---

COMPETITION TRIBUNAL OF SOUTH AFRICA

Case No: LM001Apr20

In the large merger between:

Senwesbel Limited & Senwes Limited                                                                                      Primary Acquiring Firm

and

Suidwes Holdings (Pty) Limited                                                                                                Primary Target Firm

Panel                                                                      : Mondo Mazwai (Presiding Member)

                                                                              : Enver Daniels (Tribunal Member)

                                                                              : Imraan Valodia (Tribunal Member)

Heard on                                                                : 1; 3; 8; 19; 23 July and 5 August 2020

Final submissions recieved        : 14 August 2020

Order issued on                         : 18 August 2020

Reasons issued on                     : 12 April 2021

REASONS FOR DECISION

CONDITIONAL APPROVAL

[1]                        On 18 August 2020, the Competition Tribunal (“Tribunal”) conditionally approved the proposed transaction whereby Senwesbel Limited (Senwesbel), and its subsidiary Senwes Limited (Senwes), would acquire the entire issued share capital of Suidwes Holdings (Ring Fenced) (Pty) Ltd (Suidwes). Post-merger Senwes will control Suidwes.

[2]                        Our reasons for conditionally approving the proposed transaction follow.

BACKGROUND

[3]                          On 30 March 2020, the merging parties notified the Competition Commission (“Commission”) of the proposed transaction.

[4]                       The merging parties requested the Commission to consider the matter on an urgent basis as they claimed that Suidwes was in severe financial distress and would cease operating by the end of May 2020 if the transaction were not expeditiously approved. The merging parties submitted that if this happened Suidwes would likely be put under business rescue which would lead to uncertainty for farmers, disruption to the maize value chain and the retrenchment of approximately 1,246 employees located primarily in rural areas.

[5]                          The Commission duly conducted the investigation on an expedited basis to accommodate the parties. It concluded that Senwes’ acquisition of Suidwes would substantially prevent or lessen competition in the market for grain storage in concrete silos in three overlapping geographic areas. In particular, the Commission found that the merger would raise storage and handling costs at the different silos and would potentially lead to lower prices being offered to farmers for their grain. This is because on the Commission’s market definition, the merged entity would acquire a dominant position and become a monopoly in the respective geographic markets identified, which would give rise to these unilateral effects. The Commission recommended the approval of the merger subject to conditions, including the divestiture of silos, which were agreed with the merger parties. Given the expedited nature of the proceedings to date, the Commission had not tested the proposed remedies with market participants.

[6]                          However, prior to the commencement of the hearing, following questions raised by the Tribunal at the pre-hearing regarding, inter alia, about the appropriateness of the remedies (including the size and location of the silos to be divested),[1] the Commission tested the conditions with market participants.[2]

[7]                          Since the divestiture condition was motivated on public interest grounds inter alia that it would enable a B-BEEE entity to enter the market, the Commission indicated that its investigation on conditions revealed that the divestiture offer would not provide the scale for a new (B-BBEE) entrant to enter the grain storage market and compete effectively. The relevant silos proposed for divestiture were allegedly not profitable and had low-capacity utilisation.

[8]                          The Commission indicated on the eve of the hearing that it had changed its recommendation to one of prohibition. The Commission further submitted that while it recognised that Suidwes may be in financial distress, it was of the view that the failing firm defence did not hold. This was because there were alternative buyers for Suidwes who posed less competition concerns than Senwes. In addition, even if it were to be found that there were no alternative buyers of the Suidwes business, the silo assets of Suidwes would not exit the market since third parties were likely to buy these assets.

[9]                          The merger parties levelled criticism against the thoroughness of the Commission’s investigation which was clearly affected by the fact that it was undertaken an expedited investigation. The Commission on the other hand questioned the good faith of the merger parties.

[10]                      While disputing the Commission’s conclusions on the basis that the Commission had not discharged the onus of proving a substantial prevention or lessening of competition, the merging parties nevertheless tendered conditions to divest of three silos (in addition to pricing conditions) which the Commission initially accepted.

[11]                      hearing had been set down for 1 July and 3 July 2020 as an uncontested matter. However, the change from a conditional approval to a prohibition on the eve of the hearing, necessitated the hearing of evidence and further dates for hearing.

[12]                      Given this late development, the parties were directed to provide brief statements indicating the issues to be covered by each of their witnesses, which the parties duly did.

[13]                      We heard the matter on the following further dates: 8; 19; 23 July and 5 August 2020, with final argument heard on 14 August 2020.

[14]                      The Commission called three factual witnesses, each of whom was an unsuccessful bidder for Suidwes’ business (or parts thereof). They were:

[14.1] Mr Stefan Oberholzer (“Oberholzer”), the managing director of Oos Vrystaat Kaap (OVK), a competitor of the merging parties in the Free State;

[14.2] Mr Xolani Nhlapo (“Nhlapo”), a director at West Street Capital (WSC), an investment holding firm which focuses on investing in agriculture and more specifically agricultural infrastructure;

[14.3] Mr Theo Ernst Rabe (“Rabe”) the CEO of NWK Ltd, a competitor of the merging parties in the North West.

[15] The merging parties called two factual witnesses, each from the merging parties:

[15.1] Dr Herman Van Schalkwyk (“Van Schalkwyk”), the CEO of Suidwes.

[15.2] Mr Corne Kruger (“Kruger”), the CFO of Senwes.

[16]                      We note that the merging parties’ tendered conditions evolved over time in response to concerns raised by the Commission, the Tribunal and the independent expert. This culminated in the merging parties tendering a final set of conditions.

[17]                      We found, in the absence of conclusive evidence of a substantial prevention or lessening of competition, that the tendered conditions would address the potential competition and public interest harms arising from the transaction. We therefore conditionally approved the proposed transaction.

The Independent Expert

[18]                      Neither the Commission nor the merging parties called expert economists as they were initially in agreement on the conditions, obviating the need for a robust market definition or a showing of a substantial prevention or lessening of competition. However, in light of the remedies in particular being highly contested, and taking into account the worsening financial position of Suidwes, we notified the parties on 14 July 2020 that the Tribunal intended to call Professor Johann Kirsten (Prof. Kirsten), an agricultural economist at the University of Stellenbosch as an independent expert witness to provide an opinion, more specifically regarding remedies.[3]

[19]                      Following the guidance of the Competition Appeal Competition in Anglo/Kumba[4] regarding procedural fairness, the Commission and merging parties were duly given the opportunity to indicate any conflict with the appointment. It was made clear that the parties would (i) have access to Prof. Kirsten’s report; and (ii) have the opportunity to cross examine him on any submissions made therein.

[20]                      Prof. Kirsten filed an expert report on 17 July 2020 which was entered into the record as an exhibit.[5] He was made available for questioning by the merging parties and the Commission on 19 July 2020.

[21]                      We note that the Commission and merging parties’ position toward the expert changed during the proceedings. While the merging parties had initially reserved their rights regarding the expert’s appointment, they submitted that Prof Kirsten’s report and evidence showed that the market was a lot more dynamic than what the Commission had contended for.

[22]                      While the Commission had no objection to the appointment of the expert, it submitted that Prof. Kirsten had “opined on competition matters using the lens of an agricultural economist and not that of a competition economist”. The Commission submitted that his views were merely an opinion, which was not informed by the basic principles of competition economics, evidence or data and on this basis should be dismissed.

PARTIES TO THE TRANSACTION

Primary acquiring firm

[23]                      The primary acquiring firms are Senwesbel and its subsidiary, Senwes. Senwesbel is a public company listed on the Johannesburg Stock Exchange (“JSE”) and is not controlled by any single firm or shareholder. Senwesbel, Senwes and its subsidiaries are referred to below as ‘Senwes’.

[24]                      Senwes is one of the largest agricultural businesses in South Africa, with its central head office in Klerksdorp, in the North West Province. Senwes has serviced the agricultural industry since 1909 with agricultural production inputs and market access for commodities, as well as other value-adding agricultural services.

[25]                      Senwes’ operations are located primarily in the Free State, Gauteng and North West provinces of South Africa. It supplies its products and services largely to commercial farmers, processors (millers and oil seed processors) and traders. Its main activities include, amongst others, grain handling and storage, financing, grain trading, grain transport, equipment sales, agricultural retail stores, insurance, agriculture inputs and agriculture services.

Primary target firm

[26]                      The primary target firm is Suidwes, a private company. Approximately 89.36% of the shareholding in Suidwes is held by farmers. Suidwes and its subsidiaries are referred to below as ‘Suidwes’.

[27]                      Suidwes is a 111-year-old agricultural company in South Africa with its central office situated in Leeudoringstad, North West Province. Its business activities include grain storage and handling, grain trading, retail outlets, financing, and agricultural services, amongst others.

TRANSACTION

[28]                      In the proposed transaction, Senwes will acquire the entire issued share capital of Suidwes, through a scheme of arrangement in terms of section 114 of the Companies Act 2008.

[29]                      The transaction comprises two legs. In the first leg, Senwes will provide a loan to Suidwes amounting to R508 million. The Senwes loan is interest bearing (prime plus 5%) and repayable at the earlier of June 2021, or if the merger between Senwes and Suiwes were not to proceed for any reason, the loan would become a current liability in terms of the relevant accounting standards.[6]

[30]                      The second leg of the transaction is a scheme of arrangement in terms of section 144 of the Companies Act 2008, where Senwes acquires the entire issued share capital of Suidwes. Upon completion of the proposed transaction, Senwes will have sole control over Suidwes.

