NonConfidential version
COMPETITION TRIBUNAL OF SOUTH AFRICA
Case No: 57/LM/Aug09
In the matter between:
Santam Limited Acquiring Firm
and
Emerald Insurance Company Limited; and
Emerald Risk Transfer (Pty) Ltd Target Firms
Panel : N Manoim (Presiding Member)
A Ndoni (Tribunal Member)
A Wessels (Tribunal Member)
Heard on : 28 October 2009
Order issued on : 30 October 2009
Reasons issued on : 27 January 2010
Reasons for Decision
APPROVAL
[1] On 30 October 2009, the Competition Tribunal (“Tribunal”) approved the
acquisition by Santam Limited of Emerald Insurance Company Limited and
Emerald Risk Transfer (Pty) Ltd. The reasons for approval follow.
THE PARTIES
[2] The primary acquiring firm is Santam Limited (“Santam”). Sanlam Limited
(“Sanlam”) controls Santam. Sanlam is controlled by a number of shareholders.
[3] Two target firms are relevant to the proposed deal, namely (i) Emerald
Insurance Company Limited (“Emerald”) and (ii) Emerald Risk Transfer (Pty) Ltd
(“ERT”), trading as Emerald Underwriting Managers (“EUM”). Emerald is a
wholly owned subsidiary of Super Group Limited (“Super Group”). Super Group
1
NonConfidential version
is a public company with no single shareholder holding sufficient shares to
control it. The premerger shareholders of ERT are Super Group (38%), Gary
Steven Corke (27.75%), Dave Manuel (27.75%) and the Senior Management
Trust (6.5%).
THE TRANSACTION
[4] Santam proposes to acquire 100% of the issued share capital in Emerald from
Super Group , as well as 100% of the issued share capital of ERT, 38% from
Super Group and 62% collectively from management, i.e. Gary Steven Corke,
Dave Manuel and the Senior Management Trust. Once Emerald has been
acquired the Emerald short term insurance licence will be runoff and its
insurance book transferred to Santam, i.e. a portfolio transfer will occur into the
Santam licence. Furthermore, a n internal restructuring will be effected in terms
of which all the Santam short term corporate insurance underwriters will be
transferred to Emerald, which will continue as a wholly owned underwriting
manager of a Santam investment company, Swanvest 120 (Pty) Ltd. This
specialist corporate underwriter will operate as “Emerald underwritten by
Santam”.
BACKGROUND TO THE HEARING
Summary of view s
Competition Commission
[5] The Competition Commission (“Commission”) recommended to the Tribunal
that the proposed transaction be approved, primarily on the following grounds: (i)
the likely failure of Emerald and the resultant effect thereof on the market (see
paragraphs 50 and 51 as well as 56 to 65 below); and (ii) the retention through
the proposed deal of a significant portion of the underwriting capacity in the
South African short term corporate insurance sector as well as the innovative
underwriting method/strategy of Emerald/EUM, which would benefit consumers
underwriting method/strategy of Emerald/EUM, which would benefit consumers
(i.e. corporate clients). However, the Commission also concluded that the
proposed deal would result in the removal of a maverick Emerald/EUM as an
effective competitor in the said sector. The Commission identified no competition
2
NonConfidential version
concerns regarding coordinated conduct in any relevant market as a result of
this proposed transaction.
Third parties
[6] Certain third parties, i.e. competitors, brokers and their clients, expressed
concerns regarding the effects of the proposed transaction on market capacity
and competition between progressively fewer significant players in the short
term corporate insurance sector. The issues raised include concerns that the
proposed deal would (i) reduce capacity in the short term corporate insurance
sector because reinsurers would not extend reinsurance to a total equal to the
sum of the separate entities’ reinsurance; (ii) limit capacity/appetite for certain
business, for example retail; and (iii) remove from the market a cooperative
(flexible) and innovative competitor who is solution driven and willing to take on
larger risk exposure.
[7] On the other hand , other third parties foresaw no competition concerns
resulting from the proposed deal or were of the opinion that the merger was pro
competitive given inter alia that (i) it would move Emerald to a more sustainable
business model with greater emphasis on risk management; (ii) a union with
Santam would give Emerald access to sufficient solvency capital to write more
risk; and (iii) it would provide greater financial security for corporate clients given
the Super Group difficulties and the impact of this on Emerald.
[8] The Tribunal invited the brokering firms Marsh and Glenrand MIB, as well as
Shoprite Checkers, to partake in the hearing of this matter; they however
declined this invitation.
Witnesses
[9] The Commission did not call any witnesses to testify at the hearing. At the
request of the Tribunal, the following representatives of the merging parties gave
evidence at the hearing:
• Mr. Gary Steven Corke (“Corke”), the managing director of Emerald and
• Mr. Gary Steven Corke (“Corke”), the managing director of Emerald and
CEO of ERT; and
• Mr. Quinton Matthew (“Matthew”), Santam’s head of specialist business.
3
NonConfidential version
RATIONALE FOR THE TRANSACTION
[10] As rationale for the proposed deal , Super Group submits that it is refinancing
its business and selling its noncore assets and Santam submits that it will allow
it to acquire Emerald’s insurance book and ERT’s skilled underwriters. Santam
further submits that this is aimed at improving its short term corporate insurance
business by adopting a new business/underwriting approach which incorporates
the best of the ERT and Santam’s core skills and philosophies.
[11] A review of Santam’s strategic documentation and the testimony of Matthew
make it quite clear that Santam wishes to revitalise its approach to short term
corporate insurance by adopting the “innovative” underwriting manager models
and (reinsurance) strategies of Emerald. Matthew gave a rather grim account of
Santam’s short term corporate insurance results over the past three years and
also confirmed an underwriting loss in the current year to date. He testified that
Santam needs to “ change the business model in terms of the flexibility, the
solution orientation, the way that we [Santam] conduct our business, the use of
our insurance markets ...”; that “... going forward we [Santam] marry the
philosophy of Santam with the entrepreneurship of Emerald to our future
business”; that “ the structure of the Emerald basis of working under EUM as an
underwriting manager is one that we [Santam] are looking to continue into the
future”; and that “ we [Santam] ... look to sustain the model and the business as
EUM have at present ”.
[12] All of this signifies a substantial post merger preservation of the current
Emerald/EUM approach to short term corporate insurance and thus alleviates
the concerns raised by certain third parties that the more innovative and
entrepreneurial approach of Emerald/EUM would necessarily be lost post
entrepreneurial approach of Emerald/EUM would necessarily be lost post
merger (see paragraph 6 above).
THE RELEVANT MARKET (S)
Overlapping activities
4
NonConfidential version
[13] Santam is a diversified general short term insurance provider with product
offerings and services in the corporate, commercial and personal business
segments, providing a range of insurance products to individuals, small
businesses and corporate clients. Emerald’s focus conversely is much narrower:
its product offerings and services relate to short term corporate insurance,
primarily the corporate property and engineering spheres, and as such is
classified as a socalled “monoline” insurer.
[14] ERT (t/a EUM) is an underwriting management company that fulfils inter alia
the function of internal infrastructure of insurance underwriting, claims handling
and accounting in terms of regulatory requirements. It has two main business
streams: (i) an underwriting manager business (providing underwriting
management services to Emerald and its cell captives exclusively); and (ii) the
underwriting of risk for its own account in cell captives. Regarding these
underwriting management services, EUM and Santam’s inhouse corporate
underwriting managers premerger exclusively provide underwriting services to
the Emerald and Santam groups respectively. Post merger, the new
underwriting unit will only provide services to Santam.
[15] Therefore, the overlap between the activities of the merging parties is in
respect of short term corporate insurance.
Background to short term corporate insurance
Sector regulation
[16] Short term insurers are regulated in South Africa by the Financial Services
Board (“FSB”) under the ShortTerm Insurance Act, 1998 (Act No. 53 of 1998)
(the “STIA”). In terms of the STIA, a short term insurer is inter alia required to be
registered locally with the FSB, and to maintain its business in a financially
sound condition (through having assets, providing for liabilities and in the
manner it conducts its business). More specifically, a short term insurer must
maintain a minimum solvency level of 15%, i.e. its net assets as a percentage of
its net premium written in the preceding 12 month period must be at least 15% at
all times (also see paragraph 58 below).
5
NonConfidential version
Corporate insurance structures
[17] Given the magnitude of corporate risks, no individual insurer wishes to be
overexposed to the risks of any one client or any one sector. Individual insurers
may also not have sufficient capacity (themselves or because of treaty
restrictions, see paragraph 20 below) to insure a client fully. For this reason,
insurers in practice share risks on the asset and other insurance programmes of
larger corporate clients. This spreading of risk is also a consideration for the
insured who may wish to spread its risk amongst a number of insurers. As a
result, short term corporate insurance is structured in a number of ways, the
most relevant of which are (i) coinsurance contracts, which are the most
prominent structure in practice; and (ii) layered programmes. 1
(i) Coinsurance contracts
In t hese contracts each insurer takes a specified percentage of the risk of
loss on a specific insurance programme so that the overall risk is spread
amongst several insurers. In practice this results in socalled “lead” and
“follow” positions taken by the individual insurers:
(a) Lead insurer positions
Typically the lead insurer would take the largest portion of the risk, perform
the risk assessment, set the premium rates, negotiate the terms of the
contract, handle any loss negotiations and decide on any settlement in the
event of loss occurring. This lead insurer would typically earn a 2.5%
coinsurance fee for performing these tasks (also see paragraph 92 below).
(b) Follow insurer positions
Follow insurers would take up the remainder of the risk not taken by the
lead insurer. These follow insurers generally take smaller percentages of
the risk and follow the premium rates and terms negotiated by the lead and
the risk and follow the premium rates and terms negotiated by the lead and
1 Other structures include separate contracts. However, this will not be discussed since it
does not appear to be common, given that it is undesirable, costly and cumbersome from a
client perspective to deal separately with each insurer.
6
NonConfidential version
also accept the settlements negotiated by the lead insurer (also see
paragraphs 91 and 92 below).
(ii) Layered programmes
Layered programmes are similar in their intent to coinsurance contracts. In
this case the asset protection or other programme would be layered into
primary and secondary risk layers with insurers taking a percentage of the
risk in each layer. The primary layers would be those that would take the
initial losses on the happening of any risk event with the secondary layers
taking losses above the thresholds set for the layer.
Reinsurance as secondary market
[18] Given the relative risk magnitude and local market capacity, most South
African insurers of corporate clients will utilise a reinsurer to reduce their own
risk. An insurer’s underwriting capacity is largely driven by the reinsurance
capacity granted to it by reinsurers. Consequently, a large portion of the cost of
corporate insurance in the primary markets can be attributed to the cost and
terms of reinsurance in the secondary market.
[19] Most of the larger reinsurers , whether registered in South Africa or not, are
global/international reinsurers. The Commission’s market enquiry confirmed that
the cost of reinsurance is driven primarily by international factors that often are
entirely unrelated to local claims and/or conditions. Reinsurance pricing is mostly
affected by (natural) disasters and other major events, for example hurricane
damage in America; floods in Europe; and mining disasters in South America.
The 11 September terrorist attacks in America for example had a significant
effect on the availability and price of reinsurance. Matthew indicated during his
testimony that rates and premiums increased by as much as 50% to 70% as a
result of withdrawn capacity following this event.
result of withdrawn capacity following this event.
[20] South African insurers of corporate clients use reinsurance extensively and
this reinsurance is of two basic types (of which a combination is mostly utilised)
viz:
7
NonConfidential version
(i) Facultative reinsurance
This is reinsurance acquired for a specific risk. The reinsurer assesses a
particular risk and quotes a price for reinsuring that specific risk, for example
a hotel in the centre of Johannesburg, an aircraft hanger at an airport or a
specific manufacturing plant.
(ii) Treaty reinsurance
This reinsurance is provided in terms of a reinsurance treaty under which the
reinsurer obliges itself to reinsure any risks written by the primary insurer up
to certain limits and under certain conditions. Given the greater scope of
treaty reinsurance and the fact that the reinsurer does not assess the risk
itself, the reinsurer will generally want to be satisfied of the skills of the
corporate underwriters and the underwriting philosophies of the insurers to
whom they grant treaties. This factor reinforces the need to have skilled
underwriters with a satisfactory industry track record (also see entry barriers
discussed in paragraph 41 below). Furthermore, for the same reasons,
treaties will often impose restrictions and conditions on the risks for which
reinsurance will be provided, for example (i) treaties often restrict the
geographical area within which an insurer may underwrite risks (for example
South African insurers may be restricted to underwriting South African and
perhaps subSaharan African risks); and (ii) certain higher risk industries may
be excluded from the treaty or may have to be specifically referred to the
reinsurer and accepted before cover is extended.
[21] In the context of the instant transaction, given the issue of an alleged failing
firm (see paragraphs 56 to 65 below), the implications of the use of socalled
FSB “nonapproved” foreign reinsurers are particularly relevant. FSB “approved”
FSB “nonapproved” foreign reinsurers are particularly relevant. FSB “approved”
reinsurers inter alia have an office and a bank account in South Africa. It is
noted that the use of foreign reinsurers not registered in South Africa although
permissible, places a strain on the FSB required solvency/capital adequacy of
insurers. Local insurers have to keep reserves for liability, but the said foreign
reinsurance is not taken into account as part of the capital of the insurer in the
way approved reinsurance would be. Consequently, an insurer would need to
8
NonConfidential version
have the requisite capital to cover this liability, even though this liability has been
passed to the reinsurer (also see paragraphs 60 and 61 below).
Pricing and current market conditions
[22] Insurers contacted during the Commission’s market investigation
emphasized the cyclical nature of pricing in the short term corporate insurance
sector, with hard phases (i.e. higher premiums and less consumer bargaining
power) and soft phases (i.e. lower premiums and more customer bargaining
power). Given the current credit crunch and significant losses suffered in
2007/2008 (fires), the market is generally seen as being a hard one or moving
into a hard market cycle. Matthew confirmed the latter general view during his
testimony, but pointed out that the cycle can vary according to market segment,
for example mining or retail.
Relevant product market (s)
Short term corporate insurance
[23] Short term corporate insurance involves insurance for large South African
and multinational firms on broker negotiated terms customised for the client and
its specific and complex risks. Despite varying internal definitions 2 of corporate
insurance, competitors and customers (including brokers) interviewed by the
Commission agree that short term corporate insurance is a distinct market from
the other two main short term insurance segments, namely the commercial 3 and
personal4 segments, due to certain core differentiating factors. The Tribunal
concurs with the Commission’s finding of a separate relevant market for short
term corporate insurance, which is distinct from the secondary market for
reinsurance, based inter alia on the following factors:
(i) Customised contracts
Contracts are unique in the sense that they have and require customized and
specialized wording tailored to the particular circumstances and risks of a
2 Based on, for example, the size of the risk, corporate characteristics, turnovers, premiums,
customisation of terms, or various combinations of these factors.
3 Insurance for small businesses/factories, run by individuals, partnerships, close corporations
and small corporates on standard term insurer policies.
4 Insurance for individuals and their property/risks on standard term insurer policies.
9
NonConfidential version
specific corporate client. The wording is not standard insurer policy wording
as in the personal and commercial segments, but generated by corporate
brokers and negotiated with insurers.
(ii) Scope of r isk evaluation
The depth and intensity of the risk assessment and the skills required for this
is significantly greater than that required for the commercial and personal
segments. Insurers of corporate clients may for example use engineers and
technical staff to assess the client’s premises for risk.
(iii) Underwriting skills requirement
There is a relative difficulty in rating and assessing the risks of a large
corporate, given the size, gravity and complexity of the risks. This is inter alia
a function of the clients’ geographical or international spread (different risk
environments), the complex or specialised nature of their operations (for
example retail or mining) and the variety of their activities. As such, a short
term corporate insurer requires highly skilled corporate underwriters, which is
largely acquired through many years of experience rather than through
academic qualifications alone. The Commission furthermore found that these
skills are in critical short supply in South Africa.
(iv) Risk size and severity
The sheer size and severity of short term corporate risks necessitate:
(a) a sharing of risk amongst insurers through coinsurance contracts and
layered programmes (see paragraph 17 above);
(b) favourable balance sheet and solvency requirements of insurers
(smaller players may have to rely extensively on reinsurance) (also see
paragraphs 18 to 21 above and paragraph 41 below); and
(c) favourable credit ratings, for example ‘triple A’ ratings of insurers by
rating agencies (also see paragraph 41 below).
Distinction between lead and follow short term corporate insurers
Distinction between lead and follow short term corporate insurers
[24] The Commission in its recommendation to the Tribunal states that a case
might be made for a further delineation of the short term corporate insurance
10
NonConfidential version
market into (i) lead and (ii) follow markets, but it does not conclude on this issue.
The Commission states that given the lack of fully satisfactory evidence on
whether Emerald could be classified as a usual lead market player, it has not
pursued this issue.
[25] The Tribunal is of the view that the Commission in the context of this
transaction ought to have pursued and concluded on whether or not the lead
market is a distinct relevant market for competition purposes (also see
paragraphs 91 to 100 below).
[26] Based on the (limited) available information it seems plausible that the lead
market could be a separate relevant market for competition purposes. It is clear
that the prevalence of coinsurance contacts and layered programmes
significantly impact the structure of the market. The available evidence also
unambiguously shows that certain players participate primarily in the lead
market as opposed to the follow markets, or vice versa. Furthermore, the
available information points to significantly larger entry barriers into a potential
lead market compared to a follow market; any potential new entry is likely to be
limited to follow positions (see paragraph 43 below). However, the typical
players in a potential follow market do occasionally take lead positions. From the
limited available information it is unfortunately impossible to pinpoint the
prevailing circumstances under which the latter could or could not occur (also
see paragraphs 91 to 100 below).
[27] Given the fact that it cannot be concluded , based on the limited available
information, that the lead and follow markets are not distinct relevant markets,
this matter will be assessed on the basis that a distinction may be drawn
between the likely impact of the proposed deal on potential lead and follow
markets (see paragraphs 88 to 100 below).
Relevant geographic market
[28] From a geographic market definition perspective t he Commission, based on
its market investigation, makes a distinction between short term corporate
11
NonConfidential version
insurance to South African corporate clients (including perhaps Africa
operations) and megacorporates/multinationals. The Commission concludes
that the scope of the relevant geographic market for the former group of
corporate clients is national, and international in relation to the latter group and
possibly in relation to certain very specialized risks areas, for example aviation.
[29] The Commission summarises the factors that are indicative of a national
market for short term corporate insurance to South African corporates as
follows: (i) legislative barriers 5; (ii) more competitive premiums locally; (iii) the
need for higher premium volumes to interest offshore insurers; (iv) higher
deductibles offshore; (v) offshore insurers’ preference for participating in
secondary risk layers; (vi) higher transactional costs offshore; and (vii) the lack
of local/African knowledge and easier settlement of claims locally.
[30] We shall analyse the transaction on the narrowest possible geographic
market definition, i.e. at national level.
COMPETITION ANALYSIS
Market participants and shares
[31] Premerger there are nine active participants in the South African short term
corporate insurance market, namely ACE, AIG, Allianz, Emerald, Etana, Lion of
Africa, Mutual & Federal (“M&F”), Santam and Zurich. ABSA and RMB are either
very recent entrants or potential new entrants (see paragraph 43 below).
[32] Since the FSB 6 does not have accurate information on short term corporate
insurance as a separate category, the Commission obtained Gross Premium
Written7 (GPW) information for short term corporate insurance from the
5 South African risks cannot be insured offshore with nonregistered insurers, unless it can be
shown that (a) local insurers do not have the capacity to take the risk; or (b) better premiums
and terms can be obtained offshore.
6 For the purposes of the legislative scheme of the STIA, insurance has been divided into
eight statutory categories viz: (i) accident and health; (ii) engineering; (iii) guarantee; (iv)
liability; (v) miscellaneous; (vi) motor; (vii) property; and (viii) transportation. Statistical
information is provided to the FSB by insurers in terms of these categories.
7 GPW is a generally accepted industry benchmark amongst insurers.
12
NonConfidential version
individual insurers. 8 (Note that no market share information is available for the
lead and follow markets as potential distinct relevant markets.)
Table 1 Market shares for 2008 for a South African market for short term
corporate insurance
Market participant Market share
Santam [1020]
Emerald [1020]
Combined entity [3040]
Mutual & Federal (M&F) [2030]
AIG [1020]
Allianz [010]
Zurich [010]
Lion of Africa [010]
ACE [010]
Etana [010]
Source: Confidential information submitted by each participant, based on GPW.
[33] If market share information is analysed over a three year period (i.e. from
2006 to 2008), Santam’s market share decreases very significantly (also see
paragraph 11 above), whilst the market shares of specifically Emerald and ACE
increase significantly.
[34] Regarding Santam’s loss of market share Matthew testified that of some
R250 R300 million worth of Santam lost business, less than R30 R40 million
went to Emerald, and the balance to the other market participants, specifically
AIG, ACE and M&F. Matthew attributed this lost business mainly to Santam’s
stringent risk management requirements (for example insisting on the client
installing a sprinkler system and putting fire extinguishers in certain parts of a
building), rather than to price or deductibles. According to Matthew, Santam’s
learning is that clients need to budget for some of these requirements which
require longer lead times to implement them. To address this issue going
forward, Santam may take smaller shares in riskier industries and/or structure a
reinsurance program around it, according to Matthew.
8 As noted in footnote 2 above, i nsurers have differing definitions of corporate business and,
therefore, the indicated market shares may not be entirely accurate reflections of relative
market shares.
13
NonConfidential version
[35] Zurich and Etana are relatively new entrants and thus no information is
available pre 2008. It is noted that Zurich has attained a significant market share
in a very short period (i.e. nine months of trade in 2008). There is no reason to
doubt that it would maintain this market position in future.
[36] In its recommendation to the Tribunal t he Commission accepts that the
market share of the merged entity will post merger significantly decrease as a
result of a reduced book. The Commission in this regard relies on a Santam
board briefing note proposing the merger to the Santam Board 9 which indicates
that Santam plans to reduce the Emerald book by [...]%, which according to the
Commission would reduce the merged entity’s combined market share by circa
[010]%. The Commission also avers that the combined entity’s underwriting
capacity (and therefore GPW) would in any event not equal that of a separate
Emerald and Santam due to a drop in reinsurance that reinsurers would extend
to the merged entity.
[37] Not only is the Tribunal highly sceptical of this predicted post merger decline
in the merged entity’s market share, but also notes that it is a short term point of
view which is inappropriate in a merger context. There is no reason not to
believe that the merged entity’s market share may even increase post merger,
depending inter alia on the effectiveness of the proposed merger’s
implementation.
[38] In regard to shared accounts, Matthew testified that “ we [Santam and
Emerald] have a pretty low clash in terms of common accounts that are going to
impact that downscaling of capacity ”. The accounts that Santam and Emerald
have a common line on are limited to circa 15 accounts, according to Matthew.
This was corroborated by Corke who confirmed that Emerald and Santam in fact
This was corroborated by Corke who confirmed that Emerald and Santam in fact
have only 13 common clients between them (out of a total of circa [...] Emerald
clients). Corke attributed this relatively small number of common accounts to
relative differences between Emerald and Santam in their risk selection criteria
and the manner in which they reinsure. This very limited overlap in terms of
shared accounts does not support aversions of a significant post merger decline
in the merged entity’s market share.
9 Board Meeting of 27 May 2009.
14
NonConfidential version
[39] Furthermore, Matthew affirmed that the merged entity would look to make up
any lost market share as a result of shared business and practices around risk
management by utilising facultative reinsurance, to the extent that it is
obtainable. In regard to reinsurance, t he South African Insurance Association
(SAIA) expressed the view that the merged entity may because of its larger size
post merger have the ability to negotiate better pricing for reinsurance. The
merging parties further submit that a Santam/Emerald union would give Emerald
access to sufficient solvency capital “ to write more risk ”. This defies any factual
basis for any assumed significant post merger decline in the merged entity’s
market share.
[40] From a geographic market perspective, i t is noted that the market share of
the merged entity (as shown in Table 1 above) would dilute very significantly if
short term corporate insurance placed offshore is considered. The merging
parties estimate that approximately 40% of this insurance is placed offshore,
inter alia through the use by larger corporates of cell captives placed globally.
New e ntry and entry barriers
[41] According to the Commission’s market investigation th e barriers to entry into
the short term corporate insurance market are significant and can be
summarised as:
(i) existing l egal (sector regulatory) barriers and proposed new FSB
solvency criteria which may place even further pressure on the smaller
players;
(ii) considering the size and severity of corporate risks, sufficiently large
capital in order to maintain solvency and financial stability levels;
(iii) large corporate clients’ requirement that an insurer should have a high
rating level, i.e. an ‘AAA’ or ‘AA’ rating. Reinsurers would also have an
rating level, i.e. an ‘AAA’ or ‘AA’ rating. Reinsurers would also have an
interest in an insurer’s rating insofar as treaty capacity is extended to
an insurer;
(iv) critical mass to obtain a sufficiently large and diversified portfolio
across various sectors, countries and clients to spread out risks and
15
NonConfidential version
generate sufficient premium income to offset losses. Mass affords
negotiating power for reinsurance;
(v) a reputation for specialized corporate underwriting skills (which are in
scarce supply), including risk assessment and management skills (for
example technical engineering skills to evaluate client risks), as well
as a track record that would satisfy brokers that business can be
safely entrusted to them, i.e. “broker trust”. Matthew in regard to skills
and capacity testified that the “ ability of new players to enter the
market is certainly one that is driven by the perception of having the
right underwriting skill and being able to access capacity, be that from
your parent company or your parent underwriter ”; and
(vi) the ability to obtain competitively priced reinsurance . Extending treaty
reinsurance carries risks for the reinsurer and they would generally,
particularly in the anticipated hard market conditions, need to be
satisfied of the appropriateness of the insurer’s underwriting
philosophy, their approach to risk management and most importantly
the skills and reputation of those who will write to the treaty (which
further reinforces the abovementioned skill requirement).
[42] However, d espite the apparent high barriers to entry at least two new firms
have entered the short term corporate insurance market, namely Zurich and
Etana (part of Hollard Insurance). Matthew also cited ACE as a new entrant in
the past three years. The Commission however warned that current market
conditions and future capital adequacy requirements may depress future entry.
SAIA indicated to the Commission that it sees the future prospects of smaller
insurers as very tough (given new solvency regimes and difficulties to access
capital). The Commission also pointed out that a number of large international
capital). The Commission also pointed out that a number of large international
insurers that entered South Africa have exited relatively quickly, for example
Winterthur, St Paul and XL.
[43] On the other hand, ABSA and RMB, given their access to capital, are more
hopeful prospects of additional competition specifically in a follow market. In this
regard Matthew testified that RMB, with the support of exSwiss Rhee staff, is in
the process of setting up an underwriting business that would have “ a minimum
16
NonConfidential version
of R150 million worth of capacity as a startup”. The Commission’s market
enquiries also confirm that ABSA has been hiring corporate underwriters in the
market (including former Emerald underwriters). Matthew however conceded
that the likes of new entrants such as RMB and ABSA would be “ more
conservative and follow”. Be that as it may, both said players appear to be
(potential) new entrants that would provide additional follow capacity.
Alleged r educed capacity/loss of innovative approach
[44] The apparent key to Emerald’s perceived innovative underwriting model is
the use of a layered reinsurance programme (which makes extensive use of
layers of facultative reinsurance in the primary liability layers), very low retention
of risks for its own account and the use of treaty reinsurance only once available
facultative reinsurance has been exhausted. Thus, on any loss occurring only a
very small portion would be borne by Emerald, since most loss would be passed
on to the facultative reinsurers and the treaty reinsurance would also be
protected.
[45] Certain market participants indicated that if the removal of Emerald
eliminated its style of underwriting from the market, that this would be less than
desirable. Others again held the view that if Emerald was not transferred to
Santam that Emerald’s innovative approach would certainly be lost to the
market.
[46] As already indicated in paragraph 12 above, any concerns regarding the loss
of Emerald’s innovative approach following the merger appear to be unfounded
and this will not be discussed any further. Furthermore, Matthew testified that
even in the event of the transaction not proceeding “ we may well equally
consider other options in terms of replicating some of the learnings that we have
picked up in terms of doing business on a similar basis to that of Emerald and
hence compete in that business model going forward ”. Therefore, concerns that
Emerald’s business model cannot be replicated by other market participants
appear to be a red herring; there is no factual basis whatsoever for this
assertion.
17
NonConfidential version
[47] Regarding the issue of capacity, t he Commission concludes that if the
proposed merger was not effected (if Emerald was for example liquidated), that
this would result in a loss of corporate underwriting capacity driven by a drop in
reinsurance extended to the local market. It is alleged that reinsurers would be
cautious in extending capacity to new entrants. On the other hand, certain
market participants expressed the view that underwriting capacity will in any
event reduce post merger, because reinsurers will be unwilling to extend the
new entity the reinsurance of its constituent parts.
[48] Corke in his testimony fervently refuted the notion that a prohibited deal
would have a significant impact on market capacity. He stated “ anybody has
access to international reinsurance ”; “[w] hether a number of other entrants want
to continue to reinsure locally or overseas is a matter of their choice and even
since these talks ... one or two other competitors have already moved into our
market space and one or two of them will use overseas reinsurance. So, I don’t
see that as a barrier to this deal being accepted or not accepted. I think it is
largely irrelevant ”; “ if ... we [Emerald] went away tomorrow, Everest Tree would
come over here and market and try and give one of the new entrants some
capacity”; and “... if we were to float away tomorrow, other people would move
into that market space and take advantage of the other supply channels ”.
Matthew furthermore contended that “ one of the rationales for the transaction
also rests in combining and putting a book of business together that actually has
a more attractive combined effect for reinsurers ...”. Therefore, no factual basis
exists to conclude, either absent or with the proposed deal, that reinsurance
extended (and thus underwriting capacity of the insurers) would significantly be
impacted.
Failing firm
[49] Emerald’s current financial situation (driven by same of Super Group) and
the chain of events that have led to its current sectorregulatory predicament
represent a central theme in the Commission’s analysis of this case, as well as
during the hearing of oral evidence. In the sections below we give a detailed
account of the failing firm issue in the context of the proposed deal and the high
burden of proof required (from merging parties) to credibly invoke such claim.
18
NonConfidential version
Commission’s view
[50] Despite established Tribunal and international precedent 10 that the onus is
on the merging firms 11 to provide the evidence necessary to invoke the doctrine
of the failing firm, in this case curiously it was not the merging parties who
invoked this argument but the Commission.
[51] As stated in paragraph 5 above, the Commission recommended an approval
of the proposed transaction based inter alia on its finding that Emerald meets all
requirements of a ‘failing firm’. It is pointed out that this argument does not
extend to EUM (also see paragraph 69 below). However, information requested
by the Tribunal from the merging parties (after the Commission had submitted its
recommendation but before the hearing) revealed that the Commission had not
been informed by the merging parties of an alternative bidder for Emerald/EUM.
Consequently, the Commission (at the Tribunal hearing stage of the matter)
based on this new evidence, altered its former stance and concluded that
Emerald does not meet the failing firm test. We shall elaborate on the aspect of
an alternative purchaser in more detail below (see paragraphs 66 to 71 below).
Tribunal’s assessment
[52] The failing firm doctrine enjoys express statutory recognition in the
Competition Act, 1998 (Act No. 89 of 1998) (the “Act”). Section 12A(2)(g) of the
Act directs us to consider “ whether the business or part of the business of a
party to the merger or proposed merger has failed or is likely to fail ” as part of a
nonexhaustive list of factors that must be considered in merger assessment. As
pointed out by the Tribunal in the merger between Iscor Limited and Saldanha
Steel (Pty) Ltd the failing firm doctrine, as such, in the Act is not a ‘defence’ to a
merger that has been found on an initial market analysis to be anticompetitive.
merger that has been found on an initial market analysis to be anticompetitive.
Rather, it is recognised as one of a nonexhaustive list of factors that must be
10 See case no. 67/LM/Dec01, paragraph 110.
11 As comparison: the burden of proof for invoking the failing firm doctrine rests with the
merging parties according to the jurisprudence of inter alia the United States of America; this
is also the position adopted by the European Union in its Horizontal Merger Guidelines (2004/
C31/03).
19
NonConfidential version
taken into account before one can determine whether or not a particular merger
is likely to substantially prevent or lessen competition. 12
[53] In times of financial and economic distress, such as we are currently
experiencing, many firms could find themselves in some sort of financial difficulty
and these firms may seek to safeguard their longterm survival possibly by
merging with (healthier) competitors. The task of the competition authorities is to
assess whether the claim that a firm has failed or is likely to fail is genuine or a
contrivance to obtain approval for an otherwise anticompetitive merger.
[54] The failing firm doctrine is internationally recognised in competition law
jurisprudence and, although not applied uniformly in all jurisdictions, has
nevertheless been applied with a considerable degree of uniformity regarding
the salient criteria for a credible failing firm claim. Satisfaction is required of each
of the following criteria, namely that:
(i) the firm is a failing one;
(ii) the reorganisation of the alleged failing firm is not a realistic option;
and
(iii) a less anticompetitive outcome than the proposed transaction is
absent.
[55] The Tribunal in the abovementioned Iscor Saldanha Steel matter held that
the merger criteria for a failing firm as set out in the tests of other jurisdictions
will carry serious weight in our assessment. 13 It is thus incumbent upon the
Tribunal to examine each of the abovementioned criteria commonly used in
assessing the salience of a credible failing firm finding, and we do this in the
following paragraphs.
Likely firm failure
[56] Tribunal jurisprudence highlights the fact that it is not necessary in terms of
the Act to show that a firm has already failed (as required in some other
jurisdictions); failure also need not equate to insolvency. Evidence is required to
12 Idem footnote 10, paragraph 101.
13 Idem footnote 10, paragraph 110.
20
NonConfidential version
substantiate likely failure. 14 However, likely failure is a complex factual analysis
and amounts to showing much more than a degree of financial distress. A
common standard found in other jurisdictions is that the alleged failing firm must
prove that it would be “ unable to meet its financial obligations in the near
future”.15
[57] It is stressed that a strict evidentiary approach to likely firm failure is entirely
justified given the alleged failing firm’s distinct and substantial incentive to
establish the semblance of a failing firm in order to alleviate competition
authorities’ opposition to an ordinarily anticompetitive merger. Such illusion can
be created inter alia by creative accounting methods and therefore proper
scrutiny, on a casebycase basis, is required of the true financial position of the
alleged failing firm, regardless of the type of industry in question and regardless
of whether or not that industry is subject to sectorspecific regulation.
[58] Counsel for Super Group at the hearing conceded that “ technical insolvency ”
is not an obvious “ failing firm scenario ”. However, the Commission concluded
that Emerald is a failing firm based in the main on the fact that it has for some
time not met the FSB’s minimum regulatory requirements. More specifically,
Emerald no longer complies with a minimum (local) solvency level of 15% and is
therefore considered by the FSB to be technically insolvent. The FSB calculates
solvency as the firm’s net written premiums as a percentage of its “qualifying” 16
net asset value. It is important to note that Emerald’s actual audited financial
statements thus differ from its statutory balance sheet for FSB regulatory
purposes (also see paragraphs 60 and 61 below).
[59] As stated in paragraph 56 above, the likely failure of a firm is a question of
[59] As stated in paragraph 56 above, the likely failure of a firm is a question of
fact and as such one would have expected the Commission to perform its own
analysis of the financial position and comparative market performance of
Emerald before reaching its ultimate conclusion that Emerald is likely to fail. This
14 Idem footnote 10, paragraph 109.
15 See, for example, the 1997 Horizontal Merger Guidelines published by the USA
Department of Justice and the Federal Trade Commission, as well as the Merger Guidelines
of the Office of Fair Trading of the United Kingdom.
16 The FSB excludes certain “nonqualifying” assets as per the company’s balance sheet from
this net asset value calculation.
21
NonConfidential version
analysis could have related to inter alia (i) Emerald’s financial documents such
as balance sheets, income statements and cash flow statements; (ii) Emerald’s
past and recent performance compared to other market participants; (iii) the
level of investment required in Emerald to address the solvency and other
regulatory requirements; (iv) Emerald’s relationships with creditors; and (v)
Emerald’s access to internal funds and external capital.
[60] From a competition law perspective it is entirely proper to broaden this
analysis to the root causes of Emerald’s sectorregulatory difficulties, including
the identification and analysis of the differences between Emerald’s actual and
regulatory balance sheet. Corke during his testimony summarised the
fundamental causes of Emerald’s credit rating downgrade by Global Credit
Rating (GCR) and solvency issues as follows: (i) “stripped capital” in the form of
paid dividends; (ii) a very substantial intercompany loan to Super Group, of
which only 2.5% is recognised by the FSB for solvency purposes; and (iii) a
large amount of FSB “nonapproved” reinsurance which is not an admitted asset
for FSB regulatory purposes 17 (also see paragraph 21 above).
[61] Emerald’s latest available financial statements confirm these facts and
attributes its lack of compliance with the FSB’s capital requirements and
solvency ratios mainly to the treatment of outstanding loss reserves on foreign
reinsurance in the regulatory balance sheet where the aforementioned is
disallowed in terms of STIA. 18 Other factors that contribute to the shortfall in
Emerald’s regulatory qualifying assets include a very substantial intercompany
loan to Super Group and an investment in a subsidiary. 19 It is noted that the
loan to Super Group and an investment in a subsidiary. 19 It is noted that the
settlement of these amounts owed by Super Group to Emerald in any event is a
prerequisite of the proposed deal, as confirmed by Corke.
[62] An initial analysis of Emerald’s financial statements for the years ended 30
June 2008 and 2009 reveals the following results:
17 Section 29 of the STIA relates to the aggregate value of assets held and section 30 to the
kinds and spreads of assets.
18 Emerald financial statements for the year ended 30 June 2009.
19 Letter of 15 January 2009 from Emerald to the FSB.
22
NonConfidential version
• dividends of approximately R[...] million were paid during the 2008 financial
year. Corke in fact testified that “ two dividends had been taken out of
Emerald Insurance Company in the last four years ”;
• on 30 June 2009 total assets exceed total liabilities by approximately R[...]
million. Not surprisingly, Corke testified that “[i]f it weren’t for the terms of
the Insurance Act, etc, I would have every confidence that the assets of
Emerald Insurance Company would exceed its liabilities ”;
• assets on 30 June 2009 include amounts owing by group companies (loans
and receivables) to the value of approximately R[...] million;
• profit after tax amounts to approximately R[...] million in the 2009 financial
year;
• cash and cash equivalents at the end of the 2009 financial year are
approximately R[...] million; and
• gross premiums increased by [...]% from 2008 to 2009 (approximately R[...]
million for the year ended 30 June 2009 and R[...] million for 2008).
[63] The above results speak for themselves and certainly, at face value, are not
indicative of a likely commercially failing Emerald. In fact, no compelling
evidence of a financial nature has been adduced by the merging parties (or the
Commission) that Emerald is likely to fail. The available evidence rather
suggests that the precarious position that Emerald finds itself in today is not of
its direct own making, but rather a consequence of Super Group’s by now
common cause financial difficulties. Emerald’s immediate and longerterm future
it seems depends entirely upon Super Group’s commitment to protect Emerald’s
compliance with regulatory requirements and it is noted that the Super Group
Board in correspondence with the FSB in no uncertain terms articulates this
commitment and support to Emerald. 20
commitment and support to Emerald. 20
[64] Furthermore, Mr. Gerald Kennedy of Super Group provided a noteworthy
summary of the prevailing circumstances surrounding Emerald and its potential
recapitalisation at the hearing: “... there is a possibility that we can recapitalise
the group, but from a Board perspective the group has decided to exit its non
core operations. It hasn’t contemplated refinancing Emerald Insurance
20 Letter from Super Group to the FSB dated 24 March 2009.
23
NonConfidential version
Company under its recapitalisation program with financiers. So, at this stage ...
it’s still not off the table. The door is not closed on that ... ”. From this submission
it is evident that Super Group does not lack the ability but rather the inclination to
recapitalise Emerald and to have done so in a timely fashion. This has resulted
in Emerald’s spiralled lower regulatory solvency levels to its current “technical
insolvency”. The fact that Emerald at the time of the hearing “ does not have
sufficient time left to it to salvage its business ”, as submitted by the merging
parties, seems entirely of Super Group’s making and should therefore be treated
with scepticism (also see paragraphs 72 to 77 below that deal with the issue of
the potential reorganisation of Emerald).
[65] We conclude that there is no factual basis to conclude that Emerald is either
failing or likely to fail.
Alternative offer (s) for target firm(s)
[66] The next issue that we shall examine, on the assumption that the firm in
question is indeed likely to fail, is whether or not there is an alternative buyer
whose purchase of the target firm(s) would raise less competition concerns than
the transaction under scrutiny. For a successful failing firm contention, the
merging parties must show that there is no less anticompetitive purchaser than
the acquiring firm. The Tribunal jurisprudence 21 is unequivocal regarding the fact
that no leniency would be afforded to this requirement, and we strongly reiterate
that here.
[67] In the above context the assumed failing firm must demonstrate inter alia
that it has made reasonable and verifiable good faith attempts to elicit
reasonable22 alternative offers and, furthermore, that there is no viable
alternative purchaser that poses less anticompetitive risk than does the
proposed transaction.
21 Idem footnote 10, paragraph 110.
proposed transaction.
21 Idem footnote 10, paragraph 110.
22 In terms of the USA Horizontal Merger Guidelines (1997) any offer to purchase the assets
of the failing firm for a price above the liquidation value of those assets will be regarded as a
reasonable alternative.
24
NonConfidential version
[68] The instant evidence is clear on the score that the target firms have indeed
endeavoured to locate alternative buyers. Emerald confirmed that it approached
inter alia Santam, Zurich Re, Gallagher and Capital Worx regarding Emerald’s
disposal. Emerald board minutes 23 also confirm that “ the company had been
approached by many insurers with expression of interest .” Furthermore, Capital
Worx, a venture capital company, made a formal offer for 100% of the equity in
Emerald. Moreover, the latter deal would give rise to fewer, if any, likely
competitive concerns since the acquiring parties would be a new entrant in the
relevant market, according to the potential purchaser (and confirmed by Corke
during his testimony). Corke also confirmed that the Capital Worx written offer
was “ identical in actual fact, I think ” to the Santam price, but that “ [w]e as a
management team turned it down ... ”.
[69] It is noted that a less anticompetitive alternative may also include the
counterfactual scenario where Emerald is allowed to fail and exit the relevant
market(s) and some or all of its assets are transferred to new or incumbent
firms. In the instant case there is no reason to believe that incumbent firms in the
market or potential new competitors would not be interested in some of the
assets of the target firms, more specifically in the ERT (t/a EUM) assets, to be
precise the intellectual capital. Corke confirmed that “ the goodwill element of the
purchase price is in the Emerald Risk Transfer element of the business ”.
Furthermore, Matthew testified that in the event of the current acquisition of both
Emerald and ERT not being approved, that Santam would consider a deal with
ERT “ on the basis that Santam is the insurance licence ”. Corke further
confirmed that ERT had indeed been approached by Zurich, Investec and
ABSA, but that “ management had agreed to sell the 62% [of their shares in
ERT] to Santam ”.
[70] Furthermore, there is no reason to believe that if Emerald were allowed to
fail and exit the market that Santam would in effect gain Emerald’s entire market
share. In the latter scenario the factual and counterfactual would thus not
produce the same resultant market structure, since the intellectual capital of
ERT may divert to one or more other industry participant.
23 Minutes of 01 June 2009.
25
NonConfidential version
[71] Nor even if Emerald were to fail, which we do not accept would have been
likely to happen, would there be systemic harm regarding the anticipated effect
on policyholders with potential claims. Corke in this regard testified that in the
event of the proposed acquisition not proceeding that “... most of the claims will
be met, if not all. ... I think the reinsurers will still back Emerald Insurance
Company. It won’t be them running away ... it’s my belief that EIC will eventually
meet all these payments ...”. In this context it is important to note that Emerald
cedes reinsurance in the normal course of business for the purpose of limiting its
net loss potential through the diversification of its risks. Its policies for mitigating
risk exposure include the use of both facultative and treaty reinsurance against
insurance risks. Thus, the reinsurance agreements spread the risk and minimise
the effect of losses (also see paragraphs 18 to 21, as well as paragraph 44
above).
Reorganisation of target firm(s)
[72] The next issue to consider is whether or not any realistic prospect exists for
Emerald’s successful reorganisation to address its alleged failure and enable it
to survive as a meaningful competitor in the relevant market(s). That is, is there
any prospect of Emerald surviving as a standalone player without the merger?
[73] In a letter addressed to the Tribunal as recent as 12 October 2009 the CFO
of Super Group states:
“we have since the 24 March 2009 in our attached letter indicated that
Super Group will resolve to adequately meet its obligations as agreed with
the FSB. To this end the Group has numerous options available to it,
including but not limited to a recapitalization, the implementation of a
reinsurance program, to runoff the existing insurance portfolio and/or to
reinsurance program, to runoff the existing insurance portfolio and/or to
resume negotiations with other interested parties. To this extent it must
also be borne in mind that the Group EIC is a 40% shareholder in Emerald
Underwriting Managers, an operation that will more than likely continue
into the near future. Further to this the conclusion of Super Group’s rights
issue as announced in the media, will place the Group in a financially
stable position .”
26
NonConfidential version
[74] Furthermore, nowhere in this letter of said date is it suggested that Emerald
is a failing or likely failing firm, which supports our finding regarding Emerald’s
unlikely failure (see paragraphs 56 to 65 above).
[75] Based on the abovementioned Super Group submissions, as well as the
submissions of Mr. Gerald Kennedy of Super Group at the hearing (see
paragraph 64 above), the real prospect of successfully reorganising Emerald
cannot be disputed.
[76] In summary, we found no evidence in support of a valid failing firm argument
in this case. First, there is no evidentiary basis to conclude that Emerald is likely
to fail despite its difficulties brought about by Super Group’s financial situation.
Second, an alternative purchaser has made a reasonable offer for Emerald as a
going concern and this deal would highly unlikely give rise to any competition
concerns. Third, Super Group has submitted documentary evidence that
unambiguously state that Emerald could be successfully reorganised absent the
proposed deal.
[77] The Tribunal notes its discontentment with Super Group/Emerald for two
reasons: first, the nondisclosure of a material fact to the Commission, namely
that an alternative offer was made for the target firm(s); and second, the frankly
noncredible attempt to rely on the failing firm doctrine “ as an alternative ” toward
the end of the hearing into this matter “ because it is out there ”, whilst it had
neither been claimed prior to the hearing nor any attempt made to produce the
required evidence to meet the merging parties’ requisite burden of proof.
Removal of an effective competitor
[78] The Commission reaches two seemingly conflicting conclusions in this case,
namely that (i) the proposed deal would result in the removal of Emerald as an
effective competitor, and (ii) Emerald is a failing company or likely to fail (see
effective competitor, and (ii) Emerald is a failing company or likely to fail (see
paragraphs 56 to 65 above). The Commission gives a historic account of
Emerald as an aggressive, growth orientated, flexible, entrepreneurial and
innovative insurer, as well as an effective competitor. Third parties contacted by
the Commission were however of split opinion on the wisdom of Emerald’s
27
NonConfidential version
underwriting approach, i.e. if Emerald’s growth is as a result of a reckless and
unsustainable underwriting model or a truly innovative approach. The merging
parties, on the other hand, allege that Emerald is no longer an effective
competitor given that it is “ hamstrung” because of impaired capital requirements
and solvency ratios.
[79] There is no dispute between the Commission and the merging parties
regarding Emerald’s numerous said attributes, except for the latter, i.e. opposing
views regarding Emerald’s effectiveness. The Commission’s investigation had
however indicated that Emerald has lost some effectiveness in the immediate
recent months due mainly to Super Group’s publicly quoted financial issues
(also see paragraph 63 above).
[80] The merging parties argue that the fact that Emerald’s lack of meeting
mandatory capital/solvency requirements may be of its own making, which they
do not concede, is of no relevance in this merger context and do not make them
less compelling to Emerald’s recent lack of competitiveness. The Tribunal is
however of the view that the actions of Super Group/Emerald are extremely
pertinent in light of the fact that they have every inducement to present Emerald
as an ineffective competitor in order to gain approval for a potentially
anticompetitive deal. Very recent claims of ineffectiveness, as alluded to in this
case, must be thoroughly interrogated, especially when they appear to be
generated or caused by the actions, or lack of actions as the case may be, of a
parent company (Super Group) and its strategy, rather than on the true market
perception of Emerald’s competitive significance.
[81] We stress that the issue of the effectiveness of Emerald (or recent
ineffectiveness as argued by the merging parties) cannot be severed from the
factors considered in the alleged failing firm analysis, specifically the
fundamental causes of Emerald’s current sectorregulatory predicament,
including the role of Super Group in that. Having said this, it is by no means
suggested that the issue of sufficient capital to meet statutory solvency
requirements is not a relevant one in the context of this transaction and this
relevant market. The factual and perceived financial position of an insurer in this
relevant market is highly relevant, but in a merger regime context this must be
28
NonConfidential version
evaluated in the context of the fundamental causes of any such (recent)
perception and the potential remedies thereof. As concluded in paragraph 75
above, there can be no doubt in the instant matter of the realistic option of
successfully reorganising Emerald, given Super Group’s recorded commitments
to the FSB in this regard. Moreover, this option was reiterated in a very recent
communiqué of the merging parties to the Tribunal (see paragraph 73 above).
[82] Furthermore, although it is reasonably plausible that a truly failing firm could
not attain qualification as an effective competitor, Emerald (as concluded above)
is by no stretch of the imagination a commercially failing firm (see paragraphs 56
to 65 above). It is noteworthy that Santam’s board briefing document on the
merger supports the notion of a commercially highly successful Emerald, despite
certain capital constraints: “ EIC has reflected massive premium growth, limited
only due to its capital constraints. Gross Written Premium (“GWP”) for the [...] to
February 2009 was greater than the GWP for the [...] preceding it, and the book
has grown by almost [...]% since 2007 ”.
[83] In light of the above, t he Tribunal is highly sceptical of the merging parties’
extremely short term perspective of Emerald’s alleged ineffectiveness. Corke’s
testimony by no means points to Emerald’s pivotal reputational impairment. To
the contrary, he testified that even under the current conditions the brokers
“have shown faith and confidence ” and generally have been supportive of
Emerald/EUM.
[84] Based on the above, we have no firm basis on which to conclude that
Emerald is not an effective competitor in the short term corporate insurance
sector, given inter alia the realistic possibility of its successful restructuring
(according to the merging parties own documents). On the premise that Emerald
is an effective competitor in the overall short term corporate insurance industry,
the question that remains is if Emerald is also a credible and therefore effective
competitor in a potential lead market. This is discussed below (see paragraphs
91 to 100).
Countervailing power
29
NonConfidential version
[85] Most of the larger corporates have risk officers or insurance managers with
insurance knowledge who coordinate and manage their insurance. Furthermore,
the short term corporate insurance industry is almost entirely intermediated
through large corporate insurance brokers, for example Glenrand MIB,
Alexander Forbes, Aon, Marsh and Willis, who are in many cases local
subsidiaries/branches of international corporate insurance brokers (for example
Willis, Aon and Marsh). Corke described this relationship between the risk
manager and broker as follows: “ the risk manager would habitually go in South
Africa to one of five broking houses ... and with the broker at that broking house
they would make the decisions of how they wanted the insurance placed ”.
[86] These appointed brokers would generally on an annual basis seek quotes
from insurers and then negotiate terms with selected insurers. In the case of
coinsurance contracts such negotiations would generally be conducted with their
preferred lead insurer. Negotiations appear generally to proceed around broker
terms (with the large brokerages having their own precedents/standard terms)
rather than any standard insurer produced wording (also see paragraph 23
above).
[87] The Commission conclude d that the size of the large corporate brokers, their
expertise, and in the case of some of the international corporate brokers their
international scope, may provide some degree of countervailing power to a
customer. However, the Commission provided no examples of situations in
which this alleged customer countervailing power had been effectively
exercised. The Commission in its recommendation also suggests that there may
be current corporate skill scarcities amongst brokers that may negate this power.
be current corporate skill scarcities amongst brokers that may negate this power.
We therefore conclude that there may be some element of customer
countervailing power, but that it has not been satisfactorily established.
Conclusion
Overall short term corporate insurance market and potential follow market
[88] Although the merged entity would be the largest player post merger in the
South African overall short term corporate insurance market, various competitors
remain active in that market post merger. The Commission’s market
30
NonConfidential version
investigation and witness testimony confirmed that AIG and M&F are the largest
competitors to the merged entity in terms of capacity in an overall national
market for short term corporate insurance. M&F has a market share exceeding
20% in this market and AIG has a market share exceeding 10%. Furthermore,
there is a competitive fringe of at least five smaller competitors. Etana, Zurich,
Lion of Africa and to a lesser extent ACE were identified in the Commission’s
market investigation as predominantly participants in the follow market.
Customers also confirmed that there are a sufficient number of smaller players
that can put down follow capacity. Furthermore, two potential new entrants in the
follow market have been identified, namely ABSA and RMB. This alleviates any
likely unilateral or coordinated post merger competition concerns in a potential
follow market.
[89] Furthermore, the proposed merger would not alter certain market dynamics,
for example (i) well advised clients that make use of large brokerages who may
have a degree of countervailing power; (ii) annual insurance contracts with low
or insignificant switching costs (penalties only apply if contracts are terminated
midterm and if there was a claim); and (iii) innovative approaches and
differentiated offerings by market participants to limit their risks as far as
possible.
[90] Based on the above , no significant competition concerns arise as a result of
the proposed deal in the overall market for short term corporate insurance and in
the possible narrower follow market. The potential lead market is discussed
below.
Potential lead market
[91] From witness testimony it is blatantly evident that the lead insurance
provider sets the terms and conditions for short term corporate insurance. Corke
provider sets the terms and conditions for short term corporate insurance. Corke
articulated the significance of the role of the lead in coinsurance arrangements
as follows: “... the broker and the risk manager will come up with a strategy as to
what their preferred lead is, what the terms of the contract are, because the lead
office will set the contract terms... The broker would then go to the market, the
31
NonConfidential version
coinsurance market ... and say these are the lead terms ... on this particular risk,
would you be interested in participating?”.
[92] Corke furthermore confirmed that the lead constantly sets the insurance
price. He testified that in his experience the lead sets the price in “99% of
business” and all other insurers will be on the same pricing schedule (excluding
a 2.5% (contract) handling charge by the lead office). He further stated: “ I don’t
think there is one [of circa [...] clients] where our [Emerald’s] price is below what
the lead price was ... it’s never in our interest to be seen as somebody who cuts
the price, because our share would be a lesser price ”. He also testified that there
were only four or five accounts on which Emerald as coinsurer achieved a
higher price than the lead. He further confirmed that pricing is usually driven by
the pricing for reinsurance capacity. In fact, the price of short term corporate
insurance is a derivative of the costs of inter alia reinsurance and capital. As
pointed out in paragraph 19 above, the price of reinsurance is largely
determined by factors external to the control of the primary insurer.
[93] As stated in paragraph 26 above, while the available evidence does not
suggest a rigid division between players in lead and follow positions, there does
appear to be a very small group of players who dominate the lead market.
According to the Commission, competitors and brokers alike agree that Santam
and AIG are the most significant leads, followed by M&F. Furthermore, as
indicated in paragraph 43 above, Matthew conceded that new entry into a
potential lead market is unlikely. He explained that the ability to take lead
positions “ is based on credibility and relationships with brokers as well as your
clients in the market and for a new entrant to come in and form a leadership
position is very unlikely ... ”.
[94] Although Emerald is generally regarded as a predominantly follow insurer,
the Commission however received mixed information on whether Emerald could
be characterised as a credible lead (also see paragraph 24 above). Matthew
conceded that Emerald is a good example of an entrant that geared up from
being a follow capacity provider to “ at least competing with some of the lead
markets”. He however was of the view that the brokers still prefer the traditional
32
NonConfidential version
lead companies, i.e. Santam, AIG and M&F, for the security that they provide to
clients to meet future obligations.
[95] From the depicted market structure, pricing practice, entry barriers and other
facts relating to this case it is clear that the salient hypothetical theory of harm is
potential adverse (unilateral) competition effects in a potential lead market. In
this potential market there would be only three significant players post merger,
namely Santam/Emerald, AIG and M&F. However, for the reason stated in
paragraph 24 above, the Commission did not focus its attention on this potential
market. As a result, neither has such market been properly defined by the
Commission as a separate relevant market, nor satisfactory qualitative and
quantitative evidence provided on the likely competitive effects of this deal on
this potential market.
[96] To contextualise Emer ald’s actual lead positions, Corke stated that out of the
13 common accounts between Emerald and Santam (see paragraph 38 above),
Emerald has the lead of none of them. He further testified that Emerald has very
few lead policies on its books and in his view is not perceived as a “ credible
lead” it is the lead of less than 10% of its portfolio. The reason for this
according to Corke is that: “ whilst ... our market reputation means that we
[Emerald] have the intellectual capacity to be the lead, I don’t think many people
think we’ve got the balance sheet to be the lead ... Super Group was never
perceived as a financial service provider ... part of our strategy ... was to be a
second lead, in other words, have a big chunk of the programmes that we
wanted ... it’s easier for a broker to sell the Emerald brand as a second lead
than as a major lead ... we were seen as a capacity provider ... as somebody
than as a major lead ... we were seen as a capacity provider ... as somebody
who could come up with solutions and back it up with reinsurance ”.
[97] Corke further testified that Emerald and Santam differ in the manner in which
they segment capacity. He explained the latter comparative difference as
follows: Emerald has the ability to book only a R300 million line, compared to
Santam who can book a R750 million line. This implies that on a portfolio of R1.5
billion, Santam could book 50% down, but Emerald only 20%. This suggests that
portfolio size could be a significant factor in assessing lead credibility.
33
NonConfidential version
Unfortunately brokers have in this case not submitted comparable figures to the
Commission regarding the relative size of the risk/contract, i.e. some submitted
premium information, others insured sum information, and others no information
on the relative size of the risk/contract (also see paragraph 100 below).
[98] The Tribunal analysed the available quantitative information on lead contacts
as submitted to the Commission by various brokerages. Our analysis confirms
that AIG and Santam indeed are the main leads. M&F also has a number of
substantial lead contracts. However, the available data suggest that not only
Emerald but a number of smaller players in the industry, including Zurich, ACE,
Lion of Africa, Allianz and to a lesser extent Etana, from time to time take lead
positions, of which some are significant in terms of premium/sum insured.
[99] In conclusion, there is no evidentiary foundation that the proposed merger is
likely to substantially prevent or lessen competition in a potential lead market.
More specifically, there is no quantitative or other evidentiary support for the
aversions that Emerald played a maverick or more significant role than said
other smaller players in a potential lead market. Based on the limited and mixed
available information we cannot determine with sufficient certainty how the
proposed merger would impact the number of credible leads.
[100] A more narrow focus by the Commission on the effects of the proposed
deal on a potential lead market, including the characteristics and competitive
dynamics thereof, would have placed the evidentiary value of market positions,
the views of brokers/customers and other relevant information in context. Broker
data on the quotation and selection processes of (preferred) leads in past
awarded contracts would have been of more value to establish lead credibility
from a customer perspective than the furnished information on actual lead
positions post conclusion of these processes. Broker information was namely
submitted to the Commission regarding (i) the identity of ultimate lead(s); (ii) the
type of risk; and (iii) the sector in question, for example mining, retail or other
relevant sector. Meaningful additional data on a per quotation basis would
include inter alia (i) the broker stipulated criteria/conditions that potential leads
have to comply with; (ii) in a comparable format, for example on the basis of sum
insured, the relative size of the risk/contract; (iii) the identities of potential leads
34
NonConfidential version
who submitted quotes; (iv) the price and other terms and conditions offered by
each participant; (v); the manner in which the broker/client in question used this
information to ultimately negotiate better terms and conditions with the ultimate
lead; and (vi) the broker’s criteria for selecting the preferred lead/reason(s) for
the final lead selection.
PUBLIC INTEREST
[101] No significant public interest issues arise as a result of the proposed deal.
CONCLUSION
[102] Since there is no evidence of a likely substantial prevention or lessening of
competition in any (potential) relevant market as a result of the proposed deal,
and also no significant public interest issues arising from this deal, we
accordingly approve the transaction without conditions.
____________________ 27 January 2010
A Wessels Date
N Manoim and A Ndoni concurring.
Tribunal Researcher: I Selaledi
For the merging parties: Cliffe Dekker Hofmeyr Inc
For the Commission: F Reid and M Van Hoven (Mergers & Acquisitions)
35