1
COMPETITION TRIBUNAL OF SOUTH AFRICA
Case No: LM061Jun20
In the matter between:
DH BROTHERS INDUSTRIES (PTY) LTD Acquiring Firms
SEABOARD CORPORATION
and
RUSSELLSTONE PROTEIN (PTY) LTD
Target Firm
CONDITIONAL APPROVAL
[1] On 31 August 2021, the Competition Tribunal conditionally approved the
proposed acquisition of joint control of Russell
by DH Brothers Industries DH Brothers Seaboard Corporation
.
Panel: Y Carrim (Presiding Member)
A Wessels (Tribunal Member)
H Cheadle (Tribunal Member)
Heard on: 26 29 July 2021, 2- 3 August 2021, 6 August
2021
Last submission received on: 26 August 2021
Order issued on: 31 August 2021
Reasons issued on: 29 October 2021
REASONS FOR DECISION
2
BACKGROUND
[2]
notice of a large merger between DH Brothers, Seaboard and RSP.
[3] The Commission filed a recommendation of prohibition on 22 April 2021. The
Commission had concluded that the proposed transaction was likely to result in
a substantial prevention and/or lessening of competition and that the merger
ought to be prohibited. The merging parties, on the other hand, disputed that the
transaction led to a substantial lessening of competition but were willing to
tender conditions to address the Commission's concerns.
[4] The matter was set down for hearing and the Commission persisted with its
recommendation for prohibition on the basis that the conditions tendered by the
merging parties did not adequately address its concerns.
[5] We heard evidence on 26 29 July 2021 and 2 3 August 2021, with closing
argument on 6 August 2021.
[6] The Commission called the following witnesses to give evidence at the Tribunal
hearing:
Factual
6.1. Mr Morne Botha from Wilmar SA (Pty) Ltd.1
Expert
6.2.
[7] The merging parties called the following witnesses to give evidence at the
Tribunal hearing:
1 Wilmar SA is an edible oil and fats company which is involved in, among others, oilseed
(soybean/sunflower seed) crushing, refining, and bottling. Wilmar SA is part of Wilmar International Ltd,
an agricultural commodity business headquartered in Singapore.
3
Factual
7.1. Mr Johannes De Wet Boshoff from the Animal Feed Manufacturers
7.2. Mr Heiko Koster from Kaonne Investments (Pty) Ltd;2
7.3. Ms Lesley Heads from Seaboard; and
7.4. Mr Mohamed Zubeir Moosa from Willowton.
Expert
7.5. Professor Liberty Mncube from FTI Consulting.
[8] We note that the merging parties customers such as
Mr Heiko Koster from Kaonne Investments (Pty) Ltd. The Commission's only
witness from Wilmar, which had a toll crushing agreement with RSP for a short
period of time when its plant had burnt down, was essentially a competitor of the
merging parties. No other factual witnesses were called by the Commission
either in support of its theory of harm or in rebuttal.
[9] At the end of the hearing, we invited the merging parties to submit their proposed
remedies, which they did on 10 August 2021. On 23 August 2021, we circulated
proposed conditions for comment by the merging parties and the Commission.
[10] Following our request for comments on the proposed conditions, the
Commission made submissions indicating its discontent with the proposed
conditions. However, the merging parties noted that the Commission had had
ample opportunity to engage on the remedies but failed to do so, as its stance
has consistently been that no remedies could cure the likely substantial
prevention or lessening of competition that would be caused by the proposed
transaction.
2 Kaonne Investments (Pty) Ltd is the holding company of a number of companies involved in feed mills;
animal nutrition; specialty products for the animal feeds market; consulting with major feed companies
on feed milling; formulations; on-farm production of poultry, dairy, swine, feedlot and sheep production
in most sub-Saharan and North African countries; and supplying most feed companies in the ruminant
market.
Mr Koster was previously the Vice Chairman of the Board of AFMA and Vice Chairman of the Technical
Committee of AFMA. At the time of the hearing, Mr Koster was the Chairman of the Trade Committee
of AFMA.
4
[11] After consideration of the submissions by the merging parties and the
Commission, we decided on a set of conditions, which we are of the view
adequately address the competition concerns in the identified market and
approved the transaction conditionally.
[12] These are the reasons for our approval. The conditions are attached hereto as
Annexure A.
Procedural Issues
[13] We pause to mention here that the timetable for the further conduct of
proceedings was outlined in the Tribunal direction of 6 May 2021, which included
a process of discovery, as contemplated in section 55 of the Competition Act No
89 of 1998, as amended (
an application to compel the discovery of certain internal documents of the
discovery
Commission also filed an application to compel various documents from the
discovery
were heard on 23 June 2021. On 2 July 2021, we dismissed the merging parties
discovery discovery application on
8 July 2021, most of which the merging parties had agreed to discover. The
reasons for our decision to dismiss the merging parties discovery application
are provided here after the merits of our decision have been dealt with.
PARTIES TO THE TRANSACTION
Primary acquiring firms
[14] DH Brothers is a wholly owned subsidiary of Willowton Group (Pty) Ltd
, which is indirectly controlled by several family trusts. 3
3
5
[15]
5 In these reasons we refer to DH Brothers, its
subsidiaries, and its controllers .
[16] Willowton is active in sunflower seed and soya bean crushing and oil refinery
operations and has interests in the manufacturing and sale of edible oils,
cleaning and packaging of rice, maize milling and in the fast-moving consumer
goods (FMCG) market. Willowton sells its products under, amongst others, the
[17] Willowton has manufacturing facilities in Pietermaritzburg, Kempton Park and
Cape Town, and owns four silos located at its oil seed crushing facility in
Pietermaritzburg and a further 12 silos at its facility in Isando, Kempton Park.
[18] Seaboard is a company listed on the New York Stock Exchange
two largest shareholders6 are SPC Preferred, LLC (as to 36.7%) and Seaboard
Flour, LLC (as to 39.9%). Seaboard has shareholding interests in multiple firms
in South Africa:
18.1. As to 100%: Seaboard Overseas Trading and Shipping (Pty) Ltd (South
Africa); InterAfrica Grains (Pty) Ltd (South Africa); Paramount Mills (Pty)
Ltd (South Africa); Prize Milling (Pty) Ltd; and Showlands Investments
(Pty) Ltd.
4 Willowton SA holds over 50%, Noko Milling (Pty) Ltd, which in turn wholly controls Viva Milling (Pty)
Ltd and Noordfed (Pty) Ltd.
5
.
6 As of 8 March 2020.
5
6
18.2. As to 50%: Eurogerm South Africa (Pty) Ltd (South Africa) and Gloridge
Bakery (Pty) Ltd.
18.3. As to 30%: the target firm, RSP.
[19] Seaboard is active in South Africa in the trading of agricultural commodities such
as soya beans, soya meal, soya hulls, crude soya oil, sunflower seeds,
sunflower meal, sorghum, wheat, and maize. Seaboard mainly imports the
products but also sources these products locally. Seaboard currently has a toll
crushing agreement with RSP, in terms of which RSP crushes soya beans and
supplies soya meal, soya hulls, and crude soya oil to Seaboard, in exchange for
a toll fee. Seaboard has no crushing facilities of its own.
Primary target firm
[20] The primary target firm, RSP is controlled by African Afgri Opportunities Limited
30% is held by Seaboard. African
7 which closed in
March 2019 and no longer has any business operations. RSP controls two non-
operational companies.8
[21] RSP owns a soya bean crushing plant situated in Bronkhorstspruit. The plant
only crushes soya beans. The soya t is used in the
animal feed industry, mainly for poultry and pigs.
Activities of the parties in the soya bean industry
[22] The soya bean value chain comprises of the following market participants: 9
22.1. Producers of soya beans are commercial and smallholder farmers
mostly in the inland region who grow the soya bean crop. Soya beans
7 Elangeni Oil is a subsidiary of African Afgri Opportunities (Pty) Ltd and ceased operating in March
2021.
8
9
7
are also imported into South Africa from countries such as Argentina,
Brazil and the United States of America.
22.2. Traders are commodity traders who source soya beans or derivatives of
soya bean (soya meal, soya hulls, or crude soya oil) locally or from
imports in order to meet the needs of processors. Traders ensure that
producers have a ready market for their soya beans and derivatives, and
processors have a reliable supply for their inputs. The JSE platform,
SAFEX,10 is used by traders to achieve the best price for the produce
and to supply or negotiate with processors. Examples of traders include,
among others
Limited, and Cargill Inc.
22.3. Crushers or processors crush the raw soya bean to produce soya meal,
soya hulls and crude soya oil. Crushing refers to the process of dehulling
and extraction. The soya meal (and in certain instances, soya hulls) is
then sold to animal feed manufacturers who use it as an input for animal
feed, primarily for the poultry industry. Examples of crushers include,
among others, Willowton, RSP, Nedan, Wilmar and COFCO.
22.4. Refineries, which process crude soya oil further to yield refined soya oil.
The refined soya oil is sold to consumers as an edible oil to be used for
cooking.
[23] Willowton crushes both soya beans and sunflower seeds at its plant in Isando.
Willowton's plant can switch between soya and sunflower seeds. While
Willowton is regarded primarily as an oil manufacturer in refined sunflower and
soya oil, it also supplies soya meal to the animal feed market.
[24] Seaboard is a trader in soya meal but does not own a crushing facility. Prior to
its toll crushing agreement with RSP, Seaboard was primarily an importer of
soya meal.
10 South African Futures Exchange.
8
[25] RSP's crushing plant in Bronkhorstspruit is dedicated only to the crushing of
soya beans.
[26] RSP has been toll crushing for Seaboard since In terms of the toll
crushing agreement, all the soya meal, soya hulls and crude soya oil produced
by RSP are supplied exclusively to Seaboard.11
[27] Willowton procures all of the crude soya oil produced by RSP from Seaboard. 12
[28] Thus, the pre-merger arrangement between the three parties is that Seaboard
off-takes all the soya meal and hulls produced by RSP (the target firm) and on-
sells all the crude soya oil produced by RSP to Willowton. Hence Seaboard and
Willowton are currently in a customer-supplier relationship for crude soya oil.13
TRANSACTION AND RATIONALE
[29] The proposed transaction involves two indivisible parts. The first part being the
acquisition of 50% of the shares and loan claims in RSP by Willowton; and the
acquisition of a further 20% of the shares and loan claims in RSP by Seaboard,
(
)
creating a joint venture ("JV"). The second part being the sale of certain of the
assets of Elangeni Oil, which include land, buildings, refinery and tanks, and
plant and equipment, to RSP.
[30] Willowton submitted, among others, that the proposed transaction presents an
opportunity for it to expand its business activities in the soya bean market and
to allow it to invest in a facility dedicated to soya bean crushing in order to
guarantee a consistent supply of good quality soya meal.14
15
11 Bundle B, Record, page 81, para 14.1.4.
12 Transcript (29 July 2021), page 515 lines 3-4; and page 517 lines 9 14.
13 Transcript (29 July 2021), page 515 lines 3-4.
14 Bundle B, Record, page 108, para 3.1.1 and 3.1.5.
15 Bundle B, Record, page 108, para 3.1.5.
9
[31]
.18
RELEVANT PRODUCT MARKET
[32] In determining the relevant product market, the Commission identified horizontal
overlaps in the markets for soya meal; soya hulls; crude soya oil; and refined
soya oil.
Soya meal
[33] Soya meal is a significant source of protein in the animal feed market, in
particular in the poultry industry. Based on international case law19 and the views
of market participants, 20 the Commission found that soya meal cannot be
adequately substituted for other oilseed products such as sunflower meal and
therefore constitutes a separate and distinct relevant product market. This was
consistent with the merging parties views.
Soya hulls
[34] In relation to soya hulls, the Commission did not define the market conclusively
as there was mixed evidence showing that (i) there are other products such as
wheat and bran that could potentially compete with soya hulls,21 and (ii) supply-
16 Bundle B, Record, page 109, para 3.2.1.
17 Bundle B, Record, page 109, para 3.2.2.
18 Bundle B, Record, page 109, para 3.2.3.
19Bunge and European Oilseed Processing Facilities Case No. M.8199.
20A feed manufacturer submitted that it cannot use sunflower meal, sunflower pellets or any other
oilseed product in place of soya bean meal in the event of a 5-10% increase in the price of soya bean
meal because soya bean meal has superior nutritional characteristics compared to the other products.
Another feed manufacturer noted that while soya bean meal and sunflower meal can be substituted,
this can only be done partially because the higher protein level of soya bean meal cannot be replaced
by sunflower meal due to sunflower meal having a lower protein level. A similar sentiment was echoed
by a com
21 Competitors of the merging parties submitted that all products with a high fibre content, such as wheat
and bran, do provide a competitive constraint on soya hulls. Further, soya hulls are priced to compete
.18
10
side substitution is limited. However, the Commission assessed the effects of
the proposed transaction in a distinct soya hulls market, similarly defined by the
merging parties.
Crude soya oil
[35] The Commission found that the market for crude vegetable seed oils is wide and
includes, among others, various seed crude oils such as soya bean, sunflower,
and canola crude oil.22 Seeing that the merging parties are only involved in crude
soya oil and crude sunflower oil, the Commission limited its assessment to the
market for crude soya oil, as the only area of overlap in the activities of the
merging parties. The
are aligned.
Refined soya oil
[36] The Commission did not conclude on the precise market delineation as it found
that there will be no overlap in the activities of the merging parties in relation to
refined soya oil.
Conclusion on the relevant product markets
[37] The economic experts of the merging parties and Commission agreed that the
proposed transaction raises no competition concerns in relation to the markets
for soya hulls, crude soya oil and refined soya oil.23
[38] As we discuss in detail below, it was accepted by both the merging parties and
the Commission that the relevant market of concern in this transaction is the
market for the sale of soya meal.
RELEVANT GEOGRAPHIC MARKET
[39] In defining the relevant geographic market, the Commission considered the
market to be national with imports. The Commission based this on submissions
Recommendation, para 78).
22 Bunge and European Oilseed Processing Facilities Case No. M.8199.
23 Exhibit M.
11
by third parties that soya bean products are supplied nationally and in certain
instances, to neighbouring countries.24
[40] The merging parties submitted that the relevant geographic market can be
segmented into a coastal market comprising of the Eastern Cape and Western
Cape provinces, and an inland market comprising the provinces of Free State,
Gauteng, Northern Province, Mpumalanga, North-West, and Kwa-Zulu Natal.
According to the merging parties, the supply for soya meal in the inland region
is sufficient to meet the demand inland, and where there is any surplus in the
inland region, this is used to supply the coastal region. However, the demand in
the coastal region is met primarily by imports due to the high costs of transporting
soya meal from inland to the coast. In certain instances, a small supply of
imported soya meal may be sold in the inland region. Thus, the geographic
market could be considered to be inland with imports and coastal with imports.25
[41] Ms Ranenyeni submitted in her expert report that the relevant geographic market
is national with imports. She found that imports are not limited to the coastal
region as approximately half of the imported soya meal is transported to the
inland region.26 She did not believe that transport costs were prohibitive in this
regard.
[42] Mr Botha, on the other hand, testified that inland suppliers do not usually
compete in the coastal regions because transporting soya meal to the coast from
inland is a high cost. Therefore, this would not be viable. It would be more
economical to rather import in the coastal region than transport from inland.27
be divided into two inland and coastal.
[43] In determining whether the geographic market is inland or coastal, Prof Mncube
assessed the impact that transport costs have on pricing and based his
24 commendation, paras 101 105.
25
26
27 Transcript (26 July 2021), page 22 lines 6 10, page 113 lines 6 10.
12
assessment on submissions by various competitors indicating that due to high
transport costs a supplier of locally produced soya meal will provide soya meal
in the inland areas, while imports will supply the coastal region. He found that
prices for locally-produced soya meal transported and supplied in the coastal
region tend to be high due to transport costs and that prices of imported and
locally-produced soya meal in the inland region tend to follow import parity
pricing (IPP) closely, factoring either a discount or slight premium.28 From this,
Prof Mncube concluded that there is evidence of different competitive dynamics
in the inland and coastal regions and that while locally-produced soya meal does
not constrain prices of imported soya meal sold in the coastal region effectively,
IPP (in the coastal regions) may constrain prices in the inland regions.29
Therefore, despite the debate about transport costs in the hearing, what was not
disputed and is apparent to us is that the market is 'divided' into the coastal and
inland regions due to transport costs. The coastal region is mainly supplied by
imports and the inland region largely by local crushers and by imports (whether
from the coast or overland from countries such as Zambia).30
[44] We note that following the expert meeting, the Commission and the merging
parties agreed that the relevant geographic market is national, including
imports.31
RELEVANT MARKET SHARES
[45] In its recommendation, the Commission noted that the size of the market must
be based on the products which are physically available. Hence, the
Commission based its market share calculations on products which are (i) locally
produced, (ii) procured through toll manufacturing agreements, and (iii) directly
imported from abroad. The Commission excluded the products that are sourced
28 Bundle E, Expert Report of Prof. Mncube, paras 111 117.
29 Bundle E, Expert Report of Prof. Mncube, paras 128.
29 Bundle E, Expert Report of Prof. Mncube, paras 128.
30 Bundle D, Witness Statement of Mr Botha para 9, Witness Statement of Mr Moosa para 54; and
Witness of Statement of Ms Heads para 40-41 and 85; Witness statement of Mr Boshoff paras 23-25;
Witness Statement of Mr Koster, para 28.,
31 Exhibit M.
13
locally for resale (i.e., those that are traded) since including the traded volumes
would lead to double counting.
[46] The merging parties asked the Commission to consider the multiplier effect of
soya beans being traded multiple times, and the principle of double counting,
meaning that because a product could be traded multiple times it could lead to
double counting. The Commission noted that if this was taken into account it
could deflate market shares and thereby, reduce the true proportion of the
market shares.
[47] Based on the relevant market definition of soya meal (being the relevant market
of concern in this transaction), we consider below the market shares submitted
by the Commission, merging parties and the economic experts in respect of the
sale of soya meal.
[48] The Commission estimated market shares in the national market and inland
market for the sale of soya meal including imports, based on the merging parties
revenue and volumes from 2017 to 2019
calculations are depicted in Table 1 and 2 below:
Table 1: Market share estimates in the national market for the sale of soya meal,
including imports
32
Revenue share estimates Volume share estimates
Company name 2017 2018 2019 2017 2018 2019
Willowton
Seaboard
RSP
Combined
Free State Oil
Nedan
Majesty
COFCO
CEOCO
32 63).
14
Bester
Wilmar
Total 100% 100% 100% 100% 100% 100%
Table 2: Market share estimates in the inland market for the sale of soya meal,
including imports 33
Revenue Share Estimates Volume Share Estimates
Company name 2017 2018 2019 2017 2018 2019
Willowton
Seaboard
RSP
Combined
Free State Oil
Nedan
Majesty
COFCO
CEOCO
Bester
Wilmar
Total 100% 100% 100% 100% 100% 100%
[49] In the Commission's view, the merging parties would be the largest competitors
in the market for the sale of soya meal and would be dominant in both the
national and inland markets. The Commission
market shares as being relevant to determining potential coordination by the JV
partners given that collectively the post-merger market shares would result in
the merging parties being presumptively dominant in terms of market shares.
The Commission concluded that the high post-merger market shares may limit
competition and cause the merging parties to align their incentives to increase
inland soya meal prices to import parity levels.34
33 7 (pages 63 64). The Commissioned noted that
the market shares in this regard may be overstated because it did not have the estimated total size of
the inland market.
34
15
[50] The merging parties, on the other hand, took into account the effect of the market
size being inflated as a result of soya meal being traded multiple times over and
considered the volumes sold for end consumption versus volumes sold for
onward trading in calculating market shares and submitted the following
estimates:35
Table 3: Market shares for the sale of soya meal in South Africa including
imports 36
Producer / Trader Estimated market share of total sales
Seaboard
Willowton
RSP
Merged Entity
COFCO
Majesty / Olam
Free State Oil
Wilmar
Nedan
Various Traders & Others
Total 100%
Table 4: Market shares for the sale of soya meal in the inland region including
imports 37
Producer / Trader Estimated market share of total sales
Seaboard
Willowton
35 Bundle B, Record, pages 124 125.
36 Bundle A, Comm
37 10 (page 54).
16
RSP
Merged Entity
COFCO
Majesty / Olam
Free State Oil
Wilmar
Nedan
Various Traders & Others
Total 100%
[51] At that point in time, both the Commission and merging parties expressed the
market shares as if the proposed transaction was a full merger between
Willowton, Seaboard and RSP, while noting that it in fact would result in a JV
between Willowton and Seaboard by virtue of their 50% acquisitions of RSP.
[52] The merging parties submitted that they had expressed the market shares
assuming that this was a full merger, merely to demonstrate that the transaction
will not substantially lessen or prevent competition given any construction of the
market share estimates.38
[53] The Commission suggested two approaches to how the market share of the JV
should be dealt with post-merger. The first approach involves considering the JV
as a separate entity independent of its shareholders and as such, its market
shares should be considered in isolation. The Commission submitted that this
approach is implausible as both Willowton and Seaboard will be actively involved
in all strategic decisions affecting the JV. The second approach involves
increasing the market shares of each of the acquiring firms slightly by the
proportion of their acquired stake in RSP.
38 -21.
17
[54] However, the Commission did not believe that there would be a material change
in the outcome reached regardless of the approach elected.
Market shares after experts meeting
[55] Prof Mncube and Ms experts, agreed to
measure market shares using volumes as a proxy with reference to SAGIS 39
data on the total market size. The market shares calculated by the expert
witnesses in the national market for soya meal including imports did not differ
materially. The expert witnesses did not consider market shares in the inland
market for soya meal.
[56] Ms Ranenyeni found that in 2019
%; Seabo
was approximately %, with the combined market share post-merger estimated
at %. 40 Hence, Ms Ranenyeni submitted that the merging parties are
significant players in the market, with high market shares that result in the
merging parties being presumptively dominant.
[57] Prof Mncube
as approximately
share was approximately %. 41 Prof Mncube did not combine the market
shares to arrive at post-merger market shares as in his view the proposed
transaction does not entail a full merger and the number of market players will
remain the same post-merger. Prof Mncube also considered market shares for
market share was approximately %,
share was RSP's
market share was 0% as it has been exclusively toll crushing for Seaboard and
its volumes were thus
42
39 South African Grain Information Service.
40 Exhibit J.
41 Exhibit R.
42 Bundle E, Expert Report of Prof. Mncube, para 166.
%; Seabo
%.
was approximately %
%.
market share was approximately %,
18
[58] Prof Mncube disagreed with the Commission
are high and therefore, firms have
findings was that the market shares are asymmetric and such asymmetry does
that the market is conducive to
coordination.43 .
Conclusion on market shares
[59] In our view there was alignment between Ms Ranenyeni and Prof Ncube's
computation on the market shares of the individual firms, the only essential
difference between them was that Ms Ranenyeni combined the three figures as
if the transaction was a full merger of the three entities, arriving at a figure of
%. However even if the RSP market share was computed as 0% by Prof
Ncube in 2020, this share should accrue to Seaboard as it had exclusive access
to the RSP soya meal. This is reflected in Prof Ncube's higher market share for
Seaboard in 2020.
[60] A more helpful way to express the post-merger market shares would be to utilise
the 2019 figures and allocate the RSP market share in 2019 as to 50% each to
Willowton and Seaboard. On the Commission's calculations the post-merger
market shares for Willowton would be % 44 and for Seaboard %. 45 On Prof
Mncube's calculations, the post-merger market shares for Willowton would be
approximately % and for Seaboard would be approximately %.
[61] In essence the only significant dispute between Ms Ranenyeni and Prof Mncube
was the size of the RSP market share accretion
for Ms Ranenyeni this was
% and for Prof Mncube this was approximately %. The accretion however in
either case is relatively small.
[62]
43 Bundle E, Expert Report of Prof. Mncube, para 183.
44 9% + 8/2 = 13%.
45 25% +8/2 = 29%.
shares for Willowton would be % %.
% and for approximately %. The accretion however in
19
THEORIES OF HARM
[63] In its recommendation the Commission concluded that the transaction between
the two largest players in the soya meal market would result in a substantial
lessening of competition in the market for soya meal and based this on the
following theories of harm:
63.1. The merging parties will likely have the ability to increase the inland
prices of soya meal through information sharing and coordination as a
result of the JV in RSP.
63.2. The proposed transaction is likely to raise foreclosure concerns in the
soya meal market, as Seaboard would have the ability and incentive to
foreclose the downstream rivals of RSP and Willlowton from accessing
imported meal post-merger.
Coordinated effects
[64] The Commission's concern regarding coordination stems from the fact that two
competitors, Willowton and Seaboard, would be joint partners in the production
of soya meal, the same product that they would continue to compete in post-
merger.
[65] The merging parties submitted that post-merger Willowton and Seaboard will
continue to operate their respective businesses independently of each other
outside of the JV and will remain competitors in the market for the supply of soya
meal.46 It is anticipated that both parties will however be jointly responsible for
the decision making at RSP, with each having the ability to appoint a director to
the board of RSP.
47 In the course of the proceedings the merging
parties clarified that
46 Bundle B, Record, page 1040.
47 Bundle B, Record, pages 787 788.
clarified that
20
In other words, Willowton and
Seaboard considered the acquisition, from their perspective, as a vertical merger
where the output of RSP would be for their own use, and not for any other third
party. Considering this, the Commission formed the view that the proposed
transaction is likely to result in the following 48
65.1. Create a structural link between Willowton and Seaboard as both parties
will be jointly involved in decision making at RSP.
65.2. The joint decision making is likely to focus on key parameters of
competition including sales and production, which both Willowton and
Seaboard will have access to through RSP. This potential exchange of
competitively sensitive information will result in Willowton and Seaboard
aligning their respective interests, and thereby have the effect of
lessening competition between Willowton and Seaboard.
65.3. The decisions taken at RSP and access to information through RSP are
likely to result in (tacit) coordination of the business decisions and
competitive behaviour of Willowton and Seaboard respectively.
[66]
of Willowton and Seaboard to be aligned. This will have the effect of lessening
competition between the firms post-merger by reducing the independent pricing
and sales of Willowton and Seaboard. In other words, the merging parties would
be more likely to coordinate their pricing in their competing businesses which
would remain outside the JV than compete independently. This means that they
would be disincentivized from pricing independently because their incentives
would be 'aligned' in RSP.
[67] In closing argument, the Commission posited that while its focus was on pricing,
it was never its contention that the likely coordination arising through the platform
created by the JV is limited to pricing issues. According to the Commission, it
48 17.5.
21
has always been its case that other forms of collusive conduct or coordination
such as market or customer allocation could occur.49
[68] The merging parties contested this finding, pointing to the characteristics of the
market and that the Commission was biased in its assessment of the merger as
it relied on the minority views that the merger should be prohibited and failed to
have any regard to evidence in support of the approval of the merger.
Foreclosure
[69] The Commission did not pursue its concern about vertical effects. Further, the
expert witnesses also agreed that there were no concerns relating to vertical
effects.50 Therefore, we do not deal with this any further.
EVALUATION
Legal principles
[70] In determining whether a merger would result in coordination in the market, we
are required to consider the effects of the merger after taking into account all the
evidence before us.51
[71] The Tribunal in Main Street held that there are two instances in which a merger
may give rise to concerns of coordinated effects. Firstly, the merger can
strengthen an existing coordination and secondly, the merger can increase the
likelihood that firms will coordinate.52 While there is no set or precise test for
determining whether co-ordination is likely, there are at least some prerequisites
to be present for the likelihood of co-ordination. In Main Street, the Tribunal held
that to sustain an argument of coordinated effects, the Commission is required
to show, as a starting point and at the very least, that the participants to a
coordinated strategy are able to
49 Transcript (6 August 2021), page 972 lines 1 -
50
51 Main Street 333 (Pty) Ltd and Kumba Resources Ltd Main Street
39.
52 Main Street, para 37.
22
(i) reach an agreement;
(ii) monitor whether the agreement is being adhered to;
(iii) punish deviation so as to make it costly; and
(iv) believe that coordination is feasible (i.e., coordination will not be feasible
if there are enough firms in the market who are not part of the
coordination, or if enough firms can enter the market to make it
unprofitable for the firms contemplating coordination.)53
[72] In Primedia,54 the Tribunal recognized that even where coordination may be a
theoretical possibility, there must be strong evidence to make out a case of
coordination.
[73] We assess below whether the evidence supports the
that the merger is likely to result in coordination.
Key market characteristics
[74] We turn first to highlighting some key characteristics of the industry to
understand the market dynamics in which the merger is taking place.
[75] Soya beans are mainly produced in the inland provinces of Mpumalanga, Kwa-
Zulu Natal, Gauteng, Free State and Limpopo, with the smaller producers of
soya beans being in the Western, Eastern, and Northern Cape provinces.55
South Africa currently relies on imports for the supply of soya meal to meet the
demand, primarily in the coastal regions.
[76] Industry predictions however are that South Africa will move to become a net
exporter of soya beans within the next five to ten years56 and pricing in the
industry could in the future move towards export parity. While the industry aims
53 Main Street, para 39
54 Primedia Limited and Capricon Capital Partners (Pty) Ltd And New Africa Investments Limited Case
number: IM013May06.
55 De A profile on the South African Soybean Market
Value Chain 2.
56 Trial Bundle D, Witness Statement of Mr Moosa page 15; Trial Bundle D, Witness Statement of Mr
Boshoff, page 82.
23
to reach net export levels, it is anticipated that imports will remain a source of
supply (in the coastal areas) going forward for the short to medium term.57
[77] Due to South Africa currently being a net importer of soya beans and meal, local
prices are influenced by global developments in those countries which are
significant producers of soya beans, such as Argentina, Brazil and the United
States, amongst others.58 Recent trends however show a steady decline in
imports, owing to an increase in the availability of locally produced soya meal as
a result of increased soya bean crop production.
[78] At the moment the supply of locally grown soya beans in South Africa is
insufficient to meet the demand of the crushing industry, and as a result, there
is significant excess crushing capacity in the market.59
[79] Pricing of soya meal is based on the pricing of soya beans which are traded on
the commodities exchange, SAFEX. The price of soya meal is based on IPP.
The formula for determining IPP is based on components which are transparent
to market participants. These components are the soya meal futures quoted on
, the Argentine premium (because most of
the imported meal is sourced from Argentina), ocean freight, USD/ZAR
exchange rate, import duty, landing costs and costs of transportation to the
ultimate destination.60
[80] All participants in the industry whether these are traders or crushers will work off
IPP as a base price, adjusted for their particular business or customer. In most
instances pricing would be IPP less a discount.
Evidence
[81] In advancing its theory of harm of coordination, the Commission considered and
provided evidence in relation to the following factors:
57 Bundle B, Record, page 1188 (BFAP Baseline Agricultural Outlook 2020 2029).
58 Ibid.
59 Ibid.
60 Exhibit R, page 14.
24
81.1. local pricing and transparency of the market;
81.2. countervailing power;
81.3. excess capacity;
81.4. barriers to entry;
81.5. market shares and market concentration;
81.6. product homogeneity;
81.7. import competition as a constraint;
81.8. third party views; and
81.9. information exchange.
[82] We deal with each of these factors in turn, and also consider whether there is of
existing coordination, and the post-merger operation of the JV and the
transaction rationale.
Local pricing and transparency of the market
[83] It is common cause that soya meal is priced at IPP less discount, with IPP being
calculated according to the components discussed above.
[84] During the Commission's investigation competitors and customers alike
submitted that the elements of IPP are transparent.61 Prof Mncube provided a
useful table, which we reproduce below, summarising the elements of IPP and
noting the transparency of each of those elements.
61
25
Table 5: IPP components, source, and transparency62
[85] The Commission's only contention was that freight costs, as one of the
components of IPP, are not necessarily transparent63 and that it is this element
which Seaboard could manipulate post-merger. However, evidence to the
contrary was presented by the various witnesses.
[86] Mr Koster testified that freight rates are publicised globally and that those in the
industry have an indication of the freight rates at a particular time. He suggested
that there have been fluctuations in freight cost as the market has changed and
this is evident from the price lists that are circulated in the industry. 64 According
to Mr Koster, this is a function of suppliers wanting to offer the best value for
money and not so much from any interest of suppliers to manipulate freight
costs.65
[87] Ms Heads testimony supported that there are indicative
freight rates that can be obtained from firms active in shipping and international
trading, and that this component of IPP was sufficiently transparent.66
62 Exhibit R, Page 14, Table 14.
63 Transcript (26 July 2021), page 160 lines 8 14.
64 Transcript (27 July 2021), page 260 lines 3 19.
65 Transcript (27 July 2021), page 260 line 20 page 261 line 6.
66 Transcript (28 July 2021), page 329 lines 11 16.
26
[88] conceded that freight costs are
clearly comparable from the daily price lists sent out by traders.67 Importantly,
Mr Botha also noted that customers in this market are able to identify any costs
that are overly excessive.68
[89] Similarly, Mr Koster also provided evidence that it would not be in the interest of
any company to manipulate freight costs as this would be easily detected.69
[90] The merging parties have submitted that freight rates are dependent on various
factors including the time for booking the freight, and whether it is a roundtrip or
not. Thus, these are objective factors which would be hard for Seaboard to
manipulate without risking being found out by customers.
[91] In another argument, the Commission submitted that because Seaboard is the
main importer of soya meal in South Africa, it is able to influence or even set
prices in the market.
[92] Seaboard frequently issues price lists with IPP pricing components to
customers.
Components in red not actively traded and as such cannot guarantee
offer
influence the price of soya meal in the country.70 According to the Commission,
an unintended consequence of this is that customers and competitors use the
price lists to benchmark their prices and negotiate similar or better prices.71 An
67 Transcript (26 July 2021), page 134 line 21 to page 135 line 5.
68 Transcript (26 July 2021), page 116 line 20 to page 117 line 1.
69 Transcript (27 July 2021), page 261, lines 4 to 6.
70
71 Exhibit D, page 16.
27
Table 6: 72
[93] This was countered by the who submitted that
the only component in red (on the price list) is the Argentina premium, which is
obviously transparent, and no one has claimed can be manipulated.73 Therefore,
any reliance on this is incorrect.
[94] The merging parties contend that Seaboard would not be able to set prices
above what market rates are as this would allow other importers, such as
COFCO, to import soya meal in South Africa
market share.74
72
73 Exhibit R, page 35.
74
28
[95] price is
based on IPP and that it is merely a reference point relied on by other players in
the market, who then discount or charge premiums on that price.75 Mr Botha
further noted that the market does not blindly follow Seaboard's pricing.76
[96] Another impact on pricing is hedging. Traders are able to hedge prices, which
allows the prices of soya meal to be locked. Ms Heads testified that hedging is
a risk borne by traders as it is possible to ultimately price lower than what the
product is worth. Seaboard is active in hedging.77
[97] Furthermore, Customers in the soya meal market are large, sophisticated
players with experience in hedging. The merging parties would need to justify
any increase on any component to these large customers. In addition, customers
can fix or hedge components of the IPP-based price. It would thus be difficult to
coordinate a price increase with Seaboard if the final price consisted of an
element of hedging by customers. Hence, the ability of Willowton and Seaboard
to set prices in an anticompetitive manner is limited.
[98]
was able to recall significant price indicators in the market with ease.78 We were
of the view that this further illustrated the transparency of pricing in the market.
[99]
influence on local pricing is minimal and that the IPP pricing mechanism is
transparent.79 In other words Seaboard is not a price setter.
[100] Thus, the evidence of all the witnesses, including that of the Commission, is that
the pricing of soya meal is highly transparent, determined by market conditions,
75 Transcript (26 July 2021) page 136 and 137, line 2, page s155, line 7 to 156, line 13.
76 Transcript (26 July 2021) page 144, lines 15 to 17.
77 Transcript (28 July 2021), page 327 line 21 page 329 line 2. See also the evidence of Mr Koster
Transcript (27 July 2021), page 257 line 13.
78 Transcript (29 July 2021), page 484 line 20 page 491 line 12.
79 Transcript (2 August 2021), page 622 lines 1-4.
29
and that it is unlikely that the merging parties could manipulate freight costs after
the proposed transaction.
[101] Another important factor that would render price coordination difficult post-
merger would be the different or asymmetric production costs of Willowton and
Seaboard.
[102] Both Ms Heads and Mr Moosa testified that the business models of Willowton
and Seaboard are different. Willowton focuses on crushing and Seaboard
focuses on imports and trading. Therefore, Willowton and Seaboard face
different costs structures. 80
[103] Ms Ranenyeni acknowledges that there is cost asymmetry, however, she is of
the view that because Willowton and Seaboard will make joint decisions on
prices and costs, the JV will remove the asymmetry between the parties. 82
Relying on international jurisprudence, 83 Ms Ranenyeni maintained that the
likelihood of coordinated effects still exists where, for example, one firm is a high-
cost firm and the other is low-cost firm. 84 Prof Mncube, on the other hand,
submitted that firms with production cost differences are unlikely to readily agree
on price coordination.
asymmetry exists in
relation to pricing
Furthermore, there is also differences in the volumes sold by each and neither
of them would know - Prof Mncube does
not agree with Ms Ranenyeni that the JV would remove the asymmetries by
virtue of Seaboard and Willowton being on the same board, as post-merger
80 Bundle D, Witness Statement of Mr Moosa para 51; and Witness Statement of Ms Heads, para 46.
81 Transcript (29 July 2021), page 465 line 11 page 462 line 8.
82 Transcript (2 August 2021), page 649 lines 7 page 650 line 2.
83 Gencor and Lawn Mower 1996.
84 Transcript (2 August 2021), page 660 lines 8 11.
-
30
Seaboard and Willowton will remain independent competitors and would not
pricing information in their competing businesses.
[104] We are of the view that there was insufficient evidence provided by the
Commission to support the theory that these two firms, with their different cost
structures, but equally importantly, in light of their different pricing structures,
were likely to coordinate their prices.
[105] With regards to the argument on predation, Ms Heads testified that it would not
be economically viable for Seaboard to drop prices with the intention of driving
competitors out of the market and then raising prices again because firstly,
Seaboard will lose money from cutting prices and secondly, the crushing
facilities of the party exciting the market will remain in the market which could be
purchased by other competitors.85 Prof Mncube submitted that this would not be
rational and would further be self-defeating, considering the number of large
firms in the market, some supported by multinational firms, that could and would
move in, in the case of any increases in prices.86 We find no reason to disagree,
considering that the evidence has pointed to multiple players in the market and
their ability to be able to also import soya meal into South Africa.
Countervailing power
[106] We turn next to the countervailing power of customers.
[107] In making its recommendation for prohibition the Commission relied extensively
on the fact that customers, albeit large and sophisticated, will have their
countervailing power weakened post-merger.
[108] Despite the evidence to the contrary, the Commission argues that although
customers appear to have countervailing power pre-merger, the market
conditions post-merger will reduce the countervailing power. According to the
countervailing power is over local
85 Transcript (28 July 2021), page 331 line 8 page 332 line 3.
86 Bundle E, Expert Report of Prof. Mncube, para 266.
31
producers of soya meal, and so when customers are unhappy with local
suppliers, they tend to switch to imported soya meal. Therefore, countervailing
power tends to push customers to Seaboard as the predominant importer of
soya meal and because there are no credible alternatives to switch to.87
[109] This is one of the areas the merging parties argue that the Commission suffered
a biased selection of the evidence from customers,88 as the evidence in the
record and that of witnesses who testified before the Tribunal contradicts the
proposition.
[110] The Commission had sent a number of queries to customers during its
investigation. Most customers raised no concerns with the transaction, claiming
that they were able to switch between suppliers of soya meal when the need
arises; local soya meal could be substituted with imported soya meal on short
notice; and the local crushing industry together with imports from Zambia and
Argentina is adequate to supply competitively priced soya bean products in
South Africa.89
[111] Evidence from customers shows that switching can happen from a local supplier
to another local supplier, from an importer to a local supplier; and from a local
supplier to an importer. Several feed manufacturers submitted that they have
switched between Seaboard and COFCO.90 Another feed manufacturer
submitted that it sourced soya meal from and importer and a local supplier due
to cost competitiveness, and because having more than one supplier helps it to
spread the risk.91
[112] Mr Botha also spoke to the ability of other firms to import soya meal into the
country, which could possibly constrain Seaboard from charging higher prices
above IPP.92 Ms Heads testified that COFCO, is one such firm that directly
87
88 Exhibit R, page 23.
89 Bundle B, Record, pages 1476 1477.
90 Bundle B, Record, page 1476-1477 and 1565.
91 Bundle B, Record, page 1565.
92 Transcript (26 July 2021), page 160 line 21 page 161 line 12.
32
imports soya meal to South Africa and in other instances, imports indirectly
through other players such as Seaboard.93 There have also been imports
coming from Zambia, from Wilmar SA and Bester.94 Therefore, customers
evidently have a choice between different suppliers of imported soya meal.
[113] L
the question of the reactiveness of customers thereto. Mr Koster noted that
customers rely on the reliability of the price of the soya meal supplied,95
accordingly customers will negotiate the discount or local premium with the
supplier to obtain a suitable price.96
testimony which suggested that customers have a keen eye on prices in the
market.97
[114] Moreover, the Commission's own findings in its record do not support this
proposition. The Commission found that customers of the merging parties are
well established feed manufacturers and that the market is characterised by
short term contracts and the ability of customers to negotiate the terms of these
contracts.98
submissions made by customers, notes that customers (i) have the ability to
negotiate prices and discounts with suppliers, (ii) can play suppliers off against
each other to get the best price possible, (iii) do switch between different
suppliers from time to time with price and product quality being the main drivers
for the switching, and (iv) contracts entered into between suppliers and
customers act as a barrier to switching only for a limited period, 3-12 months,
but they are generally not exclusive and can be terminated upon expiry.99
[115] In other words, the Commission was unable support its own proposition that
switching would be more difficult post-merger and that customers' countervailing
power would accordingly be weakened.
93 Transcript (28 July 2021), page 334.
94 Bundle B, Record, page 2227.
95 Transcript (27 July 2021), page 252 lines 14 16.
96 Transcript (26 July 2021), page 32 lines 4 7; Transcript (27 July 2021), page 258 lines 17 20.
97 Transcript (27 July 2021), page 210 line 16 page 211 line 10.
98 Bund 220.
99
33
[116] However, Ms Ranenyeni persisted with the view that because the final price of
soya meal to customers is uncertain, customers would not necessarily be able
to interrogate it in the manner suggested by Mr Koster.100 This was because, on
her version, the JV in RSP will enable Willowton and Seaboard to agree on the
discount to be given to customers.101 This argument fails to consider the
evidence given that a price increase by Seaboard or Willowton post-merger
would not be sustainable, mostly due to the ability of customers to react to any
price changes.
[117] Mr Boshoff indicated that feed companies are knowledgeable about the trends
modelling.102 Therefore, should a
supplier of soya meal act in a manner not consistent with the analysis
undertaken by the feed companies, this would raise a concern.
[118] own witness, Mr Botha, conceded that should the merging
parties increase prices, customers would be able to switch to an alternative
supplier and that this will remain the same post-merger.103
[119] Another argument advanced by the Commission was that the proposed
transaction reduced the number of local producers that customers can consider
when making a decision to switch as RSP will no longer be in the market.104
However, as pointed out by Prof Mncube,105 this proposition is incorrect because
pre-merger RSP is not in the market due to the toll crushing agreement and the
merger therefore does not reduce the number of local competitors that
customers can choose from.
100 Exhibit D, page 16.
101
102 Transcript (27 July 2021), page 210 line 16 page 211 line 10.
103 Transcript (26 July 2021), page 105 lines 3 9.
104
105 Exhibit R, page 42.
34
[120] Thus, the evidence of customers obtained during the Commission's
investigation, the evidence of witnesses at the Tribunal, including the
Commission's witness and the Commission's own description of customers in its
that the countervailing power of customers would be reduced or otherwise
change post-merger.
Excess capacity
[121] Another factor relied on by the merging parties as a competitive constraint is the
significant excess crushing capacity in the market.
[122] During its investigation, the Commission found that crushers are currently not
fully utilizing their installed capacity and as such have excess crushing capacity.
The Commission calculated the following capacity utilization in the market for
the crushing of soya beans:
Table 7: Capacity utilization in the crushing market -106
Installed
Capacity
Share of
Installed
Capacity
Estimated
Crush
Share of
Crushed
soybeans
Capacity
Utilization
Willowton
RSP
Merged entity
Free State Oils
Nedan
Cofco
Majesty Oils
Wilmar
GOCM107
Drak Oil
Ceoco
Other
106
107 Gauteng Oil and Cake Mills.
35
Total
[123] The table above shows that the average crush capacity utilization stands at
approximately 63%. Therefore, the Commission concedes that there is excess
crushing capacity in the market but contends it is neither a constraint nor a
potential constraint, as its utility depends on the availability of soya beans. 108
The Commission found that there is a shortage of soya beans which is likely to
prevail in the foreseeable future, and for this reason excess capacity is
meaningless.
[124] While downplaying the role of excess capacity as a constraint on the merging
parties, the Commission at the same time argues that the excess capacity in the
109
The Commission does not however explain how such a strategy would be
sustainable in the context of competitors' having excess capacity. The
capacity on the one hand and how competitors could utilise their excess capacity
on the other hand is unexplained.
[125] confirmed that the average capacity
utilisation is approximately 62%. Ms Ranenyeni further considered the evidence
of Ms Heads and Mr Moosa that crushers produce at optimal crush capacity
which is approximately 80% of nameplate capacity, and calculated excess
capacity as follows:
108 105.
109
36
Table 8: Excess capacity based on optimal crush capacity -110
[126] According to Ms Ranenyeni, the biggest competitors of the merging parties,
COFCO and Wilmar do not have excess capacity
capacity is in the hands of small players. However, Ms Ranenyeni considers
some of the small players Majesty, GOCM and Drak Oil as being too weak
or small to have an impact on the merging parties, and that the only real
competitor is Nedan, but it too would not be able to constrain the merging
parties.111
[127] The
by Prof Mncube who found, based on his assessment,
capacity was completely removed from the industry, there would still be 30%
excess capacity remaining in the market.112 We include below a table reflecting
the excess capacity in the market as calculated by Prof Mncube.
110 Exhibit F.
111 Exhibit F,
112 Bundle E, Expert Report of Prof. Mncube, para 340.
37
Table 9: Excess capacity excluding RSP113
[128] Based on the above, he concludes that it is evident that the impact of the removal
.114
[129] Under cross examination, Mr Botha, the Commission's witness, conceded that
there would be a substantial amount of crush capacity in the market, even if
115 Thus,
confirming the views expressed by Prof Mncube above.
[130] Further, Mr Moosa testified that excess capacity in the market causes suppliers
to price aggressively in attempts to maximise the utilisation of their capacity. 116
Therefore, it is likely that excess capacity is a competitive constraint in this
market.
[131] In interpreting the Commissions calculations depicted in Table 5 above, Prof
Mncube notes that, if anything, the table highlights how large competitors are
for example, COFCO significantly larger than any player in the market,
including Willowton and RSP, separately and together. 117 He also refutes the
competitors would not be able to timeously utilize
113 Bundle E, Expert Report of Prof. Mncube, Table 19 (page 218).
114 Bundle E, Expert Report of Prof. Mncube, para 340.
115 Transcript (26 July 2021), page 123 lines 9 13.
116 Transcript (29 July 2021), page 446 lines 12 16.
117 Bundle E, Expert Report of Prof. Mncube, para 255.
38
their excess capacity in response to a price increase because there is a shortage
of soya beans in the country. In doing this, he refers to the BFAP118 report which
states, in relation to soya beans, that South Africa for the outlook period (2020 -
2029) is expected to trade close to self-sufficiency, following a projected area
expansion of over 150 000 hectares in 2021.119
[132] Furthermore, Prof Mncube submitted that economic theory acknowledges that
excess capacity makes collusion difficult when there are many firms in the
market.120 We agree with this submission.
[133] In the instant case the Commission, while acknowledging that there was excess
capacity in the market, was unable to demonstrate how post-merger collusion
between the acquiring firms could be sustained notwithstanding significant
excess capacity in the market.
Barriers to entry
[134] The Commission and its expert witness concluded that barriers to entry are high
in the soya meal market and that entry into the market is not likely to occur in
the near future. Further, even if entry was to occur, it would be unlikely to have
a timely effect on competition in the industry nor is it likely to be sufficient to
constrain any anticompetitive behaviour by the parties post-merger.121
[135] The Commission relied upon the evidence of the various competitors it
contacted during its investigation that establishing a crushing plant or facility may
range anywhere between R100 million to over R1 billion. It was submitted that a
crushing plant with a crushing capacity of 600 metric tons per day may cost
approximately R200 million to establish, while a crushing plant with capacity of
between 250 000 to 500 000 metric tons may cost over R1 billion to establish.122
118 Burea for Food and Agricultural Policy
119 Bundle B, Record, page 1188 (BFAP Baseline Agricultural Outlook 2020 2029)
120 Exhibit R, page 25.
121
122
39
[136] During evidence, Mr Botha submitted that constructing a new crushing plant,
which would have a crushing capacity of 1 000 metric tons per day, would cost
in the region of R525 million.123 Mr Moosa disputed this amount and submitted
that a new plant would not cost more than R300 million for a plant that crushes
approximately 750 metric tons per day, and not more than R400 million for a
plant that crushes approximately 1000 metric tons of soya beans per day.124
[137] Mr Botha's estimates were based on the exact cost incurred by Wilmar when it
reconstructed the crushing facility in Randfontein after the explosion in 2017.
However, he noted Wilmar had some existing infrastructure from the previous
crushing plant when they commenced reconstructions.125
[138] From the above, we can accept that establishing a new plant could cost anything
between R300 million and R500 million. The cost of the plant would certainly be
affected by the size of its nameplate capacity.
[139] Evidence by Mr Moosa shows that an increase in soya bean harvests has
resulted in the expansion of crushing facilities in South Africa in the last year and
as discussed above, there is excess crush capacity.126 This could also be taken
to that the insufficient supply of soya
beans in South Africa would be the main barrier to entry.
[140] In relation to entry into the market by new entrants, particularly crushers, Mr
Botha submitted that it would take approximately two to five years to enter the
market and become an effective competitor. This was consistent with the views
of other competitors who submitted to the Commission that it would take
anything between three to five years for a new crushing plant to operate
optimally in the market.127 A single competitor submitted to the Commission that
123 Bundle D, Witness Statement of Mr Botha, para 17 (page 8).
124 Transcript (29 July 2021), page 443 page 17 page 444 line 1.
125 Transcript (26 July 2021), page 32 line 19 page 33 line 10.
125 Transcript (26 July 2021), page 32 line 19 page 33 line 10.
126 Bundle D, Witness statement of Mr Moosa, pages 21 23.
127 Bundle B, Record, pages 1802, 1911, and 2228.
40
it would take 18 to 24 months for a new entrant to become an effective
competitor in the market.128
[141] Later in his evidence, Mr Botha conceded that while barriers to entry in crushing
capacity are relatively high, they are not insurmountable. He acknowledged the
recent entry of GOCM and the increase in crush capacity by existing players in
the market; and also accepted that entry can occur through the purchase of an
existing crush plant.129 Similarly, Mr Moosa testified to the recent entry into the
crushing market by GOCM (in the first half of 2020), and that of COFCO in the
last three years.130 The merging parties further noted the recent entries of
Wilmar in late 2019 and CEOCO in the second half of 2020.131
[142] While competitors have intimated that greenfield entry may be on the high side,
the merging parties and their expert witness have noted that it is possible for
entry to occur quickly and inexpensively through the purchase of an existing
crushing plant.132 When this proposition was put to Mr Botha, he submitted that
while he does not know whether or not it would be cheap, it is indeed possible
to enter the market through the purchase of an existing crushing plant.133
[143] Even if we are to accept that barriers to entry in greenfield crushing plants are
relatively high, the available evidence suggests that there are low barriers to
entry as a trader in the soya meal market.
[144] Ms Heads testified that the market for the trading of locally produced or imported
soya meal in South Africa is a free and open market, which allows any firm to
trade soya meal at any point in time, and that it would take a new entrant less
than a month to enter the market. Further, there are no significant investment
requirements when entering the market.134 Consequently, the barriers to entry
in the trader market are relatively low.
128 Bundle B, Record, page 1743.
129 Transcript (26 July 2021), pages 132 134.
130 Transcript (29 July 2021), page 580 line 17 page 581 line 1.
130 Transcript (29 July 2021), page 580 line 17 page 581 line 1.
131 Merging Parties Heads of Argument, para 119.
132 Exhibit R, page 18.
133 Transcript (27 July 2021), page 133 lines 15 20.
134 Bundle D, Witness statement of Ms Heads, page 65.
41
[145] Taken together, the evidence above s
that barriers to entry are so high in the soya meal market as to render post-
merger collusion sustainable. In our view the evidence suggests that were the
merging parties to embark on a post-merger collusive strategy, the ease of entry
for traders together with the other market characteristics as explained, would
render such strategy unsustainable.
Market shares and concentration
[146] The Commission reiterated its view that the combined market share of the
merging parties is too high to be ignored in the context of the JV in RSP as this
supports its argument regarding the impact of any coordination between the
merging parties.135
[147] Based on its market share calculations, the Commission concluded that the
market is highly concentrated as in 2019 the largest four competitors made up
66% of the market, and the merging parties are is a significant
portion of that.136 Ms Ranenyeni, calculated
the following concentration ratios in the soya meal industry and found the
following:
147.1. per the two-firm concentration ratio, the two largest firms in the industry
controlled 51% of the market in terms of revenue and volume market
shares and 36% of the market in terms of capacity market shares;
147.2. per the four-firm concentration ratio, the four largest firms in the industry
controlled 70% of the market in terms of revenue market share, 71% of
the market in terms of volume market shares and 59% of the market in
terms of capacity market shares; and
147.3. per the eight-firm concentration ratio the eight largest firms in the
industry controlled 97% of the market in terms of revenue and volume
market shares and 94% of the market in terms of capacity market
shares.
135
136
42
[148] According to Ms Ranenyeni, international guidelines interpret the four-firm
concentration ratio of over 70% as being highly concentrated. 137 Hence, she
concluded that the high concentration levels combined with the collective
revenue and volume market share of the merging parties of over 45% and the
high post-merger capacity market shares, suggests that the proposed
transaction may have a wide impact as a result of the likely anticompetitive
effects stemming therefrom.138
[149] Prof Mncube, on the other hand, found that given the number of players in the
market and the asymmetric markets shares of these players, it can hardly be
said that the market is highly concentrated. He found that there were at least 10
(ten) players active in the soya meal market, with asymmetric market shares as
follows:
Table 10: market participants and market shares in the national soya meal
market139
[150] Based on the above, Prof Mncube calculated the HHI140 in the soya meal market
to be 1,565 based on the 2019 national market for soya meal. Relying on
international horizontal merger guidelines, he concluded that the pre-merger, the
market is close to 1, 500 which is indicative of a competitive marketplace. He
137 Exhibit D, page 14.
138 Bund 351.
139 Bundle E, Expert Report of Prof. Mncube, Table 7.
140 -
is the sum of the squares of the market shares.
43
does not consider the market shares as a collective to determine the HHI post-
merger as this is not a full merger.141 Prof Mncube cautioned that concentration
indexes (for example, HHI) may be misleading as in some cases HHI increases
as asymmetries of market share increases.142
[151] We agree with the proposition put forward by Prof Mncube that firms with
asymmetric market shares, read with the other market characteristics such as
significant excess crushing capacity and potential imports in the case of a
hypothetical small but significant post-merger price increase by the merging
parties, are unlikely to have incentives to collude.
Product Homogeneity
[152] The Commission argued that the proposed transaction involves homogenous
products, in respect of soya meal and hulls, which increases the likelihood of
coordination.
[153] However, the merging parties and competitors of the merging parties have
provided evidence that soya meal is a heterogenous product, as a result of the
quality difference.143
[154] To determine the quality of soya meal, there are minimum product specifications
that must be met in terms of the urease levels. Three characteristics differentiate
one product from another namely protein, fibre, and moisture levels.144 The
merging parties provided in the table below, the quality parameters that have to
be met to produce quality soya meal:
141 Bundle E, Expert Report of Prof. Mncube, para 241.
142 Bundle E, Expert Report of Prof. Mncube, para 240 - 241.
143 See Table 15 in Bundle E, Expert Report of Prof. Mncube, page 195.
144 Bundle D, Witness Statement of Mr Moosa, 28 June 2021, para 43.
44
Table 11: soya meal quality parameters145
[155] We understand that these characteristics also influence branding of the soya
meal, which in turn inf
product.
[156] Mr Boshoff testified that local soya meal was now of similar quality to imported
meal,146 and that quality was important for feed manufacturers especially in the
poultry industry.147 Evidence by Mr Koster also showed that feed manufacturers
in the poultry industry are sensitive to the quality and consistency of soya meal
as it has a significant impact on the digestive system of the chicken and may
affect its growth and mortality.148 Mr Koster explained that urease levels have to
be at an acceptable level from a nutritional point of view. Urease level of over
0.3% would be too high.149 Hence, it would not be suitable for animal
consumption.
[157] Furthermore, Mr Koster
of production (for example, from soya meal to sunflower) would affect the quality
of the product. He stated that a plant that runs on a short basis and has to start
continuously and switch from one product to another, would make it difficult for
stability of production to be established and therefore, impacts on the
consistency of the quality.150 Importantly, Mr Koster testified that customers are
also willing to pay an extra premium for high quality soya meal.151
145 Bundle B, Record, page 640.
146 Transcript (27 July 2021), page 203 lines 7 10.
147 Transcript (27 July 2021), page 204 lines 4 12.
148 Transcript (27 July 2021), page 255 lines 8 16.
149 Transcript (27 July 2021), page 253 lines 8 14.
150 Transcript (27 July 2021), page 254 lines 4 13.
151 Bundle D, Witness statement of Mr Koster, 29 June 2021, para 19-25.
45
[158] What was clear from the evidence before us, was that the quality of soya meal
is important. and it differentiates the soya meal supplied in the market by the
various competitors. However, there was no factual evidence showing that any
of the current players could not supply soya meal of the desired quality to its
customers in South Africa.152
Import competition as a constraint
[159] While the Commission's theory, based on the views of several competitors was
that other players in the market could respond by importing soya meal (given
that local supply did not meet the actual demand) if there were coordinated price
increases post-merger, such import competition would serve to exercise a
limited constraint on the merging parties' incentives.153 This is due to the lack of
economies of scale, which prevent other players from economically importing
soya meal from Argentina.154 According to the Co importing
is not viable where a firm does not have enough volume to fill up a vessel the
minimum volume a firm can import is 30 000 tons. The decline in imported soya
meal along with the increase in local capacity has ensured that Seaboard
remains the incumbent, considering the critical mass it has through its existing
business. Competitors submitted that firms like COFCO (which is the largest in
the market) and Bester are not effective importers of soya meal.155 It was the
s contention throughout that Seaboard, unlike other
firms, is uniquely positioned to import soya meal.156
[160] Based on the evidence of Mr Botha when asked whether other market
participants would import if Seaboard were to increase prices after the proposed
transaction, the Commission concluded that once these competitors started
importing as a response to a hypothetical post-merger price increase, Seaboard
could then simply drop its prices, which would then deter any potential importer
from bringing imports into the country. Thereafter, Seaboard would revert to its
from bringing imports into the country. Thereafter, Seaboard would revert to its
152 Transcript (27 July 2021), page 203 line 7 10; page 207 lines 12 19.
153 167.
154
155
Report, para 242.
156 173.
46
high prices again, which would be to the detriment of competitors and in turn,
customers.157
[161] The evidence before us however seems to suggest the contrary.
[162] In the first instance, while import competition is still highly relevant and imports
make up the shortfall for local demand, import volumes are on a decline due to
the fact that the local supply of soya bean is on the increase.
[163] Furthermore, as mentioned above, Seaboard is not the only experienced
importer of soya meal. COFCO, being a large multi-national, has a crushing
plant in Argentina and is experienced in importing meal. Other players active in
South Africa - Glencore, Bunge, and Louis Dreyfus Company also own
crushing plants in Argentina.158 FR Waring is part of a large multi-national which
has a crushing plant in Argentina and has the experience and financial muscle
to retaliate timeously to any price increases. These large players could import
soya meal very easily if such a need arose.
[164] Looking at
competitors turning to imports, we find that while he noted that it is not so simple
for competitors to do so because of factors such as storage and existing
relationships, he effectively conceded that Seaboard would revert to normal
competitive prices should other players import into South Africa. Hence,
(potential) import competition would be a constraining factor on Seaboard.159
[165] When the panel asked Ms Ranenyeni whether customers had been asked
whether in the event of a small but significant price increase by the merging
parties after the proposed transaction, increased imports of soya meal would
take place, she could not recall if this was done.160 We find that the
.
157 August 2021), para 78.4.
158 Bundle D, Witness Statement of Ms Heads, para 52
159 Transcript (26 July 2021), page 145, lines 8 21.
160 Transcript (2 August 2021), page 766 line 10 page 767 line 7.
47
[166] The panel put to Ms Ranenyeni whether transport costs would be prohibitive for
importers in the regional coast to transport soya meal inland if prices of soya
meal in the inland region were increased after the proposed transaction. She
agreed that the transport costs would not be prohibitive as evidence already
indicates that half of the coastal imports fare transported inland. Her earlier
evidence had shown that transport costs are also not prohibitive from inland to
the coast.161 This evidence of the Commission's expert supports the inference
that the threat of imports would exert a constraint on any post-merger price
increases, notwithstanding whether the price increase occurs inland or in the
coast.
[167] In relation to the issue of scale, if a Tier One customer that requires a consistent
supply of soya meal were to approach the likes of COFCO in the case of
increased prices by the merging parties, this in our view could remove the
impediment of scale (raised by the Commission) and allow others to import into
South Africa. We do not find this to be an unlikely scenario considering the size
of customers in the market and their requirement for quality and consistent soya
meal. own investigation revealed that at least two feed
manufacturers have switched between importers.162 Ms Ranenyeni effectively
conceded that an importer could have sufficient scale if in a hypothetical
situation one large customer, which can switch as indicated above, requires
supply for a whole year from that one importer.163
[168] As to whether increased import competition itself was a competition concern,
this was not fully explained by the Commission.
seemed to be that Seaboard is the incumbent in the import market and cannot
be constrained by other players that may import into South Africa.
[169] However, the evidence suggests that although imports are on the decline they
are still necessary to meet the demand for soya meal in the country, Seaboard
are still necessary to meet the demand for soya meal in the country, Seaboard
161 Transcript (2 August 2021), page 774 lines 2 14.
162 Bundle B, Record, pages 1476-1477, paragraph.28 (sub paragraph.28.1 & 28.1.1.); page 1565 item
25.
163 Transcript (2 August 2021), page 775 lines 1 11.
48
is not the only importer, and there are currently a number of international players
in the market who can import.164 .
[170] One of the competitors that had submitted concerns withdrew its objection to the
merger shortly before the hearings were set to commence. The reasons cited
were, among others, that there has been a significant increase in the soya bean
crop for the 2020/21 period which reduced the demand for imported soya meal.
But as we indicated above, it is evident that notwithstanding, imports will remain
relevant for the foreseeable future and may act a sufficient constraint on the
conduct of the merging parties post-merger.
Other third-party views
[171] The
and AFMA, which represents feed manufacturers who are also customers of the
merging parties, were notable.
[172] NAMC submitted that mergers of this kind contribute to cost efficiency gains due
to economies of scale and this is important for maintaining fair competition.165
[173] AFMA submitted that the proposed transaction is unlikely to have a negative
effect on the soya meal market as the merger could lead to optimization of
capacity and increase efficiencies.166 According to AFMA, there are sufficient
players in the market and hence, the soya bean industry in its entirety is likely to
remain competitive post-transaction. Further, the proposed transaction will give
customers of soya meal assurance of supply.167 Customers, through AFMA,
anticipate that the merger will have a positive impact on the market.168
164 109.
165 257.
166
167 Ibid.
168 Transcript (27 July 2021), page 211 line 14 page 212 line 8.
49
[174] During the hearing, Mr Boshoff from AFMA confirmed that suppliers themselves
are able to switch and obtain their supply from a competitor, which they then on-
sell to their own customers.169
[175] During its investigation the Commission had reached out to competitors of the
merging parties and a few of them submitted concerns regarding the proposed
transaction. We highlight here the concerns at the level of principle rather than
attribute them to particular customers. One concern was that the proposed
transaction would result in the merged entity having high market shares,
approximately 45% - 50%, in the soya meal market and as a result would be
able to heavily discount soya meal prices, which would in turn force smaller
players out of the market.170 Further, the merged entity may foreclose
participants from accessing toll agreements, which are an important feature of
the soya meal market.171 Another concern was that due to Seaboard being the
largest importer of soya meal in South Africa, the proposed transaction would
place the merged entity in a position to act anti-competitively, and this would be
to the detriment of smaller players, who may be foreclosed post-merger.172
[176] When these concerns were put to the merging parties, they asked us to take into
account the fact that most of the competitors at that stage were under the
impression that this was a full merger between Willowton and Seaboard, and
these concerns should be assessed in that light. Furthermore, customers had
indicated that they were able to switch to other suppliers. As indicated above,
the Commission itself had conceded that the market was competitive, and that
pricing was relatively transparent. In such market dynamics, it would be difficult
for the merging parties to sustain coordination to any appreciable level of price
increases or time.
[177] As to the possible anti-competitive actions that the merging parties were likely
[177] As to the possible anti-competitive actions that the merging parties were likely
to engage in, this was posited as a significant price increase in the local market
169 Transcript (27 July 2021), page 207 lines 10 19.
170 108.
171 Ibid.
172 Ibid.
50
which was canvassed at some length during the hearing and for which no
evidence was led.
[178] Finally, we note that most of the customers did not raise concerns with the
transaction. Only one customer expressed a concern that the proposed
transaction is likely to result in Willowton being the largest investor in the oilseed
crushing market in South Africa and will use the soya oil produced by RSP
internally, thereby reducing the availability of locally produced crude soya oil in
the market.173 However this concern was related to the soya oil market which is
not relevant in the case before us.
Possibility of existing coordination
[179] As mentioned above, Seaboard has a 30% shareholding in RSP. In addition,
RSP started toll crushing for Seaboard from and does not provide any
crushing services to third parties.
[180]
coordination may have existed pre-
involvement in RSP and that this restrictive practice would now be extended to
include Willowton. 174 Ms Ranenyeni based her assertion on the fact that
Seaboard has access to the commercially sensitive information of RSPs sales
and marketing information, and that it will continue to have access to this
information post-merger.
[181] However, no evidence was adduced showing existing coordination.
[182] It was not clarified by the Commission what type of 'restrictive practice' the toll
agreement constituted, nor could it point to any evidence that the toll agreement
resulted in higher prices to customers after it was entered into. There was also
no evidence from customers that the toll agreement had somehow resulted in
higher prices or reduced quality or volumes of soya meal.
173 Bundle B, Record, page 1534 1535.
174 Bundle E, Expert Report, para 333 334.
51
[183] We turn now to consider other factors to assess whether the merger will lead to
some other form of co-ordination as suggested by the Commission.
The post-merger operation of the JV and the transaction rationale
[184] At the time the merger was filed with the Commission,
175
[185] When Ms Heads testified, she was of the view that
177 However, Ms Heads cautioned that this was still
subject to further negotiations, as the merging parties were cautious about
discussing such detail due to the legal advice that had been given to them at the
time.178
[186] Mr Moosa testified that when the transaction was first contemplated by the
parties,
175 Bundle B, Record, pages 765 and 1004.
176 Transcript (28 July 2021), page 301 line 9 - 19; page 374 lines 9 17.
177 Transcript (28 July 2021), page 375 line 1.
178 Transcript (28 July 2021), page 372 lines 13 - 15; page 374 lines 2 6.
179 Transcript (29 July 2021), page 472 line 12 page 473 line 1.
180 Transcript (29 July 2021), page 473 lines 1 5.
52
This was in fact
the pre-merger arrangement, albeit with some slight differences,
[187] The Commission submitted that the evidence of the merging parties contradicted
the rationale put up by each of the merging parties at the time the merger was
1
81
that
[188] Noting the change in rationale for the proposed transaction, we called on the
merging parties to provide us with certainty on how they saw the relationship
unfolding post-merger as this has a material influence on the evaluation of the
merger.182 Consequently, the merging parties submitted a draft Heads of
Agreement which would regulate their post-merger relationship in among others
the following terms
188.1. RSP will only sell its products to Wilowton and Seaboard, and will not
sell or market soya meal to third parties in competition with Willowton
and Seaboard;
188.2.
188.3.
181 Such as RCL Foods Limited, Country Bird Holdings (Pty) Ltd and Astral Foods Limited.
182 Transcript (29 July 2021), page 560 line 1 page 561 line 9.
53
;
188.4.
188.5. and
188.6. .183
[189] In the conditions tendered by the merging parties, the merging parties included
a clause to regulate the abovementioned structural undertakings. However, we
did not deem it necessary to impose these structural undertakings as conditions
to the merger. We discuss this in detail under the Remedies section.
Information exchange
[190] The Commission advanced an argument that competition between Seaboard
and Willowton would be restricted, and this could occur as a result of not only
coordination, but also information sharing between the parties.184
[191] The Commission considered the fact that
In addition, in terms of the shareholders agreement,
Furthermore,
It is on this basis that the Commission argues that
the JV creates a platform for the sharing of commercially sensitive information
between Seaboard and Willowton, and that through making decisions at RSP,
Seaboard and Willowton will align their competitive behaviour such that
183 15.
184
;
54
competition between them will be restricted post-merger. 185 According to the
Commission, this could result in increases in prices and predatory pricing by the
merging parties, as discussed above.
[192]
submitting that due to the existing vertical relationship between Seaboard and
Willowton, as indicated above, the merger does not create an opportunity for
them to share information. Furthermore, industry information on market trends,
such as pricing, is publicly available and so the merger does not allow the parties
to have information that they currently do not have. 186 We pause here to note
the ease with which Mr Moosa spoke of market trends during his testimony,
showing the transparency of the market.
[193] The merging parties also testified that reputation is important in the market, and
they would not risk it by engaging in anti-competitive behaviour.187
[194] During the hearing, the merging parties clarified that,
188 Furthermore, as
This is something the
eady accepted in her report.189
[195] while her
submissions were confusing in this regard, the crux of her contention was that
while Mr Moosa may not know the retail price (i.e., the ultimate price to
customers), it is the wholesale price (i.e., the direct input cost of soya meal at
RSP) that is relevant and may be used to soften competition between the
185 248.
186 Bundle D, Witness Statement of Mr Moosa, pages 26 27.
187 Transcript (29 July 2021), page 453 line 18 page 454 line 18.
188 Transcript, (29 July 2021) page 494, lines 1 to 4
189 Bundle E, Ms Ranenyeni . Para 333.
55
parties.190 Despite, Mr Moosa testifying that he 99% accuracy the
direct input costs of RSP and that he does not need to sit on the board of RSP
to work out the costs",191 Ms Ranenyeni still maintained that Mr Moosa would
not know the exact cost and therefore, there is scope for coordination. It is at
this point that she submitted that the coordination does not have to focus on
pricing, the information obtained may be used to coordinate in various other
ways, such as customer allocation.192
[196] Prof Mncube testified that firstly, while the parties may know the wholesale price,
they do not know the final price to customers, and secondly the market is clearly
transparent as a lot of the information the Commission perceives to be
commercially sensitive is already in the hands of market participants pre-merger.
Ultimately, the final price is where competition takes place, and the impact that
hedging may have on the ultimate price settled on by Seaboard, is information
that unknown to Willowton.193
[197] We have heard, as detailed above, the testimony of the various factual
witnesses regarding the transparency of the market and the information that is
available in the market. As discussed above, Ms Ranenyeni had also conceded
to the market being transparent.
[198] The fact that Seaboard and Willowton have different pricing structures and the
final price to customers may not be transparent does not, as the Commission
alleges, support the likelihood of co-ordination. What it does support is the
contrary coordination would be unlikely if the final prices to each of their
customers are not known or predictable, as in the case of those large customers
that would elect to hedge components of their price.
190 Transcript (2 August 2021) page 738 lines 9 16.
191 Transcript (29 July 2021) page 489, line 16 to page 489, line 7.
192 Transcript (2 August 2021), page 732 lines 3 17.
193 Transcript (3 August 2021), page 934 line 16 page 936 line 10.
56
Conclusion on SLC analysis
[199] the soya meal market will remain competitive
post-merger,194 and customers have indicated they will enjoy countervailing
power through their ability to switch to different suppliers in the event of any post-
merger price increases.
[200] In summary, the evidence before us was that
200.1. while soya beans are traded as a commodity on SAFEX, soya meal is
characterised by certain quality differences and specifications. Quality is
important and customers, especially in the animal feed industry, are
willing to pay a premium for high quality meal;
200.2. prices in the industry are based on IPP the components of which are
transparent;
200.3. customers are large, sophisticated players who could influence the final
price paid by them by hedging components of the IPP-based price;
200.4. there is significant excess crushing capacity in the market and
competitors of the merging parties consist of both large and small
players;
200.5. import competition is still a significant part of the market because it is
more cost effective to supply the coastal region with imports rather than
inland soya meal due to hight transport costs;
200.6. the Commission produced no evidence showing that (potential) imports
of soya meal would not take place in the event of a post-merger
hypothetical small but significant price increase by the merging parties;
200.7.
increase by 40% in the period 2020/21, which is estimated at 1.7million
tonnes;195
200.8. while there are barriers to entry particularly in greenfield crushing plants,
there appears to be limited barriers in relation to entry as a trader in the
soya meal market;
194 Transcript (26 July 2021), page 122; Transcript (2 August 2021), page 724.
195 World Grain.com, available on: https://www.world-grain.com/articles/15118-south-africa-expects-
record-soybean-crop.
57
200.9. there are a number of competitors in the market, including large players
who are part of multi-nationals with experience in importing soya meal
and the financial resources to respond to any likely co-ordinated strategy
embarked upon by the merging parties.
[201] No evidence of the history of collusion in the soya meal market was presented
although this seems to have been a concern of the Commission.
[202] We see thus that the evidence obtained in the course of the Commission's
investigation and from the witnesses who testified at the Tribunal does not
support the Commission's theory of post-merger coordination. On the contrary,
the evidence thus far supports the likelihood that were the merger parties to
engage in such co-ordination they would not be able to sustain it at all or for any
significant period of time.
[203] In our view, the Commission, while being concerned about a JV between
competitors as any competition authority would, has failed to show that the
proposed transaction will increase the likelihood of coordination; that
coordination would be feasible; and that there is an incentive on the part of either
Seaboard or Willowton to act collusively or in a coordinated manner.
[204] Therefore, we find that the Commission has not shown that the proposed
transaction is likely to result in a substantial lessening or prevention of
competition, in order to justify a prohibition.
[205]
how the proposed transaction would give the merging parties access to
information which would allow them to coordinate, we accepted the conditions
tendered by the merging parties in good faith regulating, among others, the
exchange of confidential and competitively sensitive information between them.
We discuss this in the remedies section below
58
PUBLIC INTEREST
Employment
[206] The merging parties have provided an unequivocal statement that the proposed
transaction will not result in any retrenchments in South Africa.196 The merging
parties also submitted that the second part of the transaction, being the
acquisition of the Elangeni Oil refinery, is likely to create jobs.197 During the
m the employee
representatives or trade unions.
Effect on a particular industrial sector or region
[207] The Commission approached the assessment of the proposed transaction on
the basis of the importance of soya meal to the poultry sector. As an essential
input to poultry feed, the cost of soya meal has an impact on the cost of chicken
production and the ultimate price of chicken sold to consumers. This is
particularly important because poultry is the cheapest of the available proteins,
making it a key food item for South African low-income households. Given the
importance of soya meal to poultry feed, the Commission considered whether
the proposed transaction would have a negative impact on the poultry sector.
[208] There is a policy recognition that a key challenge faced by the poultry sector is
the cost of feed, which makes up a large portion of the cost of chicken
production.198 Additionally, reliance on imported feed inputs and the fact that
local feed is based on import
two core constraints on competitive local poultry production.199As such, the
Government in partnership with stakeholders in the poultry sector have
developed the Poultry Sector Masterplan to ensure amongst others the increase
of the supply of soya bean products to the poultry sector and reduce prices.
196 Bundle B, Record, page 652.
197 Bundle B, Record, page 614.
198The South African Poultry Sector Master Plan developed in a partnership between Government
and a number of stakeholders in the industry, drawn from poultry farmers, processors, exports,
importers and organised labour. It provides a framework for a determined effort to grow the output
(and jobs) in the industry through various measures.
199 Industrial Policy Action Plan 2018/19 2020/21, by the Department of Trade and Industry.
59
According to the Commission, should the proposed transaction result in an
increase in prices of soya meal through coordinated effects, this will have an
immediate and negative impact on the poultry sector, and be contrary to the
objectives of the Poultry Sector Masterplan.200
[209] This concern could be seen as both a competition concern (higher prices of soya
meal) or a public interest concern (impact on a particular industrial region or
sector).201 Given that the Commission has not shown likely coordinated effects
resulting from the proposed transaction
contention that the proposed transaction is likely to negatively impact the poultry
sector.
[210] The proposed transaction also does not raise any other public interest concerns.
REMEDIES
[211] The merging parties argued that the merger ought to be approved
unconditionally. However, they had tendered conditions in the event that the
Tribunal found these to be necessary
[212] We are alive to the reality that the acquiring firms are two competitors that will
have an interest in a JV in the same market in which they continue to compete
namely soya meal. In such instances, competition authorities are obliged to act
with caution as it is recognised by competition authorities across the world that
JVs may in certain instances have the effect of chilling competition between the
partners in a JV.202 This issue was canvassed with the merging parties during
the hearing who accepted this proposition203 and offered up a range of
conditions.
200
201 Section 12A(3)(a).
202 OECD Policy Roundtable (2000), Competition Issues in Joint Ventures [DAFFE/CLP (2000), pages
9 10.
203 Transcript (6 August 2021), page 1026 line 7 page 1029 line 14.
60
[213] Acting prudently, we accepted, albeit with some amendments, the conditions
willingly tendered by the merging parties.
[214] The conditions initially tendered by the merging parties included confidentiality
obligations to prevent the exchange or sharing of competitively sensitive
information between the merging parties,
[215] The Commission in commenting on the tendered conditions still maintained that
the proposed conditions would not effectively address the harm that would result
from the merger and prevent a substantial lessening of competition between the
competing firms. The Commission submitted that the weakness of the proposed
conditions lies in the structural conditions which would effectively amount to
market division in contravention of section 4(1)(b)(ii) of the Act; in the information
exchange provisions which only focus on prices and customers and as such, do
not prevent the discussion of strategy in respect of the RSP business; and in the
lack of a suitable provision to prevent cross directorship.204
[216]
and did not impose a structural condition that allocates products amongst
competitors the duration of which is unknown, and which could lead to
unintended consequences.
[217] We however accepted the tendered confidentiality obligations. Although the
concern identified is in respect of soya meal (which is the market in which the
merging parties currently compete), we extended the confidentiality obligations
to also apply to soya hulls and crude soya oil as the merging parties contemplate
that they might compete in all three products in future.
204
61
[218] As the Commission could not put up a case to support the theory of harm of
coordination, we do not find it necessary to include a condition to prevent cross
directorship. We are satisfied with the condition that Seaboard would not appoint
a director involved in the sales activities of Seaboard to the board of RSP.
[219] The conditions also require notification of the said conditions to the directors and
employees of the respective parties. The merging parties have also undertaken
to adopt policies to ensure compliance with the conditions, and programmes to
promote compliance with the Competition Act.
[220] We turn now to consider the procedural issue of discovery applications that were
heard prior to the hearing of the main matter.
[221] Before the commencement of the hearing of the matter, both the Commission
and the merging parties brought an application to compel the production of
documents.
[222] In relation to the Commission's application for the merging parties' marketing
and pricing documents of the merging parties, the Tribunal dealt with this in
accordance with the established principle of relevance for merger proceedings.
There is accordingly no need for us to deal with this application any further.
[223] The merging parties' application to compel was unusual in that it sought
discovery of various of the Commission's internal documents. The documents
sought included:
223.1. the draft report by the Commission team investigating the merger and
submitted to the Executive Committee for discussion on 10 November
2020;
223.2. minutes of the Executive Committee Meeting of 10 November 2020
relating to the consideration of the merger and/or draft report;
223.3. all documents reflecting the feedback from the Executive Committee
following the meeting of 10 November 2020;
62
223.4. all documents reflecting the reasons at the time for not accepting the
remedies proposed by the merging parties in November 2020;
223.5. the draft report by the Commission team investigating the merger and
submitted to the Executive Committee for discussion on 2 March 2021;
223.6. minutes of the Executive Committee Meeting of 2 March 2021 relating to
the consideration of the merger and/or draft report; and
223.7. documents reflecting the feedback from the Executive Committee
following the meeting of 2 March 2021.
[224] The merging parties' reasons for seeking these documents was a concern that
in prohibiting the merger the Commission had taken into account irrelevant
considerations an familial
relationships
active in the soya meal, sunflower meal and edible oils market.205
[225] The merging parties furthermore were of the view that the deliberations of the
Committee and ultimate conclusions were driven by
the irrelevant consideration that Willowton was a respondent in a cartel
investigation initiated by the Commission in December 2016. The merging
influenced by the
inappropriate and unfounded suspicion that Willowton has been engaged in
cartel conduct."206
[226] In their view the Commission had not produced any evidence to support its claim
that Willowton has been involved in collusion and was biased in its assessment
of the merger.
[227] The Commission opposed the application to compel on three grounds, namely
the documents sought constitute restricted information in terms of rule 14(1)(d)(i)
and (ii) of the Rules for the Conduct of Proceedings in the Competition
205 Founding Affidavit, para 4.21.
206 Founding Affidavit, para 4.18.
63
, contain legally privileged information, and
proceedings.207
Evaluation
[228] Before turning to consider the merging parties concern of bias we highlight here
that the Commission is obliged under section 12A to have regard to a number
of relevant factors in its consideration of whether a merger is likely to significantly
lessen or prevent competition in the market.208. These include factors such as
the history of collusion in the relevant market, the extent of ownership by a party
to the merger in another firm or other firms in related markets;209 and the extent
to which a party to the merger is related to another firm or other firms in related
markets, including through common members or directors.210
[229]
Commission has acted with bias in its consideration of the merger.
[230] Even if we were to accept for arguments' sake that there might have been some
bias on the part of the Commission in its assessment, we show below that there
are sufficient checks and balances in the large merger framework of the Act to
alleviate any concerns of real or perceived bias on the part of the Commission.
[231] Sections 13A(3), 14A and 16(2) and 12A together provide that a large merger
may not be implemented until such time as the Tribunal has approved it. Unlike
while being of significant importance, retains its status as a recommendation and
tested in Tribunal merger proceedings, especially so in contested proceedings
where the Commission has recommended a prohibition. In Tribunal proceedings
the evidence of witnesses is tested through a process of cross-examination by
the other side and by the Tribunal itself through its inquisitorial powers. Hence
207
208 Section 12A(2)(c).
209 Section 12A(2)(i).
210 Section 12A(2)(j).
64
new evidence may come to light or the same evidence may be cast in a different
light.
[232] In this process the Tribunal will consider the totality of all the evidence and if the
evidence does not support the Commission's views the Tribunal will not accept
the Commission's recommendation to prohibit.211
[233] Indeed, it has happened on occasion that the Tribunal has differed with the
Commission's views and conditionally approved a merger which the
Commission has recommended to be prohibited. In other instances, the
Tribunal has agreed with the recommendation of the Commission to prohibit on
the basis of the evidence put up in its proceedings.
[234] It has also happened that during the course of merger proceedings the
Commission has obtained further information from witnesses or had an
opportunity to consider evidence in a different light, causing it to change its
recommendation during the course of Tribunal proceedings.212
[235] Notwithstanding the debate between the parties on whether the documents were
protected by Commission rule 14 or were legally privileged, in our view the
overarching principle for determining whether documents sought by an applicant
ought to be discovered is whether the documents are relevant to the main
proceedings.213
[236] In Jacobus, the Tribunal held that it is not enough for an applicant to merely
allege that the documents it seeks are relevant. An applicant must fully make
out a case as to why the documents sought are relevant to the dispute.214 It is
211 Senwesbel Limited & Senwes Limited and Suidwes Holdings (Pty) Ltd, Case No.: LM001Apr20.
212 Sasol Limited, Engen Limited, Petronas International Corporation Limited and Sasol Oil (Pty) Ltd
Case No: 101/LM/Dec04, para 12.
See also, British American Tobacco Holdings South Africa (Pty) Ltd and Twisp (Pty) Ltd, Case No.:
LM262Jan18; Senwesbel Limited & Senwes Limited and Suidwes Holdings (Pty) Ltd, Case No.:
LM001Apr20.
LM001Apr20.
213 Competition Commission v Sasol Chemical Industries Ltd (48/CR/Aug10), paras 35, 43,45 and 48.
214 Jacobus Petrus Hendrik Du Plessis and Another v Linpac Plastics Ltd (UK) and Others
(CRH126Nov11/DSC091Jun16), paras 18-20.
65
for the Tribunal to determine the scope and breadth of discovery in accordance
with principles of relevance to ensure a fair and effective hearing for litigants.
[237] In Industrial Development Corporation of South Africa Ltd and Anglo-American
,215 the Tribunal noted that documents
would only be relevant if they relate either to the factors set out in section 12A
216 In
this case, the Tribunal found that the applicant had failed to persuade it that the
documents sought to be produced (the advisory opinion of the Commission and
all minutes of meetings; internal memoranda and discussion notes relating to
the advisory opinion) necessary for the purposes of the hearingfailed
to prove that the documents were relevant in the determination of the merger.217
Here too we find that the merging parties failed to meet the relevance threshold.
They have failed to show that the documents would be relevant and would take
the Tribun any further.
[238] In our view the merging parties failed to establish how the production of internal
Commission documents, even if they did indicate bias, would have any influence
on the outcome of the merger in contested Tribunal proceedings where the
Commission's recommendation and the evidence it puts up is vigorously tested.
Ultimately if the Commission is unable to put up the evidence to support its
recommendation for prohibition it will be unable to persuade the Tribunal.
215 Case No.: 46/LM/Jun02.
216 Ibid, page 5.
217 Ibid, page 9.
66
[239]
and better discovery.
29 October 2021
Ms Yasmin Carrim
Date
Mr AW Wessels and Mr. H Cheadle concurring.
Tribunal case managers:
Duduetsang Mogapi and Mpumelelo Tshabalala
Tribunal economist: Junior Khumalo
For the merging parties:
Adv Bhana SC, Adv Gotz and Adv Kessery instructed
by Vani Chetty Competition Law (Pty) Ltd
For the Commission:
Candice Slump, Ipeleng Selaledi, Bukhosibakhe
Majenge, Jabulani Ngobeni and Luke Rennie