Ultimately this is
the desired endpoint of capitation because it effectively incentivises the
service provider to tailor his treatment regime to the limits imposed by the
capitation fee. And there are variations on this theme. In his evidence Dr.
Walters of Medscheme/Solutio spoke of ‘gain sharing’ options which
combined capitation – where the managed care organisation or service
provider assumed all the ‘downside’ risk with an arrangement that enabled
the scheme to share some of the ‘upside’ with the managed care organisation
or the provider.
107. Although the merging parties argued that there are managed care
mechanisms that are substitutable for capitation, we will show that it is widely
47 Areeda defines supplyside substitution as “when manufacturers can shift their output between
products A and B simply by shifting the settings on their machines”. See Areeda Hovenkamp Solow
“Antitrust Law” Vol IIA at page 56 1
23
recognised that the product that is most effective for the provision of lowcost
insurance options is indeed capitation. Variants of this approach are steps
along what Dr. Walters described as a ‘journey’ toward the attainment of the
provision of a fully capitated product, one in which the service providers
assume the risk. Although the precise boundaries of the market for which the
Commission contends are sometimes blurred our definition is certainly close
to the Commission’s view and to the view for which the merging parties
initially contended. It is certainly narrower than the parties’ revised view of
the market which effectively argues that all managed care products are
substitutable, and are adequate to the task of securing health care options for
the poor. 48
108.Almost as many definitions of ‘managed care’ and ‘capitated managed care’
have been proposed in these proceedings as there were witnesses. Dr Walters’
definition seems to capture the essence of the concept of managed care:
“there are three things about managed care and I am sounding like a professor,
but it is quite simple. You set standards, you set financial standards, you set
clinical standards. That’s the first thing. The second one is you must have
systems, processes and systems to actually administer those standards and the
third thing is that you must analyse the outcomes. That’s managed care.” 49
109. As for capitated managed primary care the most succinct definition is
contained in the witness notes handed up by Mr. Dorfling at the hearing.
There ‘capitation’ is defined as,
“a method of payment for health services in which a provider is paid a fixed,
per capita, amount in advance for each enrolee without regards to the actual
number of nature of services provided to each member in advance. This
involves a great deal of risk sharing.” 50
involves a great deal of risk sharing.” 50
110.The Medicross website also offers a succinct insight into the content of
managed healthcare and the role of risksharing under capitation:
“As well as providing healthcare services at industry negotiated feefor
service tariff rates for the medical aid and private patient, Medicross offer a
range of unique, comprehensive managed healthcare plans on a capitated basis
(a fixed monthly fee) to look after the patient's individual healthcare
requirements. Managed Healthcare is a means of providing healthcare services
48 In truth the parties vacillate and are seemingly unable to draw the boundaries of the relevant market
for which they contend with any precision. In his closing argument Mr. Unterhalter, for the parties,
defines the market by reference to ‘those who understand themselves to be in the business of offering
managed care options in the marketplace.’ But in the very next sentence he significantly narrows this
market when he states: ‘in other words, they are managed care organisations with network capabilities
that manage risk and make offerings whether to open or closed schemes in respect of primary care
offerings.’ Transcript page 1209.
49 Transcript p523
50 Exhibit 10 page 3
24
within a defined network of service providers , who in turn assume the
responsibility and therefore the risk of providing quality, costeffective care,
while ensuring that only appropriate services are delivered. Under this model,
emphasis is placed on keeping the patient well, rather than treatingepisodes of
illness. In this environment the primary healthcare practitioner is responsible
for managing downstream utilisation of services and effectively becomes the
custodian of the patient's healthcare funds.” (Our emphasis added)
111.Mr. Strauss’ outline of the workings of Discovery’s capitated primary care
option, the Key Care plan, also pinpoints the essential features of capitation:
‘ Adv Berger : What was then the arrangement between the scheme and
Carecross?
Mr . Strauss: The arrangement was that Carecross would provide all the primary
care benefits though a network that they would put together. There would be a
capitated payment, a fixed payment per member per month from the scheme to
Carecross in return for which they would provide a list of services according to
specific medical codes and according to specific medial ferneries (should read
‘formularies’) and pathology tests and radiology tests.’
112. In concept capitated managed care is a species of managed care that is
characterised by a ‘ fixed payment per member per month from the scheme’ –
with the inevitable consequence that there is a transfer of risk from the
medical scheme to the managed care organisation . In its implementation,
capitated primary care and other forms of risktransferring managed care
require that the managed care entity which contracts with the medical scheme
undertakes the supervision or management of a network of primary healthcare
providers who generally assume a part of the risk. If they in turn receive their
providers who generally assume a part of the risk. If they in turn receive their
remuneration by way of a capitation payment, there is risk transfer to their
level or tier in the healthcare matrix. In other cases, the managed care entity
may retain the whole or a part of the risk at its level by paying the service
providers on a feeforservice basis. The Commission’s expert, Mr Hodge,
elaborated further on the concept of capitation and the concept of the passing
of risk:
“So in terms of the third model, which is what we are talking about, the
providers today with primary capitated managed care , that company takes on
risk. So it offers a scheme for a fixed fee per member per month to manage
and typically this involves unlimited daytoday benefits. And they then
established a doctor network or in some cases clinics and they will pay the
doctors either on a capitation basis themselves and pass down the risk or on a
discounted feeforservice.
As I’ve documented there, they perform certain functions. So they overlap in terms of
the administrator in terms of doing claims processing. But the key aspects, which Dr
Nauta brought out is the ability to monitor, analyse and specifically manage
utilisation. If you’re on risk, you need to manage the utilisation and especially if
you’re paying your doctors on a feeforservice basis to ensure that essentially the
25
income you get in is not exceeded by the benefits you pay out. You’ve clearly got to
implement trained doctors. We’ve heard more, let’s say intangible aspects such as
getting their buyin to manage the concept, and just manage the doctor relationship
in general to ensure that the doctors provide costeffective and quality health care.”
51(our emphasis added.)
113.Although some witnesses argued averred that there are primary managed care
products for lowcost insurance options that do not rely on full capitation, in
truth much of the evidence before us regards capitation as an essential element
of a managed care product directed at providing health insurance for low
income consumers. This is not to deny Dr. Walter’s contention that full
capitation lies at the end of a long journey that may begin with varying
mechanisms for managing a feeforservice arrangement and that ultimately
ends with full risk transfer. But it is to insist that a high degree of risk
transfer, (in the form of capitation) is required if healthcare provision is to be
extended on a significant scale to lowincome earners. Dr. Nauta contends
that the transfer of risk”
“…is really critical to the success in this market. You’ve got to ultimately
transfer risk from you as entity in the middle, that buys this from various
options and various schemes, to the doctor.” 52
114.And for all Mr. Dorfling’s insistence on a range of alternative managed care
products for the lowincome market, it does not seem that the merging parties
disagree with Nauta’s assessment of the nonsubstitutability of anything
except risk transfer – in other words, capitation in respect of the lowincome
market. Dr. Stillman, reports that:
“Medicross shares the common view that to offer a medical scheme option
that covers prescribed minimum benefits at these price points, i.e. low price
that covers prescribed minimum benefits at these price points, i.e. low price
points, the managed care model, as opposed to the feeforservice model, is
the only practical alternative. Medicross believes moreover that to provide a
private healthcare product at these price points, it will be necessary for
managed care to be on a full risk basis, i.e. to offer a managed care plan that
covers specialists, medicines, hospitals as well as primary care.” 53
115.Stillman sums up the merging parties’ view:
“In sum the parties believe that the only effective way to deliver lowcost
medical scheme options at prices affordable to lowincome consumers, is
through a fullrisk managed care product offering”. 54
116.Thus, by the merging parties own reckoning, full risk transfer is necessary if
51 Transcript page 360
52 Transcript p16
53 Cited in Transcript p1128 and Stillman’s Mdated 20 June 2005 on page 3
54 Cited in Transcript at p1117 and Stillman’s Mdated 20 June 2005 on page 3
26
health care insurance is to be made available to lowincome consumers. And
it is this model that requires highly organised primary care provider networks
and a sufficient number of lives to incentivise the doctors and other medical
service providers to accept full risk transfer.
117. Mr. Dorfling contended that the extension of private healthcare to low
income consumers could be achieved by a range of managed care products.
He lists what he believes are a number alternative managed care products
adequate to the task of ensuring lowincome health insurance option, although ,
as elucidated earlier , even he concedes that fullrisk transfer is the ‘most
effective’ mechanism. 55
118. Indeed it is clear that the demandside of the market also recognises the non
substitutability of fullrisk capitation. The GEMS tender for its Topaz and
Sapphire options, from where the bulk of the predicted surge in demand for
lowcost insurance cover is expected to emanate, clearly specifies that GEMS
is calling for tenders for fullrisk capitation. 56
119.We should, for the sake of completeness, comment on a distinction much
relied upon in the parties’ initial filings and dealt with by the Commission in
its report, but which now seems to have been abandoned by both. This is the
distinction between the provision of capitated managed care products for low
income earners (where Prime Cure is focused), and capitated products for the
middle and higher income markets. This latter is the segment where
Medicross is focused and is often referred to as the ‘buydown’ market. It is
not clear to us that either the parties or the Commission ultimately attached
much significance to this distinction. It is clear that capitation is a mechanism
for offering lowcost medical insurance and it is the lowincome segment of
the population at whom it will be aimed. There is a prospect of part of the
the population at whom it will be aimed. There is a prospect of part of the
higherincome part of the market ‘buyingdown’ and this is already occurring
to some limited extent. But it will be limited because these options embody
limitations imposed on the scheme member that higher income options do not,
and, for that reason, these options are unlikely to attract significant support
from higher income purchasers of medical insurance. Moreover these ‘buy
down’ options will not be actively marketed or facilitated. At all levels, the
healthcare providers clearly attempt to maintain a separation between the high
income and lowincome purchasers of medical insurance, precisely because of
the prospect of highincome purchasers availing themselves of options that
cost less than they are able to afford, thereby eroding revenues and profits. In
the face of everrising healthcare costs, we conclude that while buydown is a
55 Transcript pages 988996, 1118. Although the precise workings of these options were not fully
elaborated in these hearings, our initial impression is that they all demand a wellorganised network of
primary care providers and, so, even to provide these ‘noncapitation’ managed care products, a critical
barrier to entry into the capitated primary care market would still have to be overcome.
56 In the light of this, it is puzzling that Dorfling should continue to insist that his company – and, in
this instance, at least insofar as the primary care component is concerned, we take this to mean
Netdirect – will submit a bid for the provision of a managed care product which includes a possibility
other than capitation. At another point Dorfling states that Netdirect will be tendering for a full risk
product, see transcript page 1087
27
phenomenon that is unstoppable, on present evidence it is unlikely to proceed
so far that it blurs or eliminates the boundaries set by capitation.
Geographic Market
120. The geographic market is in our view national. There is a limited market for
regional primary managed care products and this market may be penetrated by
regional providers utilising regional networks. However, it is our view that
these will service a shrinking portion of the overall health insurance market,
including the lowincome market. We will, in our discussion of entry barriers,
outline why we view the regional IPAs to be inadequate substitutes for well
organised national networks – indeed, although the evidence is not unanimous,
credible evidence pointed to important weaknesses of the regional IPAs,
weaknesses that militate against them providing the extent and character of
network management that primary managed care for lowcost health insurance
demands. Many of the medical aid schemes, large and small, submitted that a
national footprint was one of the key criteria they would look for in selecting a
managed care provider. Mr Strauss, of Discovery, was particularly emphatic in
this regard:
“ MR STRAUSS : From our perspective we’re always … one of the first
criteria will be for a national footprint, Faranani, as they’ve produced their
membership lists, have always dominated Gauteng rather than anyone else.
We’ve been into their organisation and looked at their systems and their
processes, many of which they outsource, and they just haven’t seemed to
attract the lives to make them into a robust organisation.” 57
121.In short, there will be niche opportunities for regional providers of capitated
primary managed care. But these are unlikely to constrain the competitive
behaviour of the national providers. Accordingly, we conclude that the
geographical market is national.
geographical market is national.
122.Activities in this market engage closely with related markets. In particular,
interaction with the market for the provision of private hospital services and
the market for the provision of medical scheme administration services will be
selectively considered.
The Impact of the Merger on Competition
123.The relevant product market – the market for the provision of capitated
primary managed healthcare products – is highly concentrated. The merging
parties estimate that there are 342 000 lives covered by capitated managed
care options. Of these, 10,2% are covered by Medicross, 33,6% by Prime
Cure and 43,9% by Carecross. That is, 87,7% of the number of lives covered
by capitated managed care options are accounted for by the three largest
players. Faranani is estimated to enjoy a market share of 5%, with
57 Transcript page 141. See also record File D page 91, 105
28
Metropolitan’s share standing at 1,5% while ‘other IPAs’ collectively account
for 5,8%. 58
124.This is, by any reckoning, a highly concentrated market. However a merger
cannot be judged on this fact alone. We proceed then to examine the impact
of the merger on competition.
125.The Commission has argued that this transaction has both a horizontal and
vertical dimension. The horizontal dimension arises from the merger of two
firms involved in the same product and geographical market. The vertical
dimension refers principally to the place of the acquiring firm in the Netcare
hospital group. We will analyse each of these dimensions in turn.
126.Section 12A(2) provides a nonexhaustive list of factors that are to be
considered in the assessment of the impact of a merger on competition. Those
factors that are pertinent in the consideration of the impact of the horizontal
dimensions of the merger are ‘the ease of entry into the market, including
tariff and regulatory barriers’ (12A(2)(b)), ‘the level and trends of
concentration, and history of collusion, in the market’ (2(c)), ‘the degree of
countervailing power in the market’ (2(d)), ‘the dynamic characteristics of the
market, including growth, innovation and product differentiation’ (2(e)), ‘the
nature and extent of vertical integration in the market’ (2(f)) and ‘whether the
merger will result in the removal of an effective competitor’ (2(h)).
The Horizontal Dimensions of the Merger
Price sensitivity
58 These figures are to be found in Item 16.4 of the merging parties’ CC4(2) filing. Note in the
narrative these are referred to as ‘the number of lives covered by capitated managed care options’ while
in the tabular representation (which is also part of Item 16.4) they are simply referred to as ‘lives
covered by managed care’. Clearly a great many of the approximately 7 000 000 lives covered by
private medical schemes are subject to some or other managed care mechanism – for example pre
hospital authorisation is ubiquitous. We take it then that the reference in the table to ‘lives covered by
managed care’, expressed more accurately, refers to lives covered by primary care capitation.
However it is instructive that the merging parties themselves collapse the distinction between
‘managed care’ and ‘capitation’ despite their insistence that capitation is merely one among many
managed care options. Clearly where primary care insurance for low income earners is concerned,
‘capitation’ is, even in the parties’ estimation, all but synonymous with ‘managed care’.
29
wwwww.Before turning to a detailed consideration of the factors listed in
Section 12(A)2, we examine a proposition that has, to a greater or
lesser extent, received the endorsement of several of the witnesses who
participated in these hearings. This concerns the question of price
responsiveness. In essence the merging parties assert that this is an
unusually pricesensitive product, and consequently that there is
limited capacity for the exercise of market power.
xxxxx.Firstly, it is clear that the product which is assumed to be
inordinately price sensitive is not the managed care product (in this
case, capitated managed care) at all, but rather refers to the insurance
product itself. It is then implicit in the assumption that because an
increase in the price of the capitated managed care product will be
passed through to the consumers of the insurance products their
conjectured sensitivity to price increases will restrain an exercise of
market power on the part of the managed care providers.
yyyyy. We should be clear that there is insufficient evidence to sustain
the assumption – and that is all that it is – that lowincome
consumers will be particularly sensitive to movements in the price
of health insurance. Dr. Stillman, the merging parties’ expert
witness, concedes that because these lowincome insurance
products are ‘new development(s)’ there is insufficient empirical
evidence to undertake the standard statistical and econometric
tests that would be normally employed to resolve this
disagreement.59
130.It appears to us that many of the ready assumptions that are made regarding
the price sensitivity of the insurance product confuse the entrylevel price –
what several witnesses refer to as the ‘price point’ with responsiveness to
what several witnesses refer to as the ‘price point’ with responsiveness to
changes in price. 60 It is common cause that a large class of consumers, so
called ‘lowincome consumers’, has effectively been locked out of private
health insurance because even the lowest price options remain out of their
reach. Lowincome consumers do not partake of health insurance for the same
reason that they do not partake of firstclass air travel: they cannot afford it.
They are not at all sensitive to movements in the price of these products
because they simply do not feature in the consumption baskets of lowincome
consumers.
131.Prodded by a combination of government interventions and their own
commercial interest in tapping a potentially large new market, a range of
players in the healthcare industry are only now actively exploring mechanisms
59 Transcript pages 6889. Mr van den Heever, a witness for the Council for Medical Schemes, also
notes that the fledgling character of the market renders ‘quantitative assessment of elasticities…very
difficult.’ Transcript page 627
60 Although somewhat ambiguous, it appears that Dr. Nauta’s apparent support to the acute price
sensitivity in the health insurance market referred to entry price rather than to price changes.
30
for lowering the cost of private healthcare insurance to the point where it
enters the consumption baskets of these lowincome consumers. None – in the
private sector at any rate are more actively involved in this quest than the
parties to this merger and their fellow participants in the relevant market.
However once affordable products have been developed – and managed care
products directed at the provision of lowincome healthcare insurance will be
key to achieving this – there is no a priori reason for assuming that those who
purchase these options will be particularly sensitive to price increases.
132.Indeed because this is an insurance product – albeit a shortterm insurance
product – one could reasonably conjecture that a consumer who has already
sunk a material part of her income into purchasing this product would be
reluctant to forgo the possibility of recouping this in the shape of future
payouts when these are required. We acknowledge that regulation has
attempted to lower the costs of switching from one health insurance plan to
another.61 But these efforts notwithstanding, there are cogent reasons why
switching costs will remain particularly high. Not the least of these reasons is
that this is a notoriously complex product making comparison between
alternative options difficult. Moreover, many consumers will purchase their
health insurance plan from agents who may not always have an interest in
enhancing the consumer’s ability to make the necessary comparisons between
the products on offer.
133.In any event there remains an unresolved disagreement between the parties
and the Commission regarding the intensity of competition between medical
schemes themselves. The Commission has taken the view that competition
between medical schemes is muted and, therefore, that competition in this
between medical schemes is muted and, therefore, that competition in this
market will act as a poor indirect restraint on the primary managed care
market. The merging parties assert a contrary view. There is not sufficient
evidence for us to decide the intensity of competition in the medical schemes
market here. However we do note Mr. Hodge’s observation that primary
managed care is but one component of the contributions to the cost of a
scheme option. Therefore a substantial increase in the primary care
component may then not reflect as a significant increase in the end price of the
overall option and this may enhance the ability of the provider of capitated
managed primary care products to exercise market power. 62
134.We note too that Prime Cure has recently increased its premiums by some
25% and that Medicross’ predictions for the merger reflect significantly
increased prices by Prime Cure and a simultaneous growth in membership. 63
Again this is not definitive. But it is not consistent with the notion of a highly
pricesensitive market unable to absorb even modest price increases.
135.We turn now to an examination of those factors listed in Section 12(A)(2) of
the Act that are pertinent to our consideration of the impact of this transaction
61 Transcript page 599
62 This is elaborated by Mr. Hodge for the Commission at pages 379ff.
63 Transcript page 10912
31
on competition.
Barriers to Entry
136.The Commission has concluded from its investigations that entry barriers into
the market for capitated managed care options are significant. The record
evidences considerable support for this conclusion. The most important entry
barriers that are identified include the need for significant financial backing,
administrative capacity, the existence of significant economies of scale
represented by the number of insured lives, and then finally, and in our view,
decisively, to a range of elusive factors that we collectively refer to as ‘social
capital’. These latter include the capacity to build relationships with those
institutions – notably the trade unions that hold considerable sway over the
decisions of those most likely to opt for lowcost health insurance options, as
well as relationships with doctors and other primary care providers, reflected
in the ability to assemble and maintain wellorganised networks of primary
care providers.
137. The parties themselves have identified high entry barriers. Prime Cure, the
target firm, is quite explicit in this regard. The ‘Limited Confidential
Information Memorandum’ or ‘LCIM’, a report prepared by a firm of
consultants, Sevillano, Houseman, which was commissioned by Prime Cure
and whose conclusions appear to be based entirely on interviews with Prime
Cure management and shareholders, makes several direct references to the
high entry barriers. It identifies the necessity to build relationships with the
trade unions as a particularly significant barrier. 64 It also argues that new
entrants will have to rapidly secure a significant number of insured lives. The
consultants clearly believe that Prime Cure has overcome these barriers and
thus represents an attractive acquisition opportunity. Mr. Patterson, a witness
thus represents an attractive acquisition opportunity. Mr. Patterson, a witness
representing the Prime Cure shareholders, attempted to represent this as a
‘selling document’ and thus predictably hyperbolic, but this does not strike us
as credible. It was, after all, a document presented to a potential purchaser
extremely well versed in the healthcare sector generally and one that had
experience of attempting to develop and market capitated options. It seems
unlikely that Medicross, or its controlling shareholder, Netcare, would have
been persuaded by mere puffery. Indeed, Medicross, presents these factors –
Prime Cure’s established relationships and its insured lives – as the principal
reasons for undertaking the transaction. We are assured that
Medicross/Netdirect will enter the market if the transaction does not take place
but that it will, by its own admission, take it from 1836 months to do so and
this from the firm that is, as we shall elaborate, probably best placed for rapid
entry.65
138. Dr. Nauta, the managing director of Carecross, the largest provider of
capitated managed care services, testified to the high entry barriers
64 See File C page 328, 334
65 This is according to Stillman, the parties’ expert. See Exhibit 6 page 3, and transcript page 836.
Dorfling later states that it will take them 34 years, see transcript page 1111
32
surrounding this market. He believes that there are two particularly significant
barriers to entry, these being the construction and management of a primary
care providers network, and the accumulation of the number of capitated lives
necessary to cohere the network.
139. Nauta began his testimony with an overview of earlier – although still
relatively recent – attempts to provide managed care for the lowincome
market. It is an account littered with the corpses of some very significant
national and international companies that tried and failed to enter this market.
The merging parties appeared unable to dispute Nauta’s account of
unsustainable entry as evidenced by the very high failure rate of entrants.
However they argued that this was either irrelevant in that it effectively
required us to stand in judgement of what are essentially commercial
strategies. Alternatively they insisted that it represented the workings of a
robust market characterised by easy entry and exit. Both of the arguments
advanced by the merging parties are, on the facts of this case, unpersuasive.
The inability of new entrants – some very powerful and otherwise successful
parties in the insurance and healthcare sectors – to sustain a presence in this
sector, despite the appearance of strong latent demand and the consequent
incentive to stay the course, strongly suggests that they did not succeed in
overcoming the entry barriers identified by Nauta and others and thus they
failed.66
140. Nauta’s conceptualisation of the sort of network that is required and the
difficulty of organising and maintaining such a network, differs significantly
from that proffered by many of the parties’ witnesses. 67 However, it is
conceptually persuasive and his approach has succeeded where most others
conceptually persuasive and his approach has succeeded where most others
have failed. He effectively outlines two network models. The first – which is
the Prime Cure and Medicross model – is centred on a network of primary
care clinics. In Nauta’s view, this is a very costly mode of entry. 68 The
alternative model is that pursued by Nauta’s company, Carecross. This model
does not rely on bricks and mortar investments in a network of clinics but
rather on the tight organisation of independent general practitioners and other
primary care providers.
66 In Nauta’s own words, ‘….many people enter this. They go as fast as they come. It’s not difficult
to enter this market. You basically need to have a spreadsheet and a few bob in your pocket because to
say I’m going to do this and then the examples of that and the skeletons, you know, in the past are
there.’ (transcript pages 178). We should add that several of these skeletons belong to companies with
which Nauta himself was intimately involved.
67 Although, as we shall demonstrate, Nauta’s views on the difficulties of maintaining an effective
network have much in common with those of Dr. L. Walters of Medscheme, arguably the parties’ most
important witness on these matters.
68 In this, at least, Dr. Nauta and Mr. Dorfling, the witness from Medicross and Netdirect, seem to
concur. Dorfling outlined how in the merged entity the Medicross and Prime Cure clinics would
comprise the ‘hub’ of its network with the thousands of members of the Netpartner and Prime Cure
networks who are not connected to the clinics comprising the ‘spokes’ of a network significantly larger
than that which the clinics alone offered. He did not elaborate this ‘hub and spokes’ concept further
save to reject emphatically the prospect of building any more clinics. He too seems to have concluded
that these represent an unsustainable cost. See transcript page 1171
33
141. Several witnesses insisted that the formation of a network of doctors was a
simple task. None of the networks – including the Carecross network – is able
to insist on the exclusive loyalty of its members. 69 It costs individual
practitioners nothing to belong to a network and so many doctors retain
membership of several networks. It seems that a mere circular letter is
sufficient to recruit nominal network members and we were presented with
many examples of networks organised precisely in this manner. Indeed it
would be fair to say that this is the mode of organisation that characterises
most networks. However this is decidedly not the mode of organisation
favoured by Nauta, nor does he believe that this mode of organisation will
bear the weight of providing capitated managed care products.
142.It appears that the Carecross primary cure providers’ network is not an open
ended affair assembled by means of a mere circular letter. Indeed Nauta
testifies to the long process of building trust amongst doctors who were going
to have to accept Carecross’ invasive management of decisions and practices
hitherto under the exclusive control of the doctors and, as they moved onto
capitation, a reduction in the fee that they earned from each individual patient
in exchange for the relative certainty that capitation offers. These factors
require a relatively concentrated network that permits of constant contact
between the network organisers and the service providers who are members of
the network. An openended and diffuse network does not lend itself to
micromanagement nor does it enable the service provider to build an insured
lowincome patient base sufficient to make the capitated fee an attractive
alternative to the normal individual fee. It is for this reason, testifies Nauta,
that after a long period of intensive oneonone recruitment of service
providers, Carecross now has more of them applying for membership of its
network than it is willing to accept:
“…it has to be a closed network to be sustainable and so as I’ve pointed out
earlier, although we now have a lot of takers for Carecross, we don’t allow
doctors in easily, because it would just dilute our potential…the margins and
our ability to manage doctors.” 70
143.Note that Nauta specifically distinguishes the requirements of a network
providing managed care to lowerincome groups – this is the Carecross
network – from that required for providing these services on a capitated basis
to a higher income group. Hence, the One Care network, which is also part of
the Carecross stable but is directed at a higher income market similar to that
targeted by Medicross, is an open network where the patient essentially elects
her primary care provider who, on acceptance of certain conditions, signs up
for membership of the One Care network.
144. We find Nauta’s analysis of the requirements of an effective primary care
network persuasive. Certainly we are persuaded that the degree of network
69 Indeed it appears that the Council for Medical Schemes, the industry regulator, does not permit
exclusivity.
70 Transcript page 36.
34
organisation required for the provision of capitated managed care at the lower
end of the market significantly exceeds the openended approach to networks
favoured by those who insist on low entry barriers to entry in this market.
Nauta’s analysis also provides a particularly clear explanation of the
importance of a base of insured lives. It is not merely a case of spreading
these lives over the fixed costs of organising the network a traditional
economies of scale argument – but rather of spreading as many lives as
possible over as few doctors as possible in order to incentivise the members of
the network to move to a capitated model of managed care. This is why
successful managed care for lowincome consumers requires a network that is
both carefully selected or ‘closed’ and is highly organised and closely
monitored. This is precisely why the organisation of the network constitutes a
substantial entry barrier and one that is not overcome by the easy and rapid
recruitment of a nominal network.
145. Nauta testified that on a practitioner basis, once each doctor is treating
between 200 and 400 lives, a capitation proposal is put. 71 According to him,
the model would not work for fewer than 100 000 lives, but this depended
upon the extent of the network: 72
“ DR NAUTA : Possibly not as much as I ... you know, I think if you have 10
doctors, then you can capitate with 20 000 lives. You know what I mean? But
nationally to make sure that your footprints stay nationally and that you don’t
lose your peripheral doctors, because nobody ever goes there and you always
need them, it’s very critical to get a big contract, if you can’t cope with the
pensioners or the wives.
In the South African world a lot of wives live in rural areas and their
husbands work. Then you can’t get the contract, because you’ve got to have
husbands work. Then you can’t get the contract, because you’ve got to have
someone in Umgoma (should read ‘Nongoma’) that’s also a Carecross doctor
to fulfil the promises that you’ve done and if you then are that big – in our
case roughly 700 sites – then you need a 100 000 lives, I would guess. I mean,
you know it’s not a scientific thing, but that gives you enough interest by all
the parties to be the glue that sticks it together and as it grows, clearly things
got easier on our side and the whole model maintains itself.” 73
146. According to Mr. Patterson, a critical mass of 90 000 100 000 lives was
required before Prime Cure was able to turn a profit. 74 Similarly, in Prime
Cure’s Limited Confidential Information Memorandum, there is reference to
this critical mass of approximately 100 000 lives having been reached. 75 The
Medicross due diligence of Prime Cure also emphasises the importance of
volumes in overcoming entry barriers. It identifies specifically that Prime Cure
has researched critical mass and that any further lives add to the bottom
71 Transcript page 27
72 Transcript page 26
73 Transcript page 31
74 Transcript page 857 referring to File C page 168, Income Statement of Prime Cure for 20034
75 See record File C page 310
35
line.76 A number of the submissions to the Commission from small schemes
or administrators state that this is typically a ‘numbers game’. 77
147.The merging parties insist that the established medical schemes administrators
are particularly well placed to overcome any barriers to participating in the
relevant market.
148.The views of Discovery Health, the country’s largest medical schemes
administrator, were represented at the hearings by the head of its health
department agreements division, Mr. Strauss, who presented a particularly
coherent analysis of barriers to entry. He categorised entry barriers under three
headings. The first is capacity in administration. This comprises the ability to
receive, adjudicate and process claims, and to pay them over to the appropriate
party. It also encompasses call centre query resolution.
149.Secondly, financial resources are required. This entails the ability to finance
an organisation that is taking risk and to provide for claims volatility.
Furthermore, infrastructure is required for an organisation that is starting up
and assuming risk, and it is necessary to finance that infrastructure while, in
the first phase of entry, income is limited because of the small number of
insured lives.
150.Finally, network management skills – that is, the ability to manage networks
of primary care providers are fundamental. Strauss believes that Discovery
does not possess these lastmentioned skills. Indeed it is precisely for want of
these skills that Discovery had not, at the time of Strauss’ written submission,
decided to enter the market. The network managers have to possess both an
intimate knowledge of the workings of the primary care market and an ability
to micromanage the doctors participating in the network. He summarises the
primary care network skills required as:
primary care network skills required as:
“… being able to predict the utilisation patterns, being able to price, having
sufficient data to enable one to price particular procedures or particular
consultation rates, to determine how much one should be paying on the one
hand to the providers of service, and how much based on utilisation then one
could charge the members. The other part on (should read ‘of’) network
management is being on the ground and meeting with doctors and being sure
that they are managing in terms of the expected unitisation (should read
‘utilisation’) and entering into any risk sharing agreements that you can with
them.”78
151.Strauss contended that Discovery possessed neither of these latter skills.
Discovery’s established schemes tended to attract middle and upperincome
individuals and hence it was not familiar with the lowincome market. And
the ‘medical savings account’ concept pioneered by Discovery was deemed
76 See record File B page 350
77 See record File D page 92
78 See transcript pages 1445
36
providerunfriendly because it provided an effective break on utilisation.
‘The results of us not being au fait with the lowincome market and not having
very good relationships with primary care providers for the reasons around the
savings accounts, made us believe that we should not be ….if there were other
organisations who had better relationships in that market and a better
understanding of that market, we should leverage off that expertise.’ 79
152. Strauss also argued that Discovery would, in attempting itself to manage a
primary care providers network, encounter particular difficulties in ring
fencing its low income options, in other words in persuading the primary care
providers to accept a fee structure that distinguished to a significant extent
between members of different Discovery schemes. 80
153.What then do we make of Discovery’s decision to enter the market, a decision
which was announced on the eve of our hearings? In essence Discovery has
announced that it has terminated its existing contracts with Carecross and
Prime Cure and that it will, as of the 1 st January 2006, manage its lowincome
option, the Key Care plan, itself.
154.We note firstly, that Discovery is best placed of all the medical scheme
administrators for relatively rapid entry into the relevant market. Discovery
entered the lowincome market – through its Key Care option – some two
years ago and, so, amongst the large medical schemes administrators, it is
certainly a first mover in this area. In so doing it has not only been able to
acquire some knowledge of the lowincome market but, more important, it has
a ready 90 000 lowincome insured lives with which to springboard itself into
the lowincome market. These refer to the lives insured through the Key Care
plan, Discovery’s lowincome option. These favourable entry conditions are
not mirrored in any of the other medical schemes administrators. It is, in
not mirrored in any of the other medical schemes administrators. It is, in
effect, reward for Discovery’s entrepreneurial approach to health care
insurance, for its willingness to test the risky lowincome market at a time
when its major rivals were, as Dr. Walters of Medscheme testified, content to
remain in their comfort zone of high and middleincome earners with a
limited exposure to certain lowincome closed schemes with limited
membership. Accordingly we do not believe that Discovery’s entry portends
an easy entry path for other medical schemes administrators. It is the product
of factors particular to Discovery’s relatively early entry into this market.
155.Secondly, Discovery’s success in this market is far from assured. It appears
that Discovery will opt for an open network, one in which the Key Care
member will select a primary care provider who, subject to agreeing to the
plan’s terms, will become a member of the network. We have already noted
Dr. Nauta’s critique of this approach to network construction – although he
acknowledges that Discovery may be somewhat aided by its brand and the
79 See transcript page 146
80 See transcript page 260
37
sheer size of its operation and by the fact that the Key Care plan is certainly
pitched at the upper end of the lowincome market, the jury is still out on
whether or not Discovery will overcome the difficulties in network
management to which Dr. Nauta refers. Certainly, the Discovery network is
nowhere near ready to operate:
[…quote confidential….. “81]
156.And there is also the vexed question of assembling an open network to service
a Discovery scheme where many, possibly all, of the members of the network
will be serving other Discovery members on significantly different terms and
conditions:
“ Adv Unterhalter : Yes, so its not as if you have to assemble this afresh. You
have already existing relationships with these doctors, or some anyway.
Mr Strauss : There is a big difference. The existing relationships we talk of is a
relationship where those same doctors service our broader population. What we are
asking them to do for this product is to ringfence differentiated pricing.” 82
157.Discovery itself is clearly circumspect in its own assessment of its prospects
for the establishment of a successful network. We share that caution and,
accordingly, conclude that Discovery’s entry and, certainly, the sustainability
of that entry, is by no means a fait accompli.
158.This brings us to our third comment on Discovery’s entry. It is clear that
Discovery has been intent upon removing the administration component from
its contracts with Carecross and Prime Cure. […….CONFIDENTIAL……] It
is also speculated that Discovery may have been concerned that its core
function – administration – was being contracted out to a third party and, at
that, to a potential competitor in administration in a growing segment of the
medical schemes administration market. Strauss makes it clear that first prize
for Discovery was that it reclaimed and internalised the administration
for Discovery was that it reclaimed and internalised the administration
component of the contract while Carecross continued to manage the network.
Carecross refused to accept this and so, it seems after intense negotiation, the
entire contract was cancelled. Given these uncontroverted facts, it does not
seem unduly speculative to suggest that when Discovery confronts the
difficulties in organising a network and Carecross contemplates whether or not
the proverbial halfloaf is better than none, that it may well transpire that the
managed care function crucial to sustainable lowcost healthcare options,
namely the management of the network, will revert to Carecross and Prime
Cure, companies with a successful track record in this area.
81 Transcript page 261 our emphasis
82 Transcript page 262 Dr. Nauta in fact argues that this factor – the difficulties of a single brand
managing options pitched at highly diverse markets and the brand devaluation and sheer confusion to
which this gives rise – will significantly inhibit the ability of the existing administrators to enter the
lowincome market. See Transcript pages 77ff
38
159.It is interesting that an important witness for the parties, Dr. Walters, the
managing director of Solutio Healthcare Management, the managed care
division of the large medical schemes administrator, Medscheme, provided
some of the most cogent testimony in support of the view that entry barriers
are indeed high.
160.We have already noted that, despite the omission of any reference to Solutio
in their initial filings, the parties now attempt to present this entity as a
particularly significant competitor, or, at least, potential competitor in the
market for the provision of managed care services to low income health
insurance options. We have shown that this represents, at best, a heroic view
of Solutio’s current position in the market and an exaggerated and highly
speculative view of its future prospects.
161. Walters’ testimony establishes that Solutio has assembled an impressive
capacity for undertaking managed care. 83 This bears out a point made earlier
to the effect that managed care concepts and instruments have played an
important role in efforts to control costs – though largely in the areas of
secondary and tertiary care provision – even in schemes directed at middle
and highincome categories. However, this formidable array of managed care
skills does not seem to have assisted Solutio in significantly penetrating the
lowincome sector thus bolstering our view that we are here dealing with what
is essentially a new market. Solutio has been assembling a network of primary
cure providers since 2002. Walters testified that they have in this time signed
up 4000 practices on the Solutio network. However the level of organisation
of the network has clearly proceeded little beyond the signing up stage – it is
of the network has clearly proceeded little beyond the signing up stage – it is
not, in other words, a well organised network. When asked how many doctors
were members of his firm’s network, Walters simply replied: ‘your guess is as
good as mine’. 84 Walters clearly acknowledges that Solutio will only
gradually evolve from a company focused on ‘benefit management’ – the
form that managed care takes in relation to the middle and high income
insurance options – to one focused on ‘relationship management’, the form
that managed care will take in relation to capitated insurance options. This
suggests that he expects a slow and gradual growth in the number of low
83 Walters described the large infrastructure that Solutio has an actuarial division, comprising 9 full
time healthcare actuaries who service Solutio’s 23 medical schemes in the form of costing benefits and
doing projections, developing annual contribution tables for those medical schemes; a clinical
department which develop the benefit structures of each medical scheme ; a medical division,
comprising 3 professors and about 20 full time doctors with experience in managed care; a Benefit
Management Departments responsible for monitoring the utilisation by providers of benefits, as well as
the price of benefits, a hospital benefit management department with over 200 fully employed
doctors, nurses and pharmacists, tasked with preauthorising admissions to hospitals on a casebycase
basis; medicine management, comprising about 150 professionals which deals with dental benefit
management, optometry benefit management, pathology benefit management according to clinical
rules; a Disease Management department , comprising Wellness Management looking at preventative
care , also manned by clinical people, mostly doctors and nurses. Finally, Solutio has the contract
management division which appears to manage its networks, contracts with hospitals, with general
practitioners and specialists, including optometrists and dentists. Walters asserts that its network
contract management division is actually built on all the other resources. See transcript pages 516519.
84 Transcript page 519
39
income lives signing up for GEMS. 85
162. In summary, Solutio appears to be providing a capitated managed care option
to one of the medical schemes administered by Medscheme, this being
DCMED, the closed medical scheme for DaimlerChrysler employees. By
Walters own admission this is an atypical lowincome option. Certainly the
capitation fee paid to the doctors on the network is unusually generous. 86 As
indicated above, it appears that the primary managed care component of
Medscheme’s low income options is contracted out to Prime Cure and
Carecross, […….CONFIDENTIAL……].
163.At best for the parties, Solutio’s experience suggests that relatively small,
closed schemes, and preferably those that are administered by Medscheme,
that are regional in nature, and that are complemented by an active human
resource management function in the firm whose employees are members of
the scheme, represent its most likely potential customers. We note that the
schemes which Walters claims are about to desert Prime Cure and Carecross
in favour of Solutio, appear to fit this profile. If one accepts that much of the
growth in the lowincome market is going to be in large, national open
schemes, this does not, on its own, suggest a significant future role in the
relevant market for Solutio.
164.And nor, when one considers Dr. Walters’ views of entry barriers into this
market, should his modest view of Solutio’s future role be surprising. It is
worth quoting Dr. Walters’ eloquent testimony at some length.
165.He describes, in some considerable detail, the steps involved in organising a
network of primary care providers. The network to which he refers is clearly
an ‘open’ network, similar to that which Discovery now intends assembling.
What emerges from the following quotation is that while it is easy to sign up
What emerges from the following quotation is that while it is easy to sign up
large numbers of doctors to an open network, it is clearly extremely difficult to
utilise this essentially unorganised network as an effective instrument of
managed care:
“The first step in a network, in a proper network, is to link every single
beneficiary in a medical scheme to a certain General Practitioner. The
beneficiary must make a decision that this is then the General Practitioner
that I am going to consult. The second step is for that General Practitioner to
contractually accept all the clinical responsibilities surrounding that patient,
that beneficiary. That contract is usually between the doctor and ourselves as
a managed care company, as a network company, a managed care company
with network capabilities.
So the first step, beneficiary links to a doctor. The second step, doctor
assumes responsibility. The third step, profiling that doctor to ensure that
85 Transcript page 596
86 Walters describes the capitation fee paid to the doctors on this network, essentially a single East
London based IPA, as ‘quite rich’ page 521
40
that doctor meets those contractual obligations. The next step, if the doctor
meets the contractual obligations which are both financial and clinical, you
will reward that doctor. That doctor will get certain benefits from that.
Whether it be financial or whether it be that you don’t tamper with his
practice at all, but there will be rewards in it for the doctor. But if the doctor
does not meet his contractual obligations, the doctor is warned. He gets a
time period to actually do some self improvement, change his behaviour
patterns and if he still does not change his behaviour patters, he is referred to
his peers, to his fellow doctors that contractually his (should read ‘he’s’)
already selected to be his peers that will review his case if he is errand
(should read ‘errant’).
So, in that case his case is then referred to a group of his peers, who meets
like this and they look at all the data and information and they try and help
him, but if then he is not…he doesn’t prove that he wants to be helped, there
are then penalties. He could be kicked off the network. He could ….there are
several penalties. Now that peer review mechanism you will understand is
critical if you want to have cost effective quality care within a network of
doctors. Now, usually we employ the IPA, the local doctor independent
practitioners association to perform that function.
[…….CONFIDENTIAL……].”87
166.When asked, under crossexamination, why Solutio, despite its formidable
managed care capacity, continued to contract with Medicross and Faranani,
Walters replied:
“Because we cannot do peer review. It should be the peers that should be
reviewing the peers. We cannot unilaterally sit in an ivory tower and structure
standards of good practice. We can do that, but it’s unfair and it’s not good
practice to do that, to sit in an ivory tower and say, this is what you shall do.
We’ve got to collaborate with these people. That works. They’re part of the
We’ve got to collaborate with these people. That works. They’re part of the
business. They’re part of the future. We’ve got…and if we move towards real
risk sharing, then they must be enabled to be part of the risk sharing and this
is just building up to that moment in time.” 88(our emphasis )
167.In the course of explaining why Solutio had found it necessary to hire a third
party, […….CONFIDENTIAL……], to oversee the networks with which
Solutio was contracted, Walters emphasised […….CONFIDENTIAL……]:
“Pure (this should read ‘peer) management being where we identify a certain
doctor not meeting clinical or cost effectiveness criteria, we need to refer that
doctor to peer group to make a decision about that doctor status.
[…….CONFIDENTIAL……].”89
87 Transcript page 5278
88 Transcript page 581
89 Transcript p526
41
168.Walter’s scepticism of the regional IPAs is reinforced by the store that he sets
by national networks:
“We need networks with national footprint. Currently we have too few. We
think we’ve got a network with a national footprint. I’m pretty sure Discovery
will have a network with a national footprint. We need networks with a
national footprint that can actually compete with one another, and currently
we’re playing around with small numbers of doctors within networks and
uncompetitively priced products. That’s the challenge…’’ 90
169.His prognosis for other of the fringe players in the market is equally
pessimistic. Of Faranani, he says:
“ […….CONFIDENTIAL……]: ”91
170.Again, on the difficulties of practicing managed care in the low income
market generally and, particularly in organising and maintaining networks:
“ ADV BERGER : So you took a decision in 2002 that you had to move into
this market, this capitated managed care market…
DR WALTERS : And into the total networking market, which capitation … I think you
are focussing totally on capitation and that might be not the right focus in my
opinion. In my opinion the question is have you got an effective network of providers
with proper contracts where the members are educated and understand what it is all
about. That’s the biggest problem why healthcare is failing in South Africa, is
members aren’t well educated and do not understand what these lowcost options
actually mean.
ADV BERGER : Now I assume that takes time to set up such a network.
DR WALTERS : That’s the issue. You need to educate the members. You need to have
a provincial infrastructure. You need to have client liaison offices that can actually
go and talk to the members, let alone the brokers don’t tell them the truth or tell them
the hard truth. Sorry, I strike that. You need proper brochures, which they never read,
the hard truth. Sorry, I strike that. You need proper brochures, which they never read,
which you’ve got to interpret to them. You need to link them with doctors. You need to
educate the doctors. You need to contract the doctors. The contracts need to be
quality contracts. They need to have all the obligations written in. You need to then
monitor those. You need to profile those. You need to have the ability to interpret the
results from the profiling. You need to then develop the contributions and the benefit
tables. You need to ensure that the model is viable. You need to go on like this
forever. It’s a huge amount of building, which we’ve decided to embark on in 2002
and which is evolving as we speak to meet the needs of the marketplace.
ADV BERGER : So you’ve been building this for the last 3 years.
DR WALTERS : Yes.” 92
171.And then, further, Walters provides a graphic description of the difficulty of
managing networks in one of the schemes of which he has direct experience:
90 Transcript p588
91 Transcript p563. Note that Dr. Nauta concurs with this assessment.
92 See transcript page 553, our emphasis.
42
“ DR WALTERS : You see that’s not, it’s such a difficult question because
there’s so much work to be done. You need to go on national road shows, you
know with the Sasolmed doctors I have monthly meeting with them on
business issues. I have monthly meetings with them on clinical issues, it’s a lot
of time, it’s so time consuming. Now for the IECA network we’re doing the
same. For the DCMED we’re doing the same. I’m constantly in the air of
going somewhere to meet with them. You need an infrastructure to actually
just to organise all these meetings.” 93
172.And on the importance of financial capacity and the role of the quantum of
lives in the entry process, Dr. Walters testifies:
“… ADV BERGER : Yes. And you also need to have deep pockets, at least in
the initial stages.
DR WALTERS : That’s the riskbased capital that you need to set aside, and there is
an actuarial formula based on the number of lives and the chronicity, people with
chronic diseases and gender and all those kinds of things. There is a formula that you
calculate your riskbased capital that you need to set aside.” 94
173.And further:
“But as an interim measure you can’t capitate all the doctors on the ground.
You can only capitate those that have the necessary volume of patients and
that have the necessary expertise. So it’s a process that you go through in
order to have a fully capitated environment. And that’s the model that we are
… there are also other refinement that you can get pools of doctors and pay
them on a budget, which is also a type of risk sharing, although not
capitation.”
So, I’m not trying to give you a long story about this. I’m just saying that the
capitation that you are talking about is a journey. It’s not a snap and there
you’ve got a capitated network. It’s a journey.” 95
174.And clearly it is a journey, the successful conclusion of which requires a
well organised network and a rapid growth in membership.
well organised network and a rapid growth in membership.
175.Despite this apparent state of ‘unreadiness’ Mr. Walters testified that
Medscheme/Solutio would be tendering for the Sapphire and Topaz lowcost
options in the GEMS. Clearly, by Walters own analysis of the requirements
for providing capitation – an express requirement of these two options –
Solutio is not yet ready. This seems to evidence the widely held belief that the
lowcost options on GEMS will grow very slowly.
176.We have then examined the entry prospects of South Africa’s two largest
93 See transcript page 589
94 See transcript page 562
95 See transcript p559
43
medical schemes administrators. In our view, Discovery’s entry is the product
of circumstances peculiar to Discovery, in particular its firstmover advantage
that has enabled it to begin with a membership base which, though
significantly smaller than its targeted projections, will act as an important
springboard. However, we are not yet persuaded that even Discovery will
sustain this entry.
177. Medscheme through Solutio is, in our estimation, some considerable distance
from a competitive, sustainable presence in this market. It is, by Dr. Walters’
own admission, a late and somewhat reluctant entrant into the lowincome
market.96 Solutio has, to be sure, assembled an impressive managed care
capacity but this seems to consist largely in a data gathering and analysis
capacity directed at reducing costs in Medscheme’s traditional middleto
upperincome market. Solutio has a limited track record in the successful
utilisation of primary care networks and where this occurred – with the
Daimler Chrysler scheme representing its only sustained success – it has been
assisted by unusual circumstances.
178.The merging parties contended that were other fringe players, in addition, that
is, to Discovery and Medscheme, that were poised to enter the relevant
market. Medical scheme administrators Old Mutual, Sizwe and Metropolitan
were mentioned as were the IPAs, regional networks of primary care
providers.
179.The basis for the contention that Old Mutual was contemplating entering the
market appears to be its written submission to the Commission in which it
indicated a desire to enter this market and a recent letter addressed to doctors
in which it appears to be soliciting membership of a primary care providers
network. In fact Old Mutual is clearly some way from possessing the attributes
network. In fact Old Mutual is clearly some way from possessing the attributes
necessary to overcome the identified barriers to entry. We have already
indicated our scepticism of networks organised in the manner that Old Mutual
has chosen. The submission referred to makes vague mention of possible
synergies that Old Mutual’s property division may provide in the
establishment of a network of primary care clinics. Several witnesses –
notably Dorfling and Nauta – have called this mode of entry into question.
Even if the establishment of a bricks and mortar clinic network is a viable
mode of entry it will clearly take some considerable time to set this up. It is
clear to us that Old Mutual has not given much thought to entering this
market.
180.Old Mutual does not even seem to have entered the market for the
administration of lowincome health insurance options. The strongest
indication of its intention to do so is its recent acquisition of Sizwe, a small
medical scheme administrator that does have exposure to the lowincome
market. However Sizwe clearly does not have the capacity to provide
capitated managed care products for this market. Sizwe’s witness in the
96 See Walters’ apologetic attitude to Medscheme’s reluctance to tackle this market . Transcript page
566
44
hearings – Mr. B. Singh – insisted that the regional IPAs constituted a ready
supply of primary care networks. He argued that the regional limitations of
these networks could be overcome by the expedient of entering into contracts
with expanding numbers of these in order to achieve national coverage. Other
witnesses have, as noted above, already indicated their scepticism of the IPAs’
ability to provide risktransfer managed care products. We share this. Sizwe’s
experience of providing risk transfer managed care products to lowcost
options appears, not unlike Medscheme, to have been confined to small,
regionally confined medical schemes and there is no evidence to suggest that
this will translate into a ready ability to serve large, open, national schemes.
181.We should note that even if the large medical schemes administrators are
easily able to overcome the barriers to entry – and we do not accept that this is
the case – their presence will provide cold comfort to the smaller
administrators and to schemes not administered by the large administrators.
The merging parties insisted that the large administrators would be willing to
sell their managed care services to those schemes and administrators who do
not possess these capabilities. In fact both Mr. Strauss, on behalf of
Discovery, and Dr. Walters on behalf of Medscheme, indicated that while they
would, in principle, be willing to provide these services to schemes that they
did not administer, neither expected this to occur on a significant scale. They
reasoned that the administrators of these schemes would fear losing their
administration business to the large administrators. 97 Dr. Walters also
identified technical difficulties in Solutio providing managed care products to
plans not administered by Medscheme:
plans not administered by Medscheme:
“Now when you ask me about other schemes, the problem is usually the
systems. How do you interact and interface with schemes not administered by
Medscheme. How do you get the processes to align and the systems to talk to
one another. Now, that is a problem. But we’ve had extensive experience
over the years...We’ve provided services to Prosana, to Open Plan, to
Transmed, to Bestmed, to Lamaf (should read ‘Camaf’), to Munimed, to
Selfmed and to Topmed. Over time we’ve made a corporate decision. A
corporate Medscheme decision was made to focus more on the schemes that
are administered by Medscheme.
It makes it easier. It makes for more efficiency and therefore our costs come down.
Our prices come down. So, we have to a large extent got rid of these schemes.” 98
97 See Transcript page 225. Also see page 298 where Strauss alludes to transfer of sensitive member
information to a primary care service provider who could potentially be a competitor of theirs. Note
that this question also ignited a long and inconclusive debate regarding the independence of trustees
visavis the administrators. Hence it was said that if the trustees wanted a managed care provider
attached to a rival administrator, then their will would prevail over their own administrator’s protective
instincts. We cannot resolve this debate here. Suffice to say that while the trustees are certainly
supposed to be independent, this is clearly more plausible with respect to the small closed schemes
than with the large open schemes. Dr. Walters’ account of how Solutio has gone about persuading the
trustees of several closed Medscheme administered schemes is, in our view, evidence of the power of
the administrators over the trustees . See transcript page 534
98 See transcript page 524
45
182.In summary we do not believe that the barriers to entry will be overcome
easily or rapidly by either the administrators or by other primary care networks
such as the IPAs. We do, however, believe that the most likely source of
competition for the two largest players in the relevant market that we have
identified, namely Carecross and Prime Cure, is likely to emanate from
Medicross/Netdirect. Medicross is already in this market albeit at the higher
end. It also has an established clinic network which Mr. Dorfling indicated
would constitute the ‘hub’ of a larger primary provider network. The Netcare
group has, for long, attached importance to the assembly of a network of
primary care providers and Netdirect is living proof of this. Although the
network is not tightly organised, the networks of clinics that belong to the
merging parties alone will facilitate the development and tighter coordination
of the more extensive doctors’ network. Also Netcare has attempted to
cement the ties between the Netcare group and the doctors’ network through
enabling the doctors to participate in Netcare equity. In short, conditions
favour an early Medicross/Netdirect entry, a process which, in fact, appears to
be well under way.
183.We conclude then that the entry barriers surrounding this market are indeed
formidable. Nor, despite Discovery’s recent decision to enter the market, are
the large players like Medscheme well positioned to enter this market in the
near term. Discovery’s ability to do so is strongly conditioned upon its early
entry into the lowincome health insurance market.
The level and trends of concentration, and history of collusion, in the market
184.This is, as we have earlier observed, a new market. It is also a market in
which sustainable entry has proved manifestly difficult. The evidence is that
which sustainable entry has proved manifestly difficult. The evidence is that
several large and reputable firms have entered the market only to exit, having
failed to develop a sustainable presence. The market is, accordingly, highly
concentrated. This merger would serve to increase that concentration. Only
three firms (Carecross, Prime Cure and Medicross) have managed to sustain a
presence in the market with a small number of others operating on the fringes
of the national market (Faranani) or in regional niches (several IPAs). Not
only does this transaction merge the second (Prime Cure) and third
(Medicross) largest of the three firms that have proved capable of sustaining a
presence in the national market, thus accounting for the Commission’s
description of this as a ‘three to two’ merger, but it will result in the co
ordination of the merged entity with the only wouldbe entrant, Netdirect, that
is well positioned for entry in the relatively short term. Hence the level of
concentration is high and this merger exacerbates this. To the extent that the
embryonic nature of this market permits of any trend analysis, the high failure
rate of wouldbe participants may be said to point towards a trend towards
greater concentration.
185.There is no evidence that suggests collusion between the existing participants
in the relevant market. We note, however, that should one of the nonNetcare
46
hospital groups, namely Mediclinic and Life Healthcare, wish to participate in
a full risk capitation scheme (that is, a scheme that offers capitation at the
primary, secondary and tertiary care levels) it will be obliged to offer its
services to a managed care provider capable of delivering primary care
services to low income options. To the extent that Netcare’s rival hospital
groups are reluctant to enter into a fullrisk capitation arrangement where the
primary care component is in the hands of a member of the Netcare group, it
will be forced to turn to the only remaining provider of these services, namely
Carecross, which will thus enjoy considerable market power in relation to the
those who will find it necessary to partner with it in order to participate in the
provision of fullrisk capitation. Mr. Brian Davidson, the Life Healthcare
group representative who testified at these hearings, clearly indicated his
discomfort at Life Healthcare assuming risk in respect of tertiary provision
where the primary component was in the hands of an ‘unfriendly’ party:
“Now there is a, I am using this by way of example to answer your question,
there is a provision made for a primary care network service the same as
Gems option and we’re saying to ourselves, hang on, if say we have a non
friendly, to use that word, primary care network or a primary care network
that belong to one of the competitors, is that additional risk to us? Is it
possible that they could pass on or cost shift the risk to the hospital’s little
cost centre. I don’t know who is going to be the managed care organisation
who is going to be managing the risk between all of the parties who will be
contracting with this particular option, because that again is also out for
tender as is the Medical Scheme Administration itself.
tender as is the Medical Scheme Administration itself.
I would like to, we would like to think that a neutral efficient and effective
managed care organisation should be able to correctly manage and prevent
any cost sifting between the various institutes. I am not an expert at that
process, therefore I don’t know how it is going to work. So, to answer your
question a long way, we would be worried about contracting with one of our
competitors’ subsidiaries for that reason” 99.
186.However, even more disturbing from a competition point of view is the
prospect that Netdirect’s entry into this market will facilitate collusion in the
allimportant private hospital market. Three possibilities arise from the
participation of Netcare group interests in primary care provision for low
income consumers. Either this will effectively preclude – or, at least severely
discourage the other hospital groups from participating in the tertiary
component of full risk capitation (thus giving Netcare associated companies
market power in relation to the purchasers of fullrisk capitation) or, as
outlined above, it will effectively oblige those of Netcare’s rival hospital
groups who wish to participate in the provision of fullrisk capitation to
purchase the primary care component from Carecross (thus giving it
considerable market power). But possibly the most disturbing prospect is
precisely that Netcare’s rivals will find their way clear to negotiating their
participation in the tertiary component of a fullrisk product with a managed
care company that is part of their rival group. This will mean that Netdirect,
99 See transcript page 315
47
part of the Netcare group, will be negotiating capitation fees with Mediclinic
and Life, Netcare’s rivals in the private hospital market, thus further
facilitating the flow of information between the hospital groups and this
explicitly in the areas of costs and prices and covering the core competitive
strategies of the three groups. This is of particular concern in a market that
wellplaced commentators have already described as a cartel. A previous
decision of this Tribunal noted the following assessment by an investment
banker of the private hospital market:
“The strategic behaviour of these groups has historically been characterised\
by a conscious avoidance of price competition. Rather than attempt to
aggressively win market share through price wars and intensive advertising
campaigns, the hospital groups – via their joint membership of the Hospital
Association of South Africa (“HASA”) – have managed to standardize
industry pricing by agreeing set tariffs with the Medical Aids represented by
the Board of Healthcare Funders (“BHF”)…
The key issue will be the extent to which the dissolution of the formal,
collective price setting arrangement in favour of onetoone negotiations will
increase the likelihood of price competition amongst the primary service
providers. On the face of it, the encroachment of the Government on the
private sector (via the establishment of private wards) and the diminishing
growth opportunities in the top end of the local market could provide an
incentive for one of the primary service providers to break ranks and initiate a
price war in order to increase market share and sustain the growth
performances that shareholders have grown accustomed to. This is, in our
performances that shareholders have grown accustomed to. This is, in our
view, unlikely. The primary service providers have operated as a cartel over
the past 3 years and have established exceptionally healthy profit
margins”.100
187.The Tribunal went on to note that, Mr. Richard Hogben, a previous CEO and
Chairman of Afrox Healthcare and currently a nonexecutive director of Life
Healthcare, in commenting on this assessment, conceded that the private
hospitals did not compete on price. He described the competitive dynamics of
the market in the following terms:
“…The basis of competition between private hospitals is about several
elements, of which price is not really one…The basis of competition between
a hospital is distinct units. It’s about its location. It’s about the quality of the
doctors that it has that work there and the quality of the doctors that work in
those hospitals is really driven in many ways by the quality of the hospital
facility and the quality of care that is given in that hospital…The question of
price as a competing factor between the hospitals is of lesser significance,
unless it becomes extreme.” 101
100 See Business Venture Investments 790 (Pty) Ltd and Afrox Healthcare Limited – 105/LM/Dec04
paragraph 59, referring to the transcript of 10 February 2005, pages 103104 as well as pages
12271228 of File 3 of the merging parties’ subsequent filings.
101 Refer to pages 105106 of the transcript of 10 February 2005.
48
188.We recognise that this may not be a mergerspecific effect. Netdirect will
enter the market regardless of this transaction and so the opportunity for
information sharing that it provides will be there whether or not the merger
takes place. However, the removal of a rival – Prime Cure – to Netdirect and
Medicross, increases the likelihood of a relationship between Netcare, on the
one hand, and Mediclinic and Life Healthcare on the other and certainly
aggravates our concerns regarding the future state of competition in a vital
related healthcare market.
The dynamic characteristics of the market, including growth, innovation and
product differentiation
189.We have already commented at some length on certain of the dynamic
characteristics of this market. In summary, this is a market whose
environment is unusually fluid and uncertain. Direct government provisioning
– most notably in the supply of hospital services – is a ubiquitous feature of
the market and will remain so. This impacts on the supply of all healthcare
services and products, including pharmaceuticals. Private provisioning of
healthcare services is, if anything, more pervasive but it takes place in the
context of wideranging regulation including of medical insurance and private
hospital services as well as regulation of the production, patenting, licensing,
dispensing and distribution, both wholesale and retail, of pharmaceutical
products.
190.The complexity that characterises the surrounding environment is
immeasurably compounded by the state of flux that seems to have become,
both in South Africa and elsewhere, a constant feature of the regulatory
framework as governments everywhere struggle to ensure the supply of basic
healthcare services to all of their citizens without massively compromising
healthcare services to all of their citizens without massively compromising
fiscal stability and sustainability. South Africa has certainly not escaped this
experimentation in healthcare provisioning, its own efforts severely
complicated by the AIDS pandemic. The courts, including the Constitutional
Court, have played and will continue to play a central role in determining the
character of the healthcare system with some crucial judgments pending and
further litigation undoubtedly in the pipeline.
191.Moreover, government intervention in healthcare provisioning has direct
reference to the market implicated in this transaction. We have outlined, at
some length, government efforts to relieve the overstretched public healthcare
system by moving a large proportion of those who utilise it to a private
healthcare system that, particularly in the supply of private hospital services, is
characterised by significant excess capacity. In order to realise this objective,
private healthcare funders are under considerable pressure to design insurance
options that are affordable to the large low income segment of the population
and that, in turn, can only be achieved through the development of
mechanisms that lower the cost of primary, secondary and tertiary healthcare,
including the cost of pharmaceutical products.
49
192.To this end government has registered a medical scheme – GEMS – that
includes options directed at lowincome consumers. It has called on private
sector firms to tender for providing the array of services necessary for the
effective functioning of the planned new insurance scheme. Evidence
submitted to these hearings has revealed the significant lack of certainty
amongst key players in the healthcare sector regarding the future character and
size of this scheme. And if other experiences of government intervention in
the healthcare system are anything to go by, litigation will inevitably
accompany the process of getting this ambitious intervention off the ground
and, in particular, the process of awarding the tenders.
193.As already elaborated, the parties to the transaction before us are amongst
those very few entities in the healthcare market that have successfully
delivered private healthcare to lowincome consumers. This has involved
considerable risk, the surmounting of significant entry barriers and constant
innovation and experimentation. But the market is still in the early stages of
its development. In this unusually dynamic context it is our view that
competition authorities should approach private interventions that will impact
on the structure of the market with considerable circumspection.
194.We know that the Netcare group will, through the medium of
Netdirect/Medicross, intensify its participation in this market irrespective of
whether or not this merger goes ahead. This is to be welcomed and
encouraged. Mr. Dorfling has clearly indicated that he believes that there are
sustainable lowincome options that do not rely on capitation. The clinic
networks of Medicross and Primecure feature prominently in his conception of
networks of Medicross and Primecure feature prominently in his conception of
the lowincome product that is to be offered as does Netdirect’s primary care
providers network. The vertically integrated Netcare group may permit of
modes of provision that are denied others who do not enjoy these links with
secondary and tertiary providers. There is undoubtedly significant room and
an urgent requirement for experimentation and innovation. We have little
doubt that a significant merger in this embryonic market will slow the pace of
innovation, it will reduce the number of alternative modes of provision on
offer, and it will likely slow the pace at which new forms and concepts of low
income healthcare insurance are introduced.
195.The parties insist that GEMS and other government initiatives guarantee rapid
growth in the demand for managed care products for the lowincome market
and that this will assure entry by players that have shown little appetite for
serving lowincome consumers. The record clearly shows that up until now
the development of this market has been slow and has not lived up to the
expectations of experienced healthcare providers. For example note that even
the aggressive and innovative Discovery has fallen significantly short of its
predictions for growth in Key Care, its lowincome option. In 2003 Discovery
had projected that within 23 years (that is, by 2005), the number of lives on
Key Care would be [confidential] when, in fact, Key Care only currently
covers approximately [confidential] lives.
50
196.Nor was the view that GEMS would account for massive, rapid growth in the
market shared by all of the witnesses in these proceedings, Dr. Nauta, for
example, said:
“…there’s a promise of many lives, up to millions. I’m very sceptical about
the ability to have a scheme of that size go so quickly. I just look at
Discovery’s own growth, which took them 10 years to get them to where they
are and this scheme says, I’m going to do twice as much in, you know, one
year, its going to be very difficult and keep in mind, those lives are all
essentially forced onto this.
So when you force a life into a scenario, you really need to be ready to deliver
and the doctors must be there and it must go smooth, otherwise people just
won’t go. It’s different when you voluntarily buy with all sorts of other
promises the way, you know, good, open schemes have done. So you’re
asking me what I think about them? I suppose it’s coming. It’s been
postponed by a year already in the past. To really have a big thing up and
running in 4 months from now, I think is totally impossible, but if it’s a
voluntary scheme and it says, guys you want to join, join, then it will slowly
grow and whether that growth rate is going to get them to 4 million or 2
million, whatever.” 102
197. Indeed, Nauta argued persuasively that the interest taken by most of the
established medical schemes and medical schemes administrators in the
GEMS tender – and which the parties have cited as evidence of significant
new entry into this market was centred around the prospect of losing existing
insured lives in the public sector, rather than at the prospect of gaining
thousands of new, hitherto uninsured lives. In other words, those interested in
the GEMS tender are, argues Nauta, not necessarily bidding for lowincome
lives. They are bidding for the lives of those members of the public service
lives. They are bidding for the lives of those members of the public service
who are currently insured, those able to afford insurance at present levels and
so who have joined existing schemes aimed at high and middleincome
earners:
“They have them right now. They will lose them. Forget about the new 500
000 that are going to come into the kitty still.
My personal conviction right now is that nobody really is there yet. We just need to
not lose suddenly 50 or 100 000 lives. There are at least 20 schemes that could go
out of business because they’re small. They’ll never get in here. But they have 10
000 lives. Small schemes like Conmed, run very successfully. Suddenly we know
their stats, it happens, and 80% of their lives will disappear. So everyone gets in
there to keep what they have, is my real view right now. And it’s logical that you’ll
do it. It’s easier to keep to what you have then to get the new stuff. And that’s where
particularly Discovery, who has just got Lamaf, which is a big scheme, in the semi
state, state world, they’ll lose them if they don’t get into this fold. So I think that is
the motivation right now.
102 Transcript page 44
51
And clearly if one day the State has enough money to subsidise the rest and
get them all in, that’ll come proportionally to those same players.” 103
198. Mr. Davidson, the witness from Life Healthcare, is also relatively modest in
his predictions of the conversion of currently insured public sector employees
to GEMS. The success or otherwise of the conversion – which Davidson
refers to as the ‘first phase’ – will determine how government enters the
second phase in which those who currently fall out of the net of insured lives
will be offered low income options. But that will still presuppose a significant
subsidy from government. 104
199. The Commission’s expert witness, Mr. Hodge, underlined the significant
uncertainties surrounding the growth of GEMS and in particular the inability
to predict which of those converting out of their existing schemes would opt
for the capitated options within the GEMS boutique. 105
200. Dr. Walters expressed the view that it would take ‘years’ for GEMS to grow
significantly.106 He conceded that he had ‘no idea when the market will grow
and how large the market will grow’. 107 However he clearly acknowledges
that Solutio will only gradually evolve from a company focused on ‘benefit
management’ – the form that managed care takes in relation to the middle and
high income insurance options – to one focused on ‘relationship management’,
the form that managed care will take in relation to low income insurance
options. This suggests that he expects a slow and gradual growth in the
number of low income lives signing up for GEMS. 108
201.Mr. Singh of Sizwe, called by the merging parties who identified Sizwe as an
active participant in the low income market, clearly articulated the extreme
uncertainty surrounding GEMS and the growth that it was expected to
generate:
generate:
“Mr. Singh : Chairperson, I could also say that there is a potential of 400 000
members. Like I said earlier, there is no science or any survey to say that the
models or the products offered by GEMS is going to be affordable.
Chairperson: Yes
Mr . Singh : So one of the scenarios then I could paint to you is that no one of those
400 000 members will join GEMS again because of affordability.” 109
202.The evidence then suggests not so much the certainty of rapid growth – as
103 Transcript pages 1178
104 Transcript page 319ff
105 Transcript page 385ff
106 Transcript p575
107 Transcript p577. Under reexamination he did confirm that he believed that the lowincome
market would grow.
108 Transcript p596
109 Transcript 9145
52
claimed by the merging parties but rather the significant uncertainty
surrounding the scale and character of that growth. In particular, all the
witnesses concurred that, in the initial years of the GEMS product, the focus
would be on the migration of those currently insured from their existing
schemes to GEMS. The lowincome options in which, to use Dr. Walter’s
characterisation, ‘relationship management’ rather than ‘benefit management’
was key, are some years off. For the first years of the GEMS era, the schemes
will be intent on maintaining their existing membership in the public sector.
203.We conclude then that the dynamic features of this market and its surrounding
environment serve to reinforce the likelihood that the merger will substantially
lessen competition. It is a new market surrounded by considerable regulatory
uncertainty. The further development of the product in question – capitated
primary managed care demands high levels of risktaking and investment in
innovation and experimentation. It is a market in which those concerned to
promote competition would wish to emphasise the importance of new
competitive entry and innovation of the sort promised by the acquiring party if
its attempts at entry through merger do not succeed. We find that the further
dynamic feature said to characterise this market – the rapid and significant
levels of demand growth that the parties have predicted and which they rely
upon for their argument that new entry will be significant – is less certain both
as to scale and direction than that predicted by the parties.
The nature and extent of vertical integration in the market
204. There are only three firms of significance in the market. Of these, one –
Medicross is part of a larger healthcare group, the Netcare group of
companies. Our impression is that, with the significant exception of the
companies. Our impression is that, with the significant exception of the
Netcare group, the healthcare sector has not been characterised by significant
vertical integration. This merger represents an extension of the degree of
vertical integration in the market. Discovery Health’s announced intention to
enter the market represents another instance of vertical integration. It was
suggested that as regulatory interventions limit returns in parts of the health
value chain – for example in schemes administration, in private hospitals and
in pharmaceutical distribution – the large players in these markets will look to
profit from participation in other parts of the health value chain. 110
205.The vertical issues at stake in this transaction were extensively canvassed in
the hearings and are examined below.
Whether the merger will result in the removal of an effective competitor
206.As already extensively elaborated this is a merger of two of only three
significant players in this market. There can be little doubt that, from this
perspective, the merger results in the removal of a significant competitor.
Prime Cure’s successful participation in this market rests on its network of
110 See Walters’ testimony, transcript page 568
53
primary care clinics and the larger primary care providers’ network that it has
established. Several witnesses – including Mr. Dorfling insisted that the
establishment of a new clinic network no longer offered a cost effective basis
for entry into the market. However, it is also clear that the existing clinic
networks owned by Medicross and Prime Cure would constitute the ‘hub’ of
the merged entity’s strategic approach, with the extensive Prime Cure and
Netdirect primary care provider networks comprising the ‘spokes’. The
obstacles in the way of the formation of a new clinic network, combined with
the support for an approach to managed care in the lowincome market that
relies on the combination of a clinic network and a primary care providers’
network, serve to reinforce our view that the merging of the only two entities
that command access to both clinic and provider networks removes an
effective competitor in circumstances where the competitive advantage
enjoyed by the target company will not be easily replicated.
207.Although it was suggested that the target firm, Prime Cure, has, in the recent
past, experienced financial difficulties, the failing firm defence was not
invoked by the merging parties. It appears that earlier efforts to sell Prime
Cure had foundered because of the difficulties it was experiencing at the time.
It then appeared that the Prime Cure shareholders became actively engaged
with the management of the company in an effort to place the company on a
sounder footing precisely on order to enable the shareholders to exit their
investment. Medicross’ desire to absorb Prime Cure suggests that these efforts
to turn around the target have borne fruit.
208.The successful turnaround of Prime Cure notwithstanding, its shareholders
208.The successful turnaround of Prime Cure notwithstanding, its shareholders
still intend to exit the investment. This was confirmed by Mr. Patterson, a
witness representing Prime Cure’s largest shareholder, Brait. 111 Patterson
noted, however, that a rejection of this transaction on competition grounds
would significantly hamper efforts to sell Prime Cure which, he averred,
would only attract a suitor from within the industry. However, if other
institutional investors are persuaded that Prime Cure has a foothold in a
market which is poised to grow significantly and in which entry barriers limit
the prospect of new entry – views that, as we have elaborated, appear to be
held by the both of the merging parties – then there is no obvious reason why
new buyers should not be found.
209.Nor does our finding that the merger currently proposed is likely to lead to a
substantial lessening of competition preclude other firms in the healthcare
sector from acquiring Prime Cure. This decision is predicated on the fact that
the buyer is already active in this market and that the group of which the
acquiring firm is part is well placed to intensify its involvement in the market
even in the absence of the proposed merger. There are many powerful entities
in the healthcare sector that are not in the same position as
Medicross/Netdirect and to whom the same strictures are, accordingly,
unlikely to apply. The Prime Cure shareholders may, to be sure, have to forgo
111 Transcript page 854
54
part of the ‘strategic’ premium that Medicross is willing to pay. 112 However,
in our estimation, the ‘strategic value’ amounts to little more than the market
power that will accrue to the acquiring firm from the elimination of one of its
few competitors and the heightening of entry barriers that will confront new
entrants, even those wouldbe new entrants already active in the broader
healthcare sector.
210.The acquiring company has also attempted to justify the merger on the basis
that the merged entity will, as a result of the combination of the assets of the
two companies and the financial strength of the Netcare group, be better
placed to develop costeffective products for health insurance options aimed at
lowincome consumers. This argument may have some salience in
circumstances where the market is increasingly dominated by a firm that is not
party to the merger and where the merger is then effectively a defensive
response to the growth of that rival. However, this is clearly not the case here
and there is no a priori reason why the effect of the lessening of competition
in consequence of the merger should be countervailed by the superior
resources of the merged entity. This is, in effect, an efficiency argument and,
as such, is dealt with below. We have, in a previous decision, indicated our
scepticism of this argument for a more ‘effective’ competitor and then in a
situation where the need for a defensive strategy against an increasingly
powerful competitor was more clearly established than in the case of the
transaction presently before us. 113 In the circumstances of the present market
we have two firms – Carecross and Prime Cure of broadly similar strength,
and a third – Medicross/Netdirect – that is well placed to compete effectively
and a third – Medicross/Netdirect – that is well placed to compete effectively
with the two market leaders. Each of the firms has distinct competitive
strength and strategies. They should be afforded every opportunity to develop
these.
211.We find, then, that the horizontal dimensions of this merger are likely to
lead to a substantial lessening of competition in the relevant market.
The Vertical Dimensions of the Merger
112 Transcript page 825
113 See Ellerine Holdings Ltd and Relyant Retail Ltd – 56/LM/Aug05
55
212.The Commission, as well as a number of witnesses who testified at the
hearings – notably Mr. Strauss of Discovery Health, Mr. Davidson of Life
Healthcare and the Council for Medical Schemes – have made much of the
vertical dimensions of this transaction. However, while the Commission is
clearly concerned at the impact on competition of the transaction’s vertical
aspects, it is not certain how much weight these considerations were given in
its decision to recommend that the merger be prohibited. It is our finding that
the merger falls to be prohibited on its horizontal dimensions alone. While the
Commission’s careful scrutiny of the vertical dimensions of this merger is
welladvised and the anxieties of the witnesses regarding the progressive
vertical integration of the Netcare group is appreciated, the evidence does not
lead us to conclude that the vertical dimensions of this transaction will give
rise to a substantial lessening of competition in the relevant market.
213.Simply stated, the vertical dimensions arise from the expansion, through
the merger, of the Netcare group’s association with primary care providers,
that is, with general practitioners, who are the key conduit through which
patients are referred to specialists. The overwhelming proportion of South
African specialists are associated with one or other of the three large
national private hospital groupings –indeed it appears that the private
consulting rooms of a large number of South African specialists are
typically located in a private hospital belonging to one of the three
groups.114 The concerns of the Commission and the witnesses are rooted in
the allegation that the referral practices of the practitioners associated with
the Netcare group will reflect the interest of the group’s core investment –
the Netcare group will reflect the interest of the group’s core investment –
that is, its network of hospitals – rather than the interests of consumers or of
those who fund the consumption of hospital services. By the same token, it
is alleged that the resulting distortion in referral patterns will favour the
Netcare hospital group at the expense of its rivals.
214.There can be little doubt that vertical integration lies at the heart of the
Netcare group’s competitive strategy. Equally there can be little doubt that the
core objective of this strategy of vertical integration lies in its putative ability
to influence referral patterns in Netcare’s favour. 115 Thus, it is plain to see
that the hospital group has literally surrounded itself with the key platforms of
hospital referral – an ambulance service, a pathology service, a dialysis unit
and primary care services. With respect to primary care services – the area
that we are called upon to examine in this transaction – it is important to recall
that it is not only through Netcare’s control of the Medicross clinic network
114 Transcript page 306
115 When, in 2001, Netcare evaluated the acquisition of Medicross it cited the ‘the
potential for increased referrals through increased general practitioner support’ as one
of the key strategic reasons in favour of the transaction. See Information provided by
merging parties in response to Tribunal’s request of 13 July 2005. In the present
transaction certainly Prime Cure presented the gatekeeping role of the primary care
providers as a key selling point.
56
that its relationship with primary care providers is effected. It is also
cemented through Netpartner and Netdirect. Note, as already elaborated, that
Netpartner is controlled as to 48% by Netcare and as to 52% by some 9000
healthcare practitioners, the majority of whom are primary care providers.
Netpartner is, at 17.5% (at the time of the merger hearings), the largest single
shareholder of the listed entity, Netcare. Netpartner wholly controls Netdirect,
which offers full risk capitation products including a network of primary care
providers who service the primary care component of the full risk product. It
is to be expected that many of the 800 members of the Netdirect network of
primary care providers are to be counted amongst the 9000 medical
practitioners who hold equity in Netpartner and, through Netpartner, in
Netcare itself.
215.The attempt by Mr. Dorfling to cast these arrangements with medical
practitioners as nothing more than a goodwillbuilding strategy is thoroughly
unpersuasive. There are constant references throughout the relevant parts of
the record to the ‘gatekeeper’ role played by medical practitioners. 116
Netcare itself, and specifically in relation to this transaction, reckons its
potential gains by reference to the positive impact that it will have on
referrals.117 There is little doubt then that Netcare is not merely concerned to
befriend the gatekeeper; it is concerned to align the interests of the gatekeeper
with those of the Netcare hospital group. We have to satisfy ourselves, firstly
that the gatekeeper is an effective gatekeeper – that is, can the gatekeeper
determine the identity of those who pass through the gate. And, secondly,
even if the gatekeeper is effective, we must ask ourselves whether this is likely
to substantially lessen competition.
to substantially lessen competition.
216.The interests and concerns of the industry players – Davidson and Strauss –
who professed concern at the vertical dimensions of this transaction are
reasonably clear. Mr. Davidson, who represents a competing hospital group, is
concerned that the Netcarealigned practitioners will favour the Netcare
hospitals and this in one of three ways. First he fears that these practitioners
will ‘underrefer’ patients to Netcare’s competitors or, conversely, that they
will ‘overrefer’ to Netcare hospitals. In a context where all the private
hospitals experience significant excess capacity this is, of course, a serious
concern for Netcare’s competitors. Secondly, he fears that, in certain
circumstances, the Netcarealigned practitioners will overrefer to Netcare’s
competitors or, conversely, underrefer to Netcare hospitals. Thirdly, and a
variant of the second concern, he is concerned that the Netcarealigned
practitioners may selectively refer as between the three competing hospital
groups so as to favour the commercial interests of Netcare.
217.These concerns which, on the face of it, appear mutually exclusive, do indeed
arise under different incentive regimes. Firstly, where a patient is on a full
116 Transcript pages 388, 602, 613
117 See Netcare Internal Discussion Document, dated 4 June 2001, titled “Very Cross or Very Happy”
at page 6. See also Netcare Memorandum, dated 1 August 2001, titled “Medicross Acquisition –
Salient Features Motivation and Rationale” at page 2.
57
feeforservice option, that is, where each engagement with a primary,
secondary and tertiary provider is covered by medical insurance (albeit subject
to managed care interventions such as hospital preauthorisation), then, while
the incentive of each provider is to retain the patient as long as is feasible (that
is to ‘overtreat’ at each stage), once a referral is medically indicated, then the
incentive of the referring primary care practitioner is to refer the patient to an
allied secondary and tertiary provider. In this incentive regime, the secondary
and tertiary provider will be happy to accept this referral because the medical
insurance cover of the patient fully covers both of these treatment stages.
218.Secondly, however, where a patient is on a capitated primary care option but
has feeforservice cover at the secondary and tertiary stages, the incentives
shift significantly at the primary stage of treatment but remain the same at the
secondary and tertiary stages. The primary care provider is incentivised to
refer the patient as soon as possible – to ‘undertreat’ but the incentives of
the secondary and tertiary providers remain as outlined in the previous
paragraph. That is to say, the secondary and tertiary providers are pleased to
accept the patient because her treatment at this stage is fully insured by her
feeforservice cover at these stages.
219.In the two incentive regimes described above, a primary practitioner aligned
to a secondary and tertiary provider will be incentivised to support his ally.
That is, he will refer his patient to his allied secondary and tertiary provider.
The second of the regimes described – that capitated primary care and feefor
service secondary and tertiary care – is particularly attractive to the allied
secondary and tertiary providers because the primary care provider is
secondary and tertiary providers because the primary care provider is
incentivised to ‘undertreat’ or, what is the same thing, ‘overrefer’.
220.The third incentive regime is where the patient is on ‘fullrisk capitation’, that
is where each provider – primary, secondary and tertiary – is capitated. In this
regime, each provider is incentivised to undertreat, that is, overrefer.
Accordingly here, a primary care provider allied to a secondary and tertiary
provider is incentivised both to undertreat and to (over) refer to the
competitors of his allies at the secondary and tertiary stages. As already
indicated, Mr. Davidson expressed concern that under this regime – as he put
it, a regime where the primary care provider is in ‘unfriendly’ hands – the
primary care provider may well finetune his referrals and refer treatment that
would not exceed the capitation fee to his allies with the costly treatments that
exceed the capitation fees going to the competitors. As caricatured in the
hearing, a Netcarealigned primary care practitioner may refer the
appendectomies to his allies and the liver transplants and hip replacements to
his allies’ competitors.
221.Mr. Davidson was unable to produce evidence that suggested that this sort –
or, for that matter, any sort of distortion in referral patterns actually occurred.
Where the regimes that involve capitation are concerned, this may be because
capitation is in its infancy and the evidence has not started to come through.
And of course it may be because primary care providers are neither willing nor
58
able to distort referral patterns in this way.
222.Mr. Strauss of Discovery did testify that Discovery’s data suggests that
referrals from Netcarerelated platforms exceeded the hospital group’s market
share.118 While the merging parties did not put up alternative evidence they
argued that a range of ethical, practical and contractual considerations severely
limit the ability to influence GP referral patterns.
223.The parties made much of the argument that ethical considerations would
limit the extent to which primary care providers responded to incentives
designed to influence their referral patterns. These considerations militate
against undertreatment in general as well as against a referral pattern that
privileged, for commercial gain, referral in favour of a particular secondary or
tertiary provider. Mr Dorfling testified that in terms of practice protocol laid
down by the Health Professions Council of South Africa, (“HPCSA”) – he
referred to a policy document on undesirable business practices and to a policy
statement pertaining to perverse incentives and related matters if a
practitioner is found guilty of any form of channelling, he would lose his
licence to practice medicine. Dorfling also testified that a Netcare committee,
chaired by the Chairman of Netcare, closely monitored inappropriate referral
patterns that may arise through the operation, in the Netcare group, of these
incentives.
224.The material incentive for a primary care provider to favour Netcare hospitals
appears to reside in the indirect shareholding that a large number of primary
care providers hold, through Netpartner, in Netcare. The merging parties
demonstrated that the size of the effective incentive to the doctors was
insignificant.119
225.Nor, argued the parties, were the material gains to Netcare hospitals of much
225.Nor, argued the parties, were the material gains to Netcare hospitals of much
consequence. The parties tracked referrals from Prime Cure clinics to hospitals
and showed that any impact on referral rates arising from the absorption of
Prime Cure into the Netcare stable would be insignificant. 120 They argued
that referrals from Medicross did not reveal a pattern that favoured Netcare.
Under Discovery’s Foundation plan, where there was an open network of
hospitals, the referral to Netcare hospitals approximately equated to the
hospital group’s overall market share. 121
226.There are, of course, contractual arrangements that directly require referral to
preselected tertiary providers. These are the socalled preferred provider
options in terms of which referrals to the designated tertiary providers are
mandated by the scheme. However Mr. Dorling argued that the preferred
118 Transcript page 201
119 Transcript pages 1020, 1028
120 Transcript page 705
121 Prior to 2005 Medicross, had a contract with Discovery for primary care capitation only, with an
open network of hospitals on the tertiary level. It was called the Discovery Foundation Plan.
Transcript page 1025
59
provider arrangement would have to be registered with the Registrar of
Medical Schemes. If the Registrar held that patients were prejudiced by the
preferred provider arrangement, he could refuse to register it. 122
227.The parties also insisted that it is very difficult change the referral patterns of
general practitioners. They pointed out that most referrals to tertiary care
facilities were made by specialists and that the pattern of GP to specialist
referral is governed by a number of highly idiosyncratic factors, mostly of a
personal nature – university and other social ties featured strongly amongst
these factors. 123 Accordingly, insisted Mr. Dorfling, the only effective way to
alter an entrenched referral pattern was through an enforceable scheme rule.
228.The parties claimed that a number of prosaic but important issues
overwhelmingly determined referral patterns, for example proximity, the more
so in a income community where patients and their families are sensitive to
transport costs and distances. Therefore, if patients are in close proximity to a
Life or Mediclinic hospital, they will be referred there, regardless of whether
the primary care provider belongs to a network owned by the Netcare group.
229.Finally, the merging parties argued that skewed referral patterns would be
detected and disciplined by medical schemes, the more so if Netcare hospitals
were – as the disputed evidence submitted by Mr. Strauss purported to
demonstrate – more expensive than the other hospital groups. 124
230.In short, the Commission did not present evidence that established that the
vertical integration that characterises Netcare distorted referral patterns,
although Mr. Strauss did present disputed evidence to this effect. Ironically
there can be little doubt that Netcare’s intent in building its vertically
integrated structure is precisely to influence referral patterns. However, there
is little evidence that this has succeeded. We concede that it is possible that
even influencing referral patterns at the margins may, in a business where risk
and return are so finely balanced, wreak considerable harm on competitors,
and it is clearly these calculations that underpinned Mr. Davidson’s concerns
that are outlined above.
231.However, the evidence does not justify this conclusion and, even if it did, it is
not clear that this would amount to a substantial lessening of competition. The
Commission conceded that there was no evidence or even likelihood of
foreclosure.125 While foreclosure may not be the only mechanism whereby a
vertical merger threatens competition, it is the most common and, in the
absence of an alternative theory and supporting evidence, we conclude that
there is no evidence that the vertical dimensions of this merger will give rise to
a substantial lessening of competition.
122 Transcript page 1012
123 Transcript page
124 Transcript page 1271
125 Transcript pages 7, 389, 470
60
Efficiencies
232.Having found that the merger is likely to substantially lessen competition in
the relevant market, we are required, in terms of Section 12A(1)(a)(i) to assess
whether it will result in any technological, or other procompetitive gains of a
magnitude sufficient to offset the lessening of competition. Note that the Act
specifies that we should only have regard to those efficiencies that, but for the
merger, would not have occurred.
233.In their competitiveness report and through their expert’s testimony, the
merging parties claim various efficiencies. In their closing arguments
however, they did not appear to rely very heavily on the efficiencies claimed.
The parties’ heads of argument make scant reference to efficiencies. We are
not sure whether they have abandoned these claims but deal with them here
anyway.
234.Dr. Stillman, the expert for the merging parties gave evidence of the
perceived efficiency gains that would accrue from the merger. 126 The
efficiency arguments are, as noted above, predicated on the claim that the
combined assets of the merging parties will better support the development of
lowcost options in a market that will soon witness a significant boost to
demand. The merger with Prime Cure will give Medicross access to Prime
Cure’s experience in providing primary care to the lowincome market, and
ensure that Prime Cure is assisted by the financial resources of Medicross. The
merging parties claim that Netcare’s financial resources will strengthen Prime
Cure’s balance sheet. Stillman averred that even when the medical scheme
passed risk from itself to a capitated managed care provider, it was still
obliged to concern itself with the question of the financial strength of the
provider, because if the capitated provider proves unable to meet its
provider, because if the capitated provider proves unable to meet its
obligations, residual liability remains with the scheme. Therefore, schemes are
generally more interested in a provider in a capitation arrangement if the
provider has a strong balance sheet. 127 However, Stillman himself questions
whether this is a mergerspecific efficiency – as we have already observed
Prime Cure has access to alternative sources of capital and other equity
investors, particularly if the capital market shares the merging parties
predictions regarding the growth of this market. This merger does not exhaust
potential sources of capital investment. 128
235.Stillman identified Prime Cure’s business processes for implementation of
managed care protocols as a further source of efficiency. This includes
working information technology systems and processes for micro
126 See exhibit 4, Stillman slide 27, Transcripts page 729733
127 Mr. Strauss, the Discovery witness, confirmed that residual risk remained with the medical
scheme.
128 See exhibit 4, Stillman’s slide 27 and transcript page 730
61
management of doctors that Medicross could apply to roll out its Netdirect
offering. Again Stillman acknowledges that these are facilities that Medicross
itself could purchase and implement but claims that it would take between 18
and 36 months to have these fully operational. 129 Contrary to Stillman’s
stated view we do not view these efficiencies as merger specific.
236.The third category of efficiency Stillman lists is savings in infrastructure cost.
Stillman indicates that the merger would allow the merging parties to save
some R6R8 million per year through the elimination of duplicate activities in
respect of human resource staff, call centre and finance staff, as well as other
infrastructure.130 However, Stillman conceded that these figures were not
independently analysed by him but were derived from savings estimates given
to him by the merging parties. 131 Note however that a particularly
comprehensive due diligence prepared by the acquiring company concluded
that are no “backoffice” savings to be gleaned from this merger. The LCIM,
prepared on behalf of the target company and which has been described as a
‘selling document’, supports this conclusion. A series of retrenchments
effected in recent months led to the conclusion that excess labour costs had
already been squeezed out of the target company and that there is very limited
room for further cost reduction. The due diligence report and LCIM
memorandum from Prime Cure both say that there are very limited back office
savings.132
“We are of the opinion that very little additional overhead synergies can be
extracted as a massive restructuring involving 125 employees have taken
place in the last 18 months.”
237.The commission observed that the alleged infrastructure savings of R6 to R8
million are unsubstantiated and also drew our attention to the conclusions in
million are unsubstantiated and also drew our attention to the conclusions in
Prime Cure’s limited confidential information memorandum regarding
efficiencies.
238.Dr. Stillman nevertheless insists that his discussions with Medicross reveal
that they will, in fact, eliminate duplicate back office functions. He
acknowledges that these efficiencies are not generally given much weight by
competition authorities. This is the view taken by this tribunal in an earlier
decision:
“Areeda treats plant size and plant specialization economies as those most
worthy of recognition but is more sceptical about claims for others frequently
raised which he describes as "ordinary efficiencies" e.g. distribution,
procurement and overhead economies” 133
129 See exhibit 6, page 3
130 See exhibit 6, page 3
131 Transcript page 837
132 See LCIM file C 306, Due Diligence Report File B page 351 and remarks by Hodge at transcript
page 396
133 See Trident Steel (Pty) Ltd and Dorbyl Limited – 89/LM/Oct00 paragraph 56
62
239.Stillman also mentions tax savings consequent upon the merger although he
concedes that these “are absolutely never considered a merger specific
efficiency and never recognised”. This is also congruent with the view of tax
savings and other pecuniary gains taken by us in Trident Steel :
“ Pecuniary efficiencies would not constitute real economies nor would those
that result in a mere redistribution of income from the customers, suppliers or
employees to the merged entity. Without categorically rejecting them we
would be more sceptical than the Canadian courts in accepting certain
efficiencies such as administrative efficiencies since these can be established
in most mergers .”134
240.Finally, Stillman also argued that the Prime Cure clinics would benefit from
the application of the approach taken by Medicross’ medical centre model
(that is, the clinics) if it could be applied to Prime Cure centres. It appears that
Medicross clinic facilities assemble a larger and more diverse grouping of
healthcare professionals than does Prime Cure and that this generates certain
scope and scale economies which are not generated through Prime Cure’s
more restricted utilisation of its clinics. However we are not able to evaluate
this claim on the basis of the evidence presented. He also argued that
converting Prime Cure clinics to a fuller service model would create additional
opportunities for members of the South African Medical and Dental
Practitioners, (“SAMDP”) a group of black doctors and dentists indirectly
affiliated with Medicross. 135 However, he could not adduce any more precise
evidence in support of this claim or ascribe it any economic value. 136
241.Dr. Stillman also argued that vertical integration would improve pricing
incentives. He suggested that Netcare’s participation in Netdirect’s fullrisk
offering enabled the hospital group to segment its market and to price at the
margins appropriate to each segment. Again it is difficult to see how this is a
merger specific effect – it should be equally attainable to any hospital group
that participates in any fullrisk offering. It is not immediately apparent why
the tertiary provider and the managed care provider have to be part of the same
corporate structure to achieve this efficiency. 137
242.We are not persuaded that the efficiency gains claimed outweigh the anti
competitive effects of this merger. The parties’ own expert acknowledged that
many of the efficiency claims were not mergerspecific. Certain of the claims
were contradicted in several important documents, for example in the detailed
due diligence report.
243.We therefore conclude that the efficiencies claimed do not countervail the
lessening of competition to which the transaction will give rise.
134 See Trident Steel (Pty) Ltd case at paras 55, 81
135 See exhibit 6, page 4
136 Transcript page 838
137 See transcript page 728
63
Public Interest Issues
244. These were not argued and we agree with Commission that there are no public
interest factors which would justify approval of this merger.
Order
We find that the horizontal dimensions of this merger are likely to lead to a
substantial lessening of competition in the relevant market. We also find that there
are no countervailing efficiencies or public interest considerations. We have
accordingly ordered that this merger be prohibited.
____________ 13 October 2005
D Lewis Date
Concurring: Y Carrim, L Reyburn
For the merging parties : Adv. D. Unterhalter, instructed by Webber Wentzel Bowens
Attorneys
For the Commission : Adv. D. Berger, instructed by Attorneys
64