RATIONALE FOR THE TRANSACTION

[31]                      Senwes submitted that its rationale for the transaction was as follows:

“It is Senwes’ strategic objective to be an integrated agri-business and a significant role player in the food value chain. The strategy is therefore focused on growth and diversification through expansion within the agricultural and food value chain. The proposed transaction with Suidwes is aligned with Senwes’ strategy and will allow the Senwes group to diversify within its core strengths.”

[32]                      Suidwes’ rationale was the following:

“Suidwes is confronted by a situation where, without rapid intervention from a cashflow perspective, it will likely immediately face business rescue or liquidation. Suidwes is a failing firm facing significant financial challenges.

As a result of the significant financial losses which Suidwes had incurred, it breached certain covenants which it had with its primary lenders, being the Land Bank and First National Bank (“FNB”) …

Given Suidwes’ dire financial position, it is unlikely to be able to obtain credit from other financial institutions. Suidwes’ position has worsened considerably as a result of the coronavirus pandemic. The proposed transaction is, therefore, critical to the survival of Suidwes.”

[33]                      As mentioned above, Senwes extended a loan to Suidwes to assist it with the immediate risk posed by the demands of the Land Bank. However, despite this loan, at the Tribunal hearing, Suidwes submitted that its financial position had continued to deteriorate during 2020, and that this was further compounded by the impact of Covid-19.[7]

ISSUES FOR THE TRIBUNAL’S DETERMINATION

[34]                    The Tribunal had the following issues to determine:

[34.1] The first was with respect to the market definition.

[34.2] Second, was whether the merger would harm competition in the relevant market.

[34.3] Third, we were asked to determine the relevant counterfactual. That is, what would happen absent the proposed transaction in light of the allegation that Suidwes was a failing firm.

[34.4] Finally, we needed to consider whether the conditions tendered by the merging parties would be able to cure any harm arising from the transaction.

RELEVANT PRODUCT AND GEOGRAPHIC MARKET

Relevant Product Market

[35]                    The Commission defined the relevant product market as that of concrete silos for the storage of grain and oilseed operated by commercial silo operators.[8] It excluded all the alternative forms of storage and held that these do not impose a meaningful constraint on the concrete silos.

[36]                    Based on interviews conducted with 13 farmers during its investigation, the Commission argued that there were a number of technical reasons, including the deterioration of the quality of grain, safety risks, length of storage and a lack of cleaning and drying facilities, which rendered alternative storage facilities inferior to silos. During the course of the proceedings, the Commission relied on the evidence of these 13 farmers interviewed during its investigation to support its findings that alternative storage facilities were seen as inadequate substitutes to concrete silos. However, it did not call any of the farmers interviewed as witnesses. As we discuss later, this was detrimental to the Commission’s case since the farmers’ interviews could not be tested in oral evidence.

[37]                    The merging parties on the other hand relied on its two witnesses, van Schalkwyk and Kruger who operate concrete silos and considered alternative storage a competitive threat to concrete silos.

[38]                    The Commission in turn criticised the merging parties for not calling any farmers to substantiate their claim regarding alternative storage facilities to concrete silos. According to the Commission the evidence of the 13 farmers interviewed was more important than the technical substitutability evidence of the merging parties led through van Schalkwyk and Kruger.

[39]                    The Commission had also intended to call Mr Doors Kruger of Silostrat, a trader which operates a silo bag business, as a witness, but withdrew him a few days before the hearing. This also left some critical evidence untested in oral evidence, as we discuss later.

[40]                    We pause to mention that in the Imerys decision the CAC confirmed that the Commission bears the onus to prove a substantial prevention or lessening of competition.[9] Once it has discharged this onus, the merging parties then bear the onus to show that the proposed merger has pro-competitive benefits that outweigh its anticompetitive effects; or that the merger can be justified on public interest grounds.

[41]                    We turn to first consider the evidence regarding the technical substitutability of concrete silos with silobags.

Silobags

[42]                    Silobags are semi-permanent storage facilities generally made of a three-layer plastic film with UV protection.[43]  According to the Commission there are distinct functional differences between silobags and concrete silos.

[44]                    To support its findings the Commission relied on the evidence of Oberholzer of OVK, which owns concrete silos. According to Oberholzer, silobags have certain limitations, these include: drying of grain cannot be performed in silobags, theft is a greater problem compared to concrete silos and that silobags do not allow for cleaning. He also suggested that there may be a problem with rat infestations and that the loading of silobags is a much slower process compared to loading a concrete silo. These disadvantages may affect the quality of the grain stored in silobags.

[45]                    The Commission submitted that because of these disadvantages, silobags are not used as substitutes to concrete silos but are rather used as complementary facilities, to fill a gap in the market when there is not enough storage capacity at concrete silos. This was confirmed by the Commission’s witness, Rabe of NWK, a competitor of Senwes in his evidence, when he stated: “At this moment, no, I think there is silobag facilities, but that could be more because of market gaps and contracts towards specific markets. So, no, we don’t regard it as a direct threat…”.[10]

[46]                    In cross-examination, Rabe said the following regarding NWK’s[11] response to questions posed by the Commission during its investigation[12]:

“MR NORTON: So, the question is “do other storage options such as zinc bunkers, private silos and storage bags exert any competitive constraint on your business?” Do you see that?

MR RABE: Yes.

MR NORTON: And your answer is “there is a competitive constraint to some degree. Other storage mediums offer a product in competition with NWK”. Do you see that?

MR RABE: Yes.

MR NORTON: And the obvious meaning of what you’ve said there is that other storage mediums, and I assume that that means bunkers, silobags, etc, do offer competition to your concrete silos. Correct?

MR RABE: Yes, to some degree.”

[47]                    Mr Rabe further confirmed this in cross-examination[13]:

“MR NORTON: And then if you look at paragraph 18.3, you say in 18.3 “it is anticipated that should NWK not continue to offer relatively low storage and handling fees, as is currently the case, producers may be persuaded to consider the acquisition and use of silobags or other alternative options for storage of their grain”. Do you see that?

MR RABE: Yes.

MR NORTON: In other words, what you are saying here is you have to consider the potential diversion of products to these alternative storage options if you were to push your fees up. Correct?

MR RABE: Yes, we do.”

[48]                    The Commission submitted that although silobags are present in the market and may be growing, they have not yet attained critical mass to significantly constrain traditional concrete silos due to the limitations mentioned above.

[49]        The merging parties disagreed. According to them, the use of silobags has been growing in the market as can be gleaned from the growth of Silostrat’s business. Silostrat currently has approximately 600 000 tons of silobag capacity in the Senwes / Suidwes area. Van Schalkwyk pointed out that Silostrat has 305 000tons of capacity in the Suidwes area alone and that it is easily able to up-scale its capacity as silobags are easy to erect. He pointed out that the expansion of silobag capacity simply required moving a fence in order to expand the area where the silobags will be erected.

[50]                    Oberholzer later conceded[14] that where silobag facilities were constructed, the grain which is stored in those facilities would no longer be stored in concrete silos.

[51]                    One of the primary reasons given for the switch from concrete silos to silobags was price. The cost of storing grain in concrete silos is high for some farmers resulting in these farmers exploring alternative options. […][15]

                                              […]

[52]                    As mentioned, the Commission had originally elected to call Mr Doors Kruger (“Doors Kruger”) of Silostrat but subsequently withdrew him, although he attended the proceedings.[16] The evidence of Silostrat could not be tested in oral evidence on the competitive constraint it poses on concrete silos or on whether it provides storage to third parties. The merging parties alleged that the Commission’s withdrawal of Silostrat was because his evidence did not support the Commission case.

[53]                    Kruger indicated that Silostrat is storing grain in order to supply it to various millers and that this diverts grain away from the Senwes storage facilities. Kruger pointed out that farmers can store their grain with Silostrat and price the grain at a future date. According to him, this has the same effect as third party storage.[17]

[54]            […][18]

[55]                    However, as indicated, Silostrat was not called to give its view and its views could not be tested in oral evidence. The Commission’s view was that the merging parties, in arguing that alternative storage constained silos, conflate competitors in the storage market with competitors in the trading market. It put to Kruger the question why Silostrat is not listed on SAFEX if indeed it provides third party storage. Kruger submitted that it was Silostrat’s choice whether or not to register on SAFEX. Regardless, his evidence was that Silostrat (millers and on-farm storage) was taking grain away from concrete silos. The extent to which the trading and storage markets were interrelated could not be tested with Silostrat or from the demand side, with farmers.

[56]                    Regarding the drying capabilities of silobags Oberholzer admitted that he was unaware of new technology such as the Drylobag system which allegedly makes it possible for silobags to have the same grain aeration facilities as is available in concrete silos.[19]

[57]                    He testified that OVK did not take into account Drylobag technology or silobags when determining their storage fees and rates, but rather considered their own costing aspects. The Commission pointed out that neither Senwes nor Suidwes have used Drylobags or know how much they would cost to import.

[58]                    There are clear drawbacks associated with silobags which the merging parties did not dispute. However, as van Schalkwyk testified there are also risks in in concrete silos. According to him, such risks can be managed in both types of facilities.

[59]                    In our view, we heard sufficient evidence to suggest that silobags appear to be gaining traction in the market and the Commission’s witnesses (Oberholzer and Rabe) testified that alternative storage facilities pose a competitive constraint to concrete silos to some degree. This is confirmed by the Commission’s third witness, Nhlapo who confirmed in evidence that grain previously stored in Senwes’ concrete silo had been diverted towards silobags.[20]

[60]                    The above indicates that while silobags may not be perfect substitutes to concrete silos, there is a degree of substitutability between the two.

[61]                    Apart from the technical substitutability of concrete silos with silobags, a contentious issue as foreshadowded above was whether grain that goes directly from the farmer to a trader, such as Silostrat should form part of the relevant market. The Commission submitted that since traders, millers and on-farm storage facilities store for their own account, and are not available for third party storage, they should be excluded from the market. We deal with this in the section discussing General Trends in the Grain Storage Market.

Bunkers

[62]                    Grain bunkers may be constructed from corrugated iron, wood, concrete, or steel for the angled walls, while the floor is generally covered with a plastic ground sheet to prevent ground moisture from contaminating the stored grain. Grain is then stored within the bunker, which is generally also covered with sheeting to prevent inter alia wind erosion. In addition to the construction of the actual facility, additional investments are required for loading (e.g. conveyor belts) and drying equipment.

[63]                    According to the Commission the biggest disadvantage of bunkers is that grain stored in bunkers is more exposed to theft and hail. Further, bunkers lack other features such as drying and cleaning facilities which require additional investment. Like silobags, the Commission was of the view that bunkers should be considered to have a complementary relationship with concrete silos. Further   that bunkers are not available for third party storage but provide storage for the account of the bunker operator.

[64]                    The merging parties argued that the Commission’s analysis in relation to bunkers was very thin and appeared to be parasitic on its critique of silobags as being a suitable alternative to concrete silos. According to the merging parties the Commission simply asserted that there were limitations associated with fumigation, aeration and theft which were based on incorrect assertions made by Oberholzer.

[65]                    The merging paries submitted firstly, that both Suidwes and Senwes had fumigated their own bunkers in the past. Oberholzer conceded that he had no knowledge that it was in fact possible to fumigate bunkers.[21] Oberholzer was also unable to dispute that the speed of loading at Senwes’ bunkers was as fast if not faster than at certain of Senwes’ concrete silos. Finally, with respect to theft Kruger confirmed that Senwes had not had any insurance claims in respect of theft from its bunkers in the previous five years.[22]

[66]                    The merging parties argued that many market participants have constructed bunkers, including GWK, China Oil and Foodstuffs Corporation (“COFCO”), Afgri and the merging parties themselves, and that this was attributable to the fact that bunkers are not significantly different to concrete silos. In the overlapping areas only GWK has bunker facilities.

[67]                    According to Van Schalkwyk, following the construction of the GWK bunker in the Christiana area, the Christiana silo had lost up to one third of its throughput volume.[23] The merging parties submitted that there are a number of entities which have constructed bunkers in South Africa including Afgri (which has more than 500 000 tonnes of bunker capacity), Senwes (102 000), and Suidwes (175 000 tonnes of capacity). Senwes’ Raathsvlei bunkers have a capacity of more than 60 000 tonnes (which is the same size as the Regina silo).

[68]                    With regard to the Commission’s argument that third parties are not able to store grain in bunkers, the merging parties rebutted this argument by pointing to the fact that COFCO currently stores grain for third parties.[24]

[69]                    It appears that there has been an increase in the number of bunkers constructed and that these appear to be placing a competitive constraint on the merging parties’ silos as shown in the Christiana area, discussed later.

Zinc silos (and on farm facilities)

[70]                    Zinc silos are generally available in two types, namely corrugated steel or smooth walled. Zinc silos are fully enclosed structures installed with aeration and fumigation equipment, as well as hoppers for in and out loading. The zinc silos can differ by size and capacity depending on the needs of the storer and can be set up relatively quickly. These silos are generally erected by the farmers on their farms.

[71]                    According to the Commission, zinc silos have limited storage capacity and farmers generally have to produce large volumes of grain for the investment to be economically viable and thus only the largest farmers may be able to do this. Further that this storage was often only available to the farmer and was not made available to third parties. However, of these market participants, only GWK has bunker facilities in the overlapping areas.

[72]                    The Commission found that although some farmers have their own storage capacity, farmers generally view this alternative as too risky. This was mainly due to quality concerns as grain can sprout and become toxic. The Commission submitted that it was because of this that farmers preferred using commercial silos as silo operators have the expertise to grade grain correctly and remove poisonous seeds, broken seeds and determine the moisture content. Therefore, concrete silos allow for safer storage of grain for a long-term period.

[73]                    The merging parties disputed the Commission’s contentions. They submitted that there had been an increase in farmers erecting their own zinc storage facilities and that this had threatened their business as these farmers could also store grain for their neighbours.

[74]                    Kruger also reflected on the increase in on-farm storage and the fact that this results in grain by-passing Senwes’ storage facilities. He emphasised that in the Hoopstad area there are many farmers who have erected their own facilities and also that farmers store grain for their neighbours. He submitted[25]:

“MR KRUGER: Chair, once again, and I explained it in the first with Mr Norton. Take just Deon Berg as a farmer. He is planting 5 to 6 000 hectares. He is storing product on his farm. He is delivering at Premier Kroonstad. So, previously Deon Berg 5 000 tons times 6 tons per hectare, 30 000 tons was delivered at our facility. Now it’s gone. So, that’s competition for us. They are taking … they participate in the commercial production and therefore they took away product from us. It’s not that that product is not going to for commercial use. It is going for commercial use and therefore we are missing the opportunity.”

[75]                    The Commission put up a 2006 article to Kruger and to Prof Kirsten titled, “On- farm storage, the road ahead”, written by Dr Andrea Van Der Vyver, an agricultural economist in the rural development faculty of natural science at the University of Pretoria.[26] According to the Commission, this article highlighted the disadvantages associated with on-farm storage. The Commission used this article to make the point that there had been a slow uptake of alternative storage because of the disadvantages associated with on-farm storage, such as the cost to insure grain in on-farm storage (which tends to be higher), higher finance costs, higher marketing costs and a screening reduction.[27]

[76]                    In response Kruger submitted that while there are disadvantages associated with on-farm storage, there were also advantages such as the tax reduction which is enjoyed by farmers with on-farm storage. Secondly, that the market has changed and that many farmers have resorted to constructing their own on-farm storage.

[77]                    In his evidence, Kruger confirmed the view of Professor van der Vyver that “…farming activities increased in size. Direct sales to processors became possible and transport from the farm, on-farm loading to processors has become a common practice. Therefore, the demand for on-farm storage and increased storage at processing facilities is on the rise.” When asked his thoughts on this statement, he submitted[28]:

“MR KRUGER: I think it’s exactly correct, Chair, and the farmers with their new John Deere equipment, they harvest so fast, they actually choose to erect facilities on their farms, because it’s buffers their harvest process and from there on it gives them channel to market. So, it’s exactly correct. It actually forms part of a plaaslaai action as well, Chair. It’s to buffer your product on-farm and then store it there or buffer it and send it off to the miller. That’s the two types of models you get there.”

[78]                    The merging parties pointed out that Professor van der Vyver in his article indicated that the construction of on-farm storage allows for the delivery of grain from the farm to the miller (so called plaaslaai)[29]   throughout the season and not only at harvest time. Professor van der Vyver stated that this trend was likely to   continue.

[79]                    From the Suidwes point of view, Van Schalkwyk testified that where a farmer is disgruntled with the service offered by Suidwes, the farmer will construct his or her own facility and “it will mean that we lose him for good, because he will store on his farm and he won’t bring it to our silos anymore”. He also confirmed that “we know of various examples of farmers that actually do store maize for their neighbours and other farmers in their facilities.” As the merging parties put it, on- farm storage will divert grain away from concrete silos and it is, therefore, a suitable alternative to concrete silos.

[80]                    From the submissions of the 13 farmers, there are different views with regards to zinc silos as a viable alternative of the traditional concrete silo.

[81]                    Some farmers have expressed that they have not and do not use zinc silos (and other alternatives storage facilities such as silo bags, bunkers, and grain dams). These farmers are of the view that such alternatives are not the best substitute for concrete silos due to the quality of grain and the possible risk that grain could get exposed to various elements that may cause damage.

[82]                    On the other hand, some of these 13 farmers which have used zinc silos (together with other alternative storage facilities) are of the view that they provide an alternative to traditional concrete silos. There are also farmers that consider all types of storage facilities in the same way but note that cement silos and zinc silos are expensive as an option, to construct on farms. Some farmers believe that zinc silos and concrete silos are good for long-term storage.

[83]                    Based on the above we are of the view that the evidence of the 13 farmers is mixed and inconclusive.

[84]                    Zinc silos appear to be increasing in use, however the extent to which these place a competitive constraint on concrete silos, or how rapidly they will grow in future is unclear. Without the benefit of oral testimony from the farmers, we leave question this open.

General Trends and Dynamics of the Grain Storage Market

[85]                    According to the merging parties there has been a noticeable growth in alternative storage facilities in the last 20 years, while over the same period no new concrete silos had been built.

[86]                    Kruger testified that in the period 2003 to 2020, alternative storage capacity had increased from 500 000 tons to 2.2 million tons in the area in which Senwes operates; and that on a national basis the total alternative storage capacity currently stood at 9 million tons.[30] These 9 million tons had been erected over the last 15 years as compared to the 50 years of construction it took to construct the current concrete silo capacity in the market.[31] According to the merging parties, the growth in the volume of alternative storage capacity was proof that it is exerting some competitive constraint on concrete silos.[32]

[87]                    The merging parties put up an extract from Senwes’ Strategic Plan, which shows the current dynamics of the grain storage market, as replicated in Table 1 below below.

Table 1

[…]

[88]                    Further, the merging parties submitted that the JSE does not distinguish between concrete silos and alternative storage as an approved silo is defined as: “a delivery point that could either represent an upright storage structure, bunker, silobag site or warehouse owned by a JSE approved storage operator and approved for each marketing seasons in terms of the requirements set out in Appendix D.”[33] They pointed out that since 2009 the number of registered silos has increased reflecting the growth in alternative storages.

[89]                    The Commission submitted that even if silobags (or alternative storage in general) were considered as alternatives, they only account for less than 5% of total storage capacity nationally.[34]

[90]                    As indicated, one of the issues in dispute was whether the grain that is not stored in the merger parties’ silos forms part of the contestable market. The merging parties submitted that there appeared to be a substantial amount of grain bypassing their concrete silos. Taking into account the amount of grain delivered to their silos in the Senwes and Suidwes areas relative to the total maize production in those areas, they estimatated that approximately 40% to 50% of grain bypassed their silos, as indicated in Table 2 below.

Table 2: […]

[…]

[91]                    They submitted that this demonstrated that over time an increasing proportion of the grain which is produced in the traditional Senwes and Suidwes areas by- passed the parties’ own storage facilities. According to them, this means that there are a range of other storage facilities in which grain is being stored and these impose a direct or indirect constraint on entities such as Senwes and Suidwes.

[92]                    Kruger testified that one of the significant benefits which alternative storage facilities have over concrete silos is the fact that they can be positioned in the middle of production areas, whereas concrete silos cannot be moved. He emphasised that for every ton of grain which reached the mills through an alternative channel, the owners of concrete silo facilities lose the revenue associated with that ton of grain, which they would have earned if the alternative facility did not exist.

[93]                    Further that the existence of traders and millers who purchase directly from farmers via plaaslaai also impose a constraint on the activities of the providers of storage services.

[94]                    As part of the assessment of the grain that bypasses the merging parties’ facilities, the merging parties also submitted municipal and district production data for the Free State and North West areas, aimed at reflecting the estimated grain production from the three overlapping areas. Kruger estimated that Senwes stored approximately […] of the grain produced in the overlapping area, and the balance of […] bypassed their silos in these overlapping areas.

[95]                    The Commission disputed this analysis and argued that these bypass figures did not reflect the size of storage that is lost by the merging parties to rivals in the concrete silo storage space, since the bypass figures included grain stored at traders' facilities and millers' facilities, farmers’ own storage facilities, and plaaslaai grain. According to the Commission the bypass figures are not part of the contestable market since traders, millers, own farm storage facilities and plaaslaai do not store for third parties but for their own account. The Commission submitted that the merging parties conflate loss of storage revenues to other commercial storage competitors vs the loss of revenues to all competitors in the procurement of grain. To determine the loss of storage revenues to rivals, the merging parties should not cite instances when they lose to rivals in the procurement process for grain.

[96]                    The Commission submitted further that to understand the actual sizes of the merging parties in the concrete silos storage market, the Tribunal should instead consider the evidence of the capacity that Senwes and Suidwes have in the overlapping areas. We deal with this in the market share section.

[97]                    We did not have to conclusively decide on the bypass figures in the overall determination of this matter. The Commission submitted that: “I think so much has been made by the merging parties around that actually in the physical world a lot of grain bypasses concrete silos. We are unperturbed about that, because to our view that’s irrelevant, because we have a storage market. So, we are worried about grain that actually does go into the storage. We are less concerned about grain that doesn’t go via concrete silos.”[35]

[98]        This however does not speak to the evolving dynamics in the market and the interrelationships between the distinct markets from both the supply and demand side. We heard no direct evidence from farmers regarding how they made decisions on the different storage options. This is not to suggest that the Commission’s conclusions may ultimately be found not to be correct, however,they could not be thoroughly tested with relevant factual witnesses or with detailed economic assessment to make a conclusive determination.

[99]                    As we discuss below, the evidence suggested that the market was more dynamic than contended for by the Commission.

[100]                The Commission also relied on Senwes/Suidwes’ strategic documents which considers participants in the grain storage market. It submitted that these documents, contrary to what the merging parties contended, did not reflect alternative storage facilities as competitors.[36] More specifically they failed to mention the names of the merging parties’ two biggest alleged competitors COFCO and Silostrat.

[101]                The merging parties in response indicated that they refer to ‘alternative storage’ multiple times in these documents as being competitors. Further, that the documents show that alternative storage is listed as a risk to the Senwes business.[37] Not much turned on this issue in our overall conclusion.

[102]                Similarly, the two articles (of 2006 and 2017) put by the Commission to the merging parties witnesses did very little to advance the Commission’s case without direct evidence that could be tested in oral evidence.

[103]                The 2006 article is the van der Vyver article referred to under the zinc silo discussion above. It sought to indicate that although alternative storage facilities had been in the market for some time (since 2006), their uptake was slow.[38]

[104]    The 2017 article is an article in which Senwes is reported as saying that concrete silos are still the safest method of storing grain and that Senwes will continue to use this and only when this capacity is filled up will Senwes look to alternatives. Kruger explained that this was published in a year where there was was abumper crop, and it was an attempt to encourage farmers to deliver their product directly. There is no evidence to the contrary.

[105]                The Commission’s other argument, aside from the above by-pass arguments, was that Senwes’ own documents showed that its […]. Senwes’ silos are […] which is evidenced by Senwes’ financial statements which showed that […] with an […]. Furthermore, a strategic planning document from March 2020 showed […]

[106]                The Commission was of the view that with returns of […] concrete silos were highly profitable. Kruger did not dispute the evidence and […] However, he submitted that […] that[39]:

                                  […]

[107]                This issue could not be taken further absent a detailed assessment of costs, and was on its own inconclusive to establish a substantial prevention or lessening of competition.

[108]                We now turn to consider Prof Kirsten’s evidence.

Prof Kirsten

[109]                Prof Kirsten was of the view that alternative storage facilities do indeed provide viable alternatives to concrete silos and that this trend would increase rapidly in the future.

[110]                In this regard, Prof Kirsten stated that:

“The argument for delineating the market is to my mind driven to a large extent by the incorrect perception that alternative storage systems on farms and elsewhere do not provide competition to the current silos. Like the merging parties I am also of the view that the alternative storage options do indeed present profitable alternatives – especially for the large farmers.”

[111]                However, he went much further, to state that in his opinion, the relevant product market was one for “a market solution that is provided to the farmer and that market solution revolves around the issue of trading, storage, price, delivery time, etc. So, a trader or any company can come to a farmer with a particular solution that takes the grain away from him or her in a way that is convenient, speedy and at a price that is sensible to the farmer.”[40]

[112]                It cannot be disputed on the evidence that new technologies have entered and are entering the grain storage market given the need by farmers to find a more integrated solution. It is also clear that these technologies have been disrupting the grain storage market especially with respect to concrete silos given the increased uptake of alternative storage. The extent of this disruption and changes towards integrated market solutions from the perspective of farmers was unclear. This could not be tested with farmers.

[113]                The Commission sought to challenge the expert testimony given by Prof. Kirsten, primarily on the basis that he was not a competition economist and had not applied the SSNIP test in order to define the relevant product market.

[114]                A SSNIP test seeks to analyse whether a non-transitory price increase in a candidate market would be profitable. In this respect, the test is seeking to assess the cross elasticity of demand for a given good in comparison to another alternative good. However, as Professor Kirsten indicated such tests are inherently complex and require significant amounts of data (which was not available) to be able to determine with any degree of scientific accuracy whether two products are indeed substitutes.

[115]                In our view such criticism levelled against Professor Kirsten suffered from the difficulty that the Commission itself did not conduct a SSNIP test which it advocated for. Instead it relied on submissions made by farmers, evidence which as discussed above, was mixed and inconclusive, and could not be tested with the farmers since none were called.

[116]                It is common cause in this case that no economic expert evidence was called by the Commission nor the merging parties which would have provided a more robust assessment of the relevant markets. In particular, we had no expert evidence, such as a SNNIP test, to provide a quantitative assessment of the competitive dynamics between alternative storage options. Neither were we presented with any uncontested evidence from actual market participants, such as grain farmers, about how they made decisions on the different storage options. Moreover, the debate on bypass figures was bogged down in technical arguments and left the matter moot. Absent such evidence, we aknowlege that alternative storage facilities cannot be excluded from the relevant market, however, on the evidence cannot conclusively decide whether they exercise a sufficient competitive constraint to concrete silos.

Conclusion on the relevant product market

[117]                The Commission’s approach to market definition was premised on its conclusion that alternative storage facilities should be discounted as forming part of the relevant contestable market as they do not exert a sufficient competitive constraint on concrete silos.

[118]                We are of the view that the Commission’s assertions that alternative storage facilities do not form part of the relevant product market is not sufficiently supported by the evidence of the witnesses before us. At best for the Commission the interviews of the farmers the Commission relied on is mixed. This evidence could not be tested with the farmers as none were put up as witnesses. With the debate on bypass figures bogged down in technical arguments, we rely on the evidence of the expert witness which suggests that the competitive dynamics in the market have changed substantially in the recent period, and the in seeking a comprehensive solution for their maize output, farmers do actively consider alternative storage facilities available to them.

[119]                Having also analyzed each form of alternative storage facility identified by the Commission, we are of the view that each has its own advantages and disadvantages associated with it, including concrete silos.

[120]                To our mind there is a degree of substitution across these different forms of storage, albeit to varying degrees. It bears mention that one of the disadvantages of concrete silos, which were not disputed by the Commission are that they are expensive to construct and operate. They cannot be moved, unlike alternative storage solutions, which means that when grain declines in areas where concrete silos are located this often leads to these silos being mothballed and/or operating at a lower efficiency compared to other silos located in high production areas.

[121]                The evidence further shows that there has been significant growth in alternative storage capacity over the last 15 years with no new concrete silos being built over the same period. The evidence of farmers, although mixed would suggest that they are in fact using alternative storage. While the merging parties profit margins have been high despite the increased storage capacity in the market, the Commission did not put up any evidence to refute Kruger’s explanations on margins.

[122]                Based on the evidence before us we are of the view that the market is potentially broader than the market for concrete silos and includes silobags, bunkers and zinc silos to a degree. However this broad market could not be tested with farmers and alternative storage facility providers. We have thus found the evidence inconclusive on the scope of the relevant market.

Relevant Geographic Market

[123]                According to the Commission, the farmers interviewed indicated that they were unlikely to travel further than between 25km and 35km to deliver their grain into a storage facility due to transport costs, convenience and the SAFEX transport differential.[41] This was also confirmed by competitors, who indicated that beyond this distance delivery of grain would no longer be economically feasible for a vast majority of farmers.[42] In light of this the Commission used a 40-kilometre radius as “the outer distance which would capture a typical area in which farmers are willing to travel to deliver their grain.”

[124]                The Commission concluded that there were three local geographic markets (in a 40km radius around the target silos) in which Suidwes silos overlap with silos owned by Senwes in the North West and Free State provinces.

[125]                  The areas of overlap are the Ottosdal, Leeudoringstad, and Christiana areas.

[126]                The merging parties submitted that the Commission’s approach to the geographic market was arbitrary as no witness was called by the Commission to support this approach.

[127]    In their filing the merging parties defined the geographic market as a radius of 40-60km. However, during the hearing, they submitted that the market for grain storage had changed since the competition authorities’ decisions in Afgri andSenwes. They contended that the geographic market was actually much wider than the 40-60km radius as grain could be transported as far as 100km.[43]

[128]                In response to a question by the Chair, Kruger submitted the following:

“CHAIRPERSON: Yes, before you do that, Mr Norton, I have a question for Mr Kruger about the grain that moves, as you say, even further away from the 40 kilometre radius to another Senwes facility. What would the reason be for a farmer moving all the way to that particular facility?

MR KRUGER: Ja, it’s all about the economics Chair. So, the transport differential from the place he is harvesting till Randfontein, if you move it, say we are farming here in the west, so if you move from the west to the east and to Randfontein, that transport differential is to your advantage. So, anywhere you can even deliver 5 or 400 kilometres away from where it’s been harvested, you can deliver the product and therefore if you don’t like Senwes, you can choose any facility from here to Randfontein and you can even choose the millers in Randfontein So, Chair, it is normally the economics, which makes sense and the Unigrain business model is exactly like that. They’ve got a transport solution coupled with a trading solution and they just say to the farmer, listen, I collect at your farm, I will take it wherever I want, the farmer does not even know where the product is going. He just knows that he is selling it to Unigrain and then Unigrain will settle the transport costs and he will deduct that transport cost from the transport differential of Safex. So, that’s normal economics for him or that’s how his business model is working.”

[129]                However, when asked a follow up question about the catchment area around the silos i.e. how much of the grain in the Senwes silos comes from more than a 40km radius, Kruger indicated that it was 10%, 20%. He indicated that up to 60-70% comes from within a 40 km or less radius.[44]

[130]                While Rabe testified that the catchment area for silos is approximately 20km for farmers, he also stated that it could be wider for market players such as millers who would be willing to transport product up to distances of between 200 and 250km.[45]

Conclusion on Geographic Market

[131]                It appears from the evidence that farmers choose a silo on a consideration of commercial terms, including transport costs to the storage facility.

[132]                While the transport differential may extend the distance over which grain is transported, the evidence is not conclusive as to a specific radial point at which one silo falls within or without the catchment area of another. Both Rabe and Kruger’s evidence shows that the distance differs depending on the player concerned. Traders may transport grain over longer distances, but farmers (particularly small-scale farmers) may not.

[133]                The distance travelled will depend on many factors, and while the distance may be as far as 400km, this does not apply to the bulk of grain produced in the vicinity of the concrete silos.[46] What was clear from the evidence was that the majority of the grain, between 60 and 70% entered the merging parties’ silos from within a 40km radius. This would suggest to us that the majority of farmers are only prepared to travel short distances to the closest storage facility possibly due to the high cost of transport.

[134]                We have accepted the geographic markets as delineated by the Commission i.e. as three local geographic markets (in a 40km radius around the target silos) in which Suidwes silos overlap with silos owned by Senwes in the North West and Free State provinces.

MARKET SHARES AND MARKET CONCENTRATION

Competition Analysis

Market shares

[135]                We next turn to the market share analysis to further assess the competitive constraints of other storage alternatives. The market shares assist in providing an overview of the market given the limited data analysis provided by the Commission.

[136]                This market share analysis underwent various iterations as the evidence showed that the market was potentially broader than what had been defined by the Commission. We reflect those calculations which spoke directly to our considerations.

[137]                Based on a relevant market definition of only concrete grain silos the Commission estimated the following market shares:

[137.1] The Commission found that the merging parties would have a combined market share of approximately 65% in the North region (Ottosdal).

[137.2] A market share of approximately 100% in both the East (Leeudoringstad) and West (Christiana) regions.

[138]                During the proceedings following testimony by Van Schalkwyk[47] it appeared that data from the South African Grain Information Service (“SAGIS”) may be an alternative source to answer the question of how much of the grain around each of the silos is actually stored in these concrete silos.

[139]    The Tribunal therefore requested both the Commission and the merging parties to procure this data for the preceding 5-10 years which would cover, amongst other things, the total level of production in the various geographic areas, the volumes stored of this production in the various forms of storage, as well as capacity and utilisation of each of the silos in the three geographic markets.

[140]                It turned out that this data was not available through SAGIS, and the merger parties instead purchased data from an independent third party, GeoTerra, which estimates the production volumes per magisterial district.

[141]                However, given that the Commission had defined the geographic ambit to be a 40km radius from the relevant Suidwes silos, this created methodological issues given that only portions of each of the magisterial districts surrounding the silos would fall within this 40km radius. The merging parties were of the view that including the total production figures for each of the magisterial districts would overstate the production area falling within the scope of this radius as it would potentially include portions of these districts which are located considerable distances from the relevant silos and, thereby, understate the proportion of the production of maize in these districts which flows into the relevant silos. On the other hand, only including the area falling within 40km of the relevant Suidwes silos would underestimate the relevant production areas as it would not include the additional and adjacent areas which may feed the relevant Senwes’ silos (i.e. outside of the 40km radius from the relevant Suidwes silo, but within 40km of the relevant Senwes silos).[48]

[142]                To overcome this methodological hurdle, the merging parties sought firstly, to include only the production areas which fall roughly within 40km of the relevant Suidwes and Senwes silos in order to calculate the total production in these areas surrounding the silos. The GeoTerra data was not available on a more granular level and as such the merging parties made assumptions as to the proportion of the production area in each of the districts that would fall within the approximate 40km radii.

[143]                The Commission on the basis of the production data provided by the merging parties and to overcome the hurdles identified, conducted its own exercise (as   described further below) aimed at estimating the level of production in each of the three geographic markets.

[144]                The Commission considered the (approximate) location of the districts which would fall within each overlapping area guided by the views of industry participants such as Silostrat, GWK and NWK. No farmer views were obtained. The Commission considered the maize production only from those municipal districts which can be considered to be reasonably falling within the overlapping areas Ottosdal, Leeuringstad and Christiana, and applied the merging parties’ own assumptions on the proportion of grain that can be considered to have been produced from each of those identified districts.[49]

[145]                The Commission also estimated the proportion of alternative storage facilities relative to total storage capacity per geographic area based on information obtained.

[146]                For each of the geographic areas, the Commission used the total storage capacity (which includes the storage capacity of the registered grain silos as well as alternative storage facilities that offer storage services to third parties) to estimate the market shares for grain storage facilities in the overlapping areas. Privately-owned silos were excluded given that on the Commission’s investigation they are mainly used to store farmers’ own grain and are not always available to the public or for use by other third parties. In other words, on this estimation the Commission showed what the merging parties market position was relative to alternative storage providers.

[147]                We discuss the results of this exercise below by geographic area.

Ottosdal

[148]    There are five relevant silos in this region: Suidwes Bamboesspruit, Suidwes Strydpoort and Suidwes Wolmaranstad. The two Senwes silos which are locatedwithin a 40km radius of the Bamboesspruit silo are the Melliodora and Werda silos. NWK is the only concrete silo competitor in this area.

[149]                The merging parties and the Commission estimated the percentage of the white maize produced in these areas which is receipted into the merging parties’ silos.

[150]                According to the merging parties’ estimates, white maize receipted into the merging paries concrete silos is depicted in Table 3 below.

Table 3: […]

[…]

[151]                In calculating its market shares, the Commission obtained additional information from NWK as a competitor to the merger parties in the Ottosdal area.

[152]                The Commission then excluded certain of the production areas which had formed part of the merging parties’ calculations. It (i) excluded the Lichtenburg magisterial district (which the merging parties submitted fell within a 40km radius of the Ottosdal, Werda and Melliodora silos and should be included) and; (ii) included the Wolmaransstad silo (which the merger parties had excluded as it is located more than 40km away from the Werda and Melliodora silos).

[153]                The Commission’s market share calculations are depicted in Table 4 below:

Table 4: […]

[…]

[154]                In addition, the Commission recalculated market shares inclusive of alternative storage and found that the merging parties would have a market share of approximately 65% in the Ottosdal area post-merger as depicted in Table 5 below:

Table 5: […]

[…]

[155]                The merging parties pointed out that the NWK silos are located between the Senwes and Suidwes silos and have lower storage and handling rates than the Suidwes silos. This means that they are the most proximate constraint to the Suidwes silos. This fact combined with the point that the merging parties have a combined share of receipts of white maize of less than 32% in 2019/2020 means that there is unlikely to be any concerns arising in relation to the combination of their silos in this area, even on the Commission’s estimates.

Leeudoringstad area

[156]                In the Leeudoringstad area the following four silos are relevant: Suidwes Leeudoringstad, Suidwes Wolmaransstad and Suidwes Makwassie which overlap with the Senwes Regina silo within a 40km radius.

[157]                As before, the merging parties and the Commission estimated the percentage of the white maize produced which is receipted into the merging parties’ silos. The Commission prepared its calculations based on its understanding of the market and excluded certain districts i.e. Delareyville, Schweizer-Reneke, Christiana and Viljoenskroon districts from its calculations on the basis that these districts are located too far away from this geographic market to be included. Only three districts were included in its calculations, they were the Klerksdorp, Wolmaransstad and Bothaville districts.

Table 6: […]

[…]

Table 7: […]

[…]

[158]                The Commission also re-estimated market shares taking into account alternative storage and found that the merged entity would have a market share of 72% in the Leeudoringstad area. Once again, the merging parties pointed out that on the Commission’s own estimates, they had a market share of […] in 2019/2020. This would mean that […] of the white maize produced in this area by-passed the merger parties’ silos.

Table 8: […]

[…]

[159]                In this region the merging parties further pointed out that there are a number of third parties such as Silostrat and Vyf Susters’ silobag depot offering alternative storage facilities.[50] According to Kruger, Vyf Susters has a silobag facility of between 140 000 and 160 000 ton. In addition, the merging parties have submitted that there are also a number of farmers who have built their own storage facilities in the area.

[160]                Kruger further testified that the area between the Leeudoringstad and Regina silos is a cattle farming area and that the primary area where Leeudoringstad would draw grain is the area to the south closer to Vyf Susters. Therefore, no concern arises given the combined shares of less than 35%; as there is effective competition from Vyf Susters and Silostrat to constrain the behaviour of the merged entity.

[161]                The Commission also re-estimated market shares taking into account alternative storage and found that the merged entity would have a market share of 72% in the Leeudoringstad area.

Christiana

[162]                While district level production information was available from GeoTerra for the Ottosdal and Leeudoringstad areas which fall in the North West and Free State provinces, the same was not the case for the whole of the Commission’s Christiana area. This was because, while the Christiana facility is located in the North West province, Jan Kemp Dorp is located in the Northern Cape province.

[163]                For the Jan Kemp Dorp area, the merging parties relied upon internal Senwes estimates of production volumes and not on independent third-party data. Senwes submitted further that the total maize production volumes from this area make up such a small proportion that the merging parties submitted that was unlikely to have a material impact on the overall volumes which would be produced in the Christiana overlapping area identified by the Commission. This is because the Christiana area is mostly a cattle farming area. In addition, the merging parties included in their calculation the actual maize receipts in respect of the relevant silos over an eight-year period. This data reflected the actual volume of maize which was received at each of the relevant silos.

[164]                Suidwes operates a 68 000ton capacity silo in Christiana. Senwes’ Jan Kemp Dorp silo has a capacity of 28 000 tons, and is located within a 40km radius from the Christiana silo. In this region, GWK has constructed a 30 000 ton bunker next to the Jan Kemp Dorp Silo.

[165]                GWK also owns the Westra mill, which previously belonged to Suidwes and which is adjacent to the Christiana silo. According to Van Schalkwyk, GWK stored its grain in the Christiana silo prior to it constructing its own 50 000-ton bunker alongside the Christiana silo. Once this was erected GWK stopped storing its grain at the Christiana silo, which according to Van Schalkwyk was because GWK could now store its grain at lower rate and have better access to its grain.[51]

[166]                The Commission acknowledged the construction of the GWK bunkers next to the Jan Kemp Dorp and Christiana silos, but discounted them on the basis that they are not concrete silos. The merging parties submit that there are many other third party owned storage facilities in the areas surrounding the Jan Kemp Dorp and Christiana silos.

[167]                As before the merger parties and the Commission estimated the production areas falling within a 40km radius of the relevant silos as well as the proportion of maize receipts passing through the relevant silos.

[168]                The Commission excluded the production areas of Hoopstad and Boshoff due to these districts being located too far away from this geographic market. The Commission’s production estimates only included the Jan Kempdorp, Hartswater, Magogong and Christiana districts. On this basis it found that the merging parties’ receipts accounted for over […] market shares consistently in the period 2011-2019, as depicted in the Tables 9 and 10 below:

Table 9: […]

[…]

Table 10: Proportion of maize that passes through the merging parties silos

Period	Total average production	Total receipts	Receipts relative to total production
2012/2013	65100	198 385	305%
2013/2014	115 472	121 869	106%
2014/2015	86 165	140 139	163%
2015/2016	70 330	168 826	240%
2016/2017	64 005	116 643	182%
2017/2018	52156	87674	168%
2018/2019	32954	52099	158%
2019/2020	58796	60758	103%

Source: Commission’s calculations.

[169]                The merging parties disputed the Commission’s calculations in this regard and submitted that the analysis was self-evidently wrong. It was simply not possible for the merging parties to have a market share in excess of 100% and that this was a clear indication that the Commission had excluded relevant production areas from its calculation erroneously.

[170]                The Commission acknowledged that its estimation of market shares “…would appear to suggest that the merging parties’ combined annual receipts are significantly (and consistently) more than the average maize production figures in the area. This implies that these facilities [are] likely [to] receive significant amounts of maize from areas outside the immediate Christiana area and the Commission has not been in a position to ascertain, in the time available, the full extent of the areas from which the grain is derived…”

[171]                The Commission’s own information shows that GWK has a […] bunker situated next to the Christiana silo, constructed in 2016/2017, and that it has a   throughput of […]

[172]                The Commission’s data also reflects the reduction in the throughput of the Christiana silo since the construction of the GWK bunker. This suggests that the GWK bunker is an alternative to the Christiana silo and a significant competitive constraint. However, the GWK volumes receipted by the GWK bunker were not included in the Commission’s calculations.

[173]                In the Christiana area, based on storage capacity the Commission also estimated that the merged entity would have a post-merger market share of 67%.[52]

Conclusion

[174]                We heard many iterations of the market shares over the course of the proceedings. The Commission’s initial assertion was that the merging parties would be dominant and enjoy a monopoly in at least two of the geographic areas (Leeuringstad and Christiana) identified based on storage capacity.

[175]                The merging parties on the other hand provided data in which their market shares post-merger would be less than 35% in each of the three geographic areas on the different permutations of market shares discussed above.

[176]                Given these disparities in market share calculations in the three geographic areas, and the disputes about the reliability of the data, we have not placed much reliance on them. We turn to consider the Commission’s theories of harm against this backdrop.

THEORIES OF HARM

[177]                The Commission was of the view that the merger would give rise to unilateral effects due to increased concentration in the relevant markets. It relied on three theories of harm.

[178.1] First that the transaction would allow the merged entity to raise storage and handling fees in the three overlapping areas.

[178.2] Secondly, that the merged entity would be able to procure grain from farmers at cheaper prices than was previously the case pre- merger.

[178.3] Finally, that as a consequence of the volumes of white maize stored in the merging parties’ facilities, that post-transaction the merged entity would be in a position to influence the price of white maize by unlawfully withholding maize from third parties in their facilities.

[178]                The merging parties broadly contended that there was no basis for any of the Commission’s theories, and insofar as there were concerns, the conditions tendered would adequately address the concerns.

[179]                We briefly discuss each of these theories below.

Increase in grain storage and handling fees

[180]                The Commission submitted that the merger would likely lead to a loss of competition in the relevant market as Senwes and Suidwes are the two major participants in their respective regions. Senwes is already the largest grain storage company in South Africa and the proposed transaction would remove the competition between these companies.

[181]                The Commission argued that the removal of an effective competitor, as a result of this transaction, would likely result in increases in storage and handling fees for the overlapping silos. Senwes would be the main silo owner in the overlapping areas and according to the Commission’s investigation, Senwes’ handling and storage fees are generally higher than those of Suidwes in overlapping areas.

[182]                For this first theory of harm, the Commission advanced three mechanisms through which this would occur. The first was the proposed merger may result in an increase in the storage rates in relation to the Suidwes silos in the overlapping areas.

[183]                Secondly, that the proposed merger may result in an increase in the storage rates across the Suidwes portfolio of silos. Thirdly that the proposed merger may result in an increase in the storage rates across the combined Senwes-Suidwes portfolio of silos.

[184]                The Commission submitted that Senwes had a higher annual storage tariff than Suidwes, and that this rate was above the comparator group average. Senwes also had a higher daily tariff than Suidwes, and this rate was the highest for the comparator group. On the Commission’s calculations, Senwes’ tariffs were between 15% and 21% higher than those of Suidwes.

[185]                Based on these findings the Commission argued that post-merger Senwes would have the incentive and ability to raise storage tariffs at its silos. This ability and incentive would be in addition to the expected increase of tariffs at the newly acquired Suidwes silos, as the merged entity would translate Senwes’ existing pricing policy onto all new storage facilities. Customers would have no choice but to accept these price increases as there would be no alternatives. This would likely harm the welfare of farmers who are customers of the merging parties.

[186]                The merging parties disputed the Commission’s submissions and submitted that Senwes and Suidwes apply company-tariffs, meaning that their pricing policy was not differentiated by silo.

[187]                Further that the differential between the daily storage tariffs between Senwes and Suidwes currently amounted to 1c per ton.

[188]                With respect to seasonal storage rates, Kruger testified that the average seasonal storage rate of Senwes is […]

[189]                However, the Commission disputed this, pointing out that the Senwes documents showed that for the past two years, the bulk of storage fees were actually from the annual rate and that Kruger’s testimony was not borne out by its documents.

[190]                With respect to handling fees the Commission submitted that Senwes had the highest rate amongst the five largest storage companies whereas Suidwes offered the second lowest handling fee in that group.

[191]                The merging parties submitted that the difference in handling fees of […] This was not challenged by the Commission.

[192]                With regard to both the handling fees and storage tariffs, Kruger maintained that the competition faced by the merging parties from alternative storage facilities was a constraining factor on Senwes’ pricing strategy and would remain so post- merger.

[193]                The Commission submitted that despite the discrepancies between the merging parties’ documents and witness testimony, the primary issue was that the merger would result in a permanent change to the structure of the market. The loss of competitive rivalry between Senwes and Suidwes would create an upward pricing pressure in the overlapping areas.

[194]                The Commission argued that the loss of competitive rivalry is concerning when one considers the submission from the merging parties that they charge uniform rates across all their silos which could then broaden the upward pricing pressure to all silos of the merged entity.

Our analysis

[195]                As indicated, on the evidence before us, we could not conclusively find that alternative storage facilities could be excluded as having a constraining effect on concrete silos, the only issue being the extent of this effect. In light of the lack of strong evidence by farmers in this regard, we assessed the proposed transaction on a worst case scenario that it potentially removes the competitive rivalry between Senwes and Suidwes.

[196]                Given Senwes’ national pricing policy, and to address the concern that Senwes will post-merger, have an ability to increase its pricing in the overlapping areas the merging parties tendered a pricing condition. This condition was initially accepted and recommended by the Commission.

[197]                This is further discussed under the remedies section below.

Decrease in the premia paid to farmers

[198]                The second theory of harm advanced by the Commission was that the merger would result in lower prices being paid for grain to farmers.

[199]                According to the Commission the proposed transaction would result in a direct loss of competitive rivalry between Senwes and Suidwes, with the merged entity likely to be in a position to exercise market power against farmers in the procurement of grain in the Senwes/Suidwes areas post-merger.

[200]                Currently, the merging parties compete for the procurement of grain from farmers in the overlapping areas. Further, the merging parties compete on several levels including offering volume discounts on storage to farmers. The Commission was of the view that the merged entity would likely significantly lower prices paid to farmers for their grain.

[201]                The Commission based its theory on an analysis which it performed to try to calculate the premia paid in respect of pre-season contracts in the overlapping areas as compared to the premia paid in non-overlapping areas. According to the Commission, farmers in areas of greater competition could receive prices that were up to 5% higher for their grain, as compared to other farmers in areas where Senwes faced little or no competition.

[202]                The Commission accordingly submitted that because the merger would result in a higher degree of market concentration in the three markets which it had identified, the premia paid to farmers would decrease as a result of the lessening of competition.

[203]                Kruger challenged the Commission’s calculation on the basis that “they ignored a lot of stuff. They ignored the qualities of product. They ignored the handling fees, the commissions, if it was delivered or pre-delivered, they ignored the interest. So Chair, it’s an incomplete dataset. It’s materially incorrect. Just to look at that non-overlapping minus 5%, that discount of 5%, we never paid discounts [premia] to the amount of 5% and that’s the average number.”

[204]                Applying what the merging parties considered the correct methodology the merging parties found that the premia are not systematically higher in the overlapping areas but are rather dependent on the proximity of the silo to processors.

[205]                Further the merging parties argued that the Commission’s conclusion ignored the fact that Senwes is required to treat all users of its storage facilities in an equivalent fashion. This arises both from the consent order[53] confirmed by the Tribunal and the SAFEX regulations. The consent order related to a complaint against Senwes that as a vertically integrated firm that is allegedly dominant in the grain storage market, had abused its position by charging higher storage tariffs in its silos to exclude rival firms in the downstream market for the trading of grain.

[206]                In terms of the consent order Senwes is required to offer all parties which store grain with it equal access to its various storage options on identical terms, save for such differentiation that may legitimately be made under the Competition Act.

[207]                No evidence was put up by the Commission to rebut Kruger’s explanation that the Commission’s calculation of the discount was based on an incomplete data set. Further, the evidence that the market was broader than contended for by the Commission remained an issue which could not be sufficiently tested with farmers. Kruger’s evidence which the Commission seemed to accept was that farmers first choose who to sell their grain to, with the choice of facility being secondary.[54] This appears consistent with Prof Kirsten’s view that market dynamics are changing such that farmers are looking for integrated solutions, not simply storage.

Market manipulation

[208]                This theory of harm was premised on the following, that (i) the Suidwes and Senwes silos are located in the Free State and the North West; (ii) these two provinces account for a significant share of white maize production; and (iii) Senwes will be able to influence the volume of maize which is being out-loaded.

[209]                The Commission sought to advance three sub-theories under the broad category of what came to be known as the “market manipulation theory”.

[210]                The first was that the merger could result in the merged entity having power in the overall market for the sale of maize. According to the Commission:

[211.1] The merged entity would account for the production and storage of approximately 80% of white maize in the Senwes area which could facilitate the constructive market foreclosure of competitors as the supply and stock of white maize available would be centralised in the merged entity post-transaction.

[211.2] The foreclosure could be subtle, taking the form of Senwes offering its trading rivals inconvenient out-loading slots, or using silo maintenance and fumigation procedures at certain key points and at certain key silos to the detriment of trading firms rivalling Senwes’ trading arm.

[211]                In this regard, Oberholzer mentioned that OVK had experienced difficulties in obtaining sufficient slots to out-load grain from the Senwes silos in the past for which it was pursuing a claim of up to R6 million.[55] The merging parties submitted that OVK had not (despite requests) provided Senwes with any detail in relation to the alleged complaint.[56] Nor could it indicate that a complaint had in fact been filed with the JSE in this regard.

[212]                Nhlapo similarly testified that Senwes had in the past been able to act contrary to the SAFEX rules without incurring penalties. He cited an instance where a company called Vrystaat Mielies had won a tender to supply white maize to a trading company and encountered loading issues as Senwes advised Vrystaat Mielies that it was fumigating the silo at the time that the grain was due to be loaded, and offered an alternative loading location 60km from the original silo, thereby increasing the cost for Vrystaat Mielies. It turned out that this incident was more than 10 years ago.

[213]                We thus find that the evidence does not support the Commission’s claim of market manipulation as discussed above.

[214]                The second sub-section of the market manipulation theory advanced by the Commission was that the merged entity would be able to influence maize prices in South Africa during stock shortages. Oberholzer highlighted instances of significant price movements for white maize, when he referred to:

“[a]…R1 800 fluctuation in one day due to the fact that there was a problem with obtaining physical stock and that’s our submission Sir, that any silo owner with so much insight in what the quantities in South Africa is available will have a huge effect of the maize price and at the end of the day on food prices in South Africa”.

[215]                The merging parties’ submitted that the “base price” for maize is determined from the SAFEX price. The merging parties have no ability to influence the SAFEX price. They submitted further that such price manipulation would fall within the remit of the JSE to investigate and to take appropriate remedial action.

[216]                We have found no evidence in support of this theory.

[217]                The third and final theory raised by the Commission was that the merger would result in a greater ability to engage in anti-competitive conduct. NWK testified to this concern, noting that once competition in the overlapping area between Senwes and Suidwes was removed leading to the merged entity being more profitable (as a result of paying lower prices for grain and receiving higher storage fees), the merged entity would have a greater ability to undercut NWK by offering greater premia for farmers’ grain.

[218]                While this would benefit farmers in the short-term, so the Commission argued, it could ultimately cripple NWK (in much the same fashion as predatory behaviour would) leading to exclusionary effects, or the purchase of NWK by Senwes, both of which would be to Senwes’ benefit.

[219]                The merging parties argued that behaviour of this nature on the part of Senwes would lead to a substantial increase in competition in the NWK area, which the merging parties argued, was evidence that the proposed transaction was likely to be pro-competitive.[57]

[220]                In our view, the market manipulation theories of harm were either not supported by evidence or fell outside our remit, as discussed above.

[221]                Although we are not convinced by the Commission’s assertions in this regard, we are satisfied that the conditions imposed taken as a whole would mitigate against any potential anti-competitive effects of the merged entity’s increased market share.

COUNTERFACTUAL: IS SUIDWES A FAILING FIRM?

[222]                An area of contestation in the proceedings was the relevant counterfactual, that is, what would take place in the event that the proposed merger did not proceed. The merging parties had put up a failing firm defense and indicated that Suidwes’ assets were on the verge of exiting the market if the transaction were not expeditiously approved.

[223]                The approach to the failing firm doctrine was established in Iscor/Saldanha[58] and was reaffirmed in Santam Limited and Emerald Insurance Company Limited/Emerald Risk Transfer (Pty) Ltd[59] (“Emerald”), in which the Tribunal noted the following:

“The failing firm doctrine enjoys express statutory recognition in the Competition Act, 1998 (Act No. 89 of 1998) (the ‘Act’). Section 12A(2)(g) of the Act directs us to consider ‘whether the business or part of the business of a party to the merger or proposed merger has failed or is likely to fail’ as part of a non-exhaustive list of factors that must be considered in merger assessment. As pointed out by the Tribunal in the merger between Iscor Limited and Saldanha Steel (Pty) Ltd the failing firm doctrine, as such, in the Act is not a ‘defense’ to a merger that has been found on an initial market analysis to be anti-competitive. Rather, it is recognized as one of a non-exhaustive list of factors that must be taken into account before one can determine whether or not a particular merger is likely to substantially prevent or lessen competition.

In times of financial and economic distress, such as we are currently experiencing, many firms could find themselves in some sort of financial difficulty and these firms may seek to safeguard their long-term survival possibly by merging with (healthier) competitors. The task of the competition authorities is to assess whether the claim that a firm has failed or is likely to fail is genuine or a contrivance to obtain approval for an otherwise anticompetitive merger.

The failing firm doctrine is internationally recognised in competition law jurisprudence and, although not applied uniformly in all jurisdictions, has nevertheless been applied with a considerable degree of uniformity regarding the salient criteria for a credible failing firm claim. Satisfaction is required of each of the following criteria, namely that:

(i)            the firm is a failing one;

(ii)                the reorganisation of the alleged failing firm is not a realistic option; and

(iii)              a less anticompetitive outcome than the proposed transaction is absent.”

[224]                The Commission accepted that Suidwes was unlikely to continue in its current form given its precarious financial position. It posited several scenarios. Its main contention was that Suidwes could be placed in business rescue or liquidation. Further that even if business rescue (or liquidation) were to fail, Suidwes’ assets   (i.e. silos) would not exit the relevant market as they could be purchased by third parties. The Commission submitted that the counterfactual also includes the OVK offer which was a reasonable offer, but was rejected by the Suidwes board. Another counterfactual included West Street which the Commission submitted was still interested in making another offer.

Our Analysis

[225]                It is common cause between the parties that Suidwes is in financial distress. For that reason, we do not discuss this question any further.

[226]                In its recommendation and opening, the Commission accepted that Suidwes’ attempts to re-organise the business had failed.[60] During the hearing it suggested that, in the counterfactual, Suidwes might elect to be placed under business rescue. This would permit the business to be turned around as it would enable the Business Rescue Practitioners (BR practitioners) to delay the repayment of the debts and therefore, address the liquidity problem.

[227]                The merging parties submitted that business rescue was not a viable alternative. According to Van Schalkwyk he had discussions with various BR practitioners, and according to them, business rescue was unlikely to be a viable option for Suidwes for various reasons.[61] Van Schalkwyk testified further that there was no time left for a BR practitioner to come up with a different approach to managing Suidwes, given its very significant loss-making position.

[228]                In closing argument, the Commission appears to have accepted that business rescue carries uncertainty.[62] It maintained the view that OVK presented a reasonable offer since its concrete silos did not geographically overlap with Suidwes’ silos. The Commission submitted that in any event Suidwes’ silo assets were valuable and if this transaction were not to be approved, there would be interested buyers for the silos if the transaction is not approved.

[229]    We next consider whether Suidwes made good faith efforts to find reasonable alternatives.

Has there been a good faith effort to pursue reasonable alternatives?

[230]                In Emerald, the Tribunal stressed that one of the requirements to prove a failing firm defense took the following form “the assumed failing firm must demonstrate inter alia that it has made reasonable and verifiable good faith attempts to elicit reasonable alternative offers and, furthermore, that there is no viable alternative purchaser that poses less anticompetitive risk than does the proposed transaction”.

[231]                The assessment of whether there is a viable alternative must take account of the factual situation when the alternative offers were made. As the Tribunal noted in Phodiclinics[63]:

“This brings us to the proposal or offer tabled by Netcare and Tradeworx in the course of the hearing … the existence and the terms of this belated offer are irrelevant to these proceedings and the Tribunal does not regard it as a valid offer existing at the time when the merger transaction was concluded. “Reasonable alternatives” as contemplated in the Iscor case must exist at the time when offers are procured by the liquidator and a transaction is concluded, not at some indeterminate time in the future.”

The EU and US guidelines require that a failing firm demonstrate, at the time when the transaction is being evaluated for competition implications, to the competition authority that it “has made unsuccessful good-faith efforts”. The word “has” is the singular present tense of the word “have”. In the context of the requirement that the merging parties prove the elements of the failing firm doctrine, the parties are required to show, at the time at which they seek approval from the Competition Authorities, that they “have made” good faith efforts to find reasonable alternatives to the offer they have accepted and for which they seek approval. The Act does not require parties to provide an undertaking that they “will continue to make” efforts to find reasonable alternatives. Such an interpretation would lead to an absurdity, since the authority would never be able to approve a transaction to which a party must continuously strive to find an alternative offer.”

[232]                Suidwes received offers from both local and international firms. They included offers from OVK, NWK and West Street Capital.

[233]                We concluded on the evidence before us that neither the NWK or West Street Capital offer was a reasonable alternative on the basis that each of these offers had only been for a portion of the Suidwes business. For instance, while NWK only wanted to purchase the concrete silos and the retail business, West Street Capital only wanted to purchase the concrete silos. As the Commission correctly pointed out from the economic literature, only offers made for the entire asset base of a firm should be considered to be valid offers and form part of the assessment.

[234]                At the hearing, Nhlapo made submissions that West Street Capital had subsequently approached the IDC for funding in late February 2020 and that it was still interested in purchasing Suidwes. The Commission’s view was that even though West Street’s offer was not to purchase the entire business, its interest in still making an offer shows that Suidwes’ assets will not exit the market and can find alternative purchasers. We concluded that a remedy that preserves the Suidwes business in its entirety was in the public interest.

[235]                OVK made an offer for the entire business of Suidwes, which entailed a loan […]

[236]              […][64]

[237]                The Commission argued that, “both equity and debt can be used as viable methods of financing a transaction. In fact, in the majority of instances, debt is cheaper than equity finance. The OVK route would have equally been viable financing option given that what Suidwes benefits in the Senwes option through lesser interest burden, they lose through diluted equity shareholding in the merged entity”.

[238]              […][65]

[239]              […][66] […][67] […][68]

[240]              […][69]

[241]                According to the merging parties even if a renewed offer were to be made, it would require Suidwes’ shareholder approval, a due diligence and this process could not be countenanced given Suidwes’ current financial position.

[242]                In our view, these factual disputes regarding the counterfactual do not alter our assessment of the merger.

[243]                By its nature, assessing the counterfactual is a predictive exercise. It cannot be measured with exact precision. In Imerys, the CAC said the following regarding the Tribunal’s exercise of discretion when weighing up whether to approve a merger with conditions or to prohibit it:

“I do not think the Tribunal is obliged to approve a merger just because it finds it more probable than not that the conditions will neutralise the likely SLC. One should bear in mind, in this regard, the real problem in such cases will not necessarily be competing views as to the probable future state of the market but an inability to make reliable predictions at all.”

[244]                The CAC further held:

“ [42] I do not say that the Tribunal would be obliged to reject conditional approval just because there was a reasonable possibility (falling short of a preponderance of probability) that the conditions would fail to remedy the likely SLC. The Tribunal might properly exercise its discretion in such a case to give conditional approval. In exercising its discretion, the Tribunal could be expected to take into account, on the one hand, the precise likelihood and extent of the SLC; and, on the other, the precise extent of the risk that the conditions will fail to remedy the likely SLC. The public interest may also enter into the balancing exercise, particularly the public importance of the markets which would be directly or indirectly prejudiced if the conditions failed to remedy the likely SLC.”

[245]                What would have happened with the OVK offer including its funding of the transaction (both now and at the time it made an offer to the Suidwes Board) had Senwes not entered the arena with its offer for Suidwes is speculative.

[246]                On the contrary the final set of remedies tendered by the merging parties address the potential competitive harm identified by the Commission and have public interest benefits which as the CAC pointed out, may also enter into the balancing exercise.

[247]                As reaffirmed in Emerald referred to above, the failing firm is not a ‘defense’ to a merger that has been found on an initial market analysis to be anti-competitive. Rather it is only one of the non-exhaustive list of factors to be taken into account before one can determine whether or not a particular merger is likely to substantially prevent or lessen competition. As indicated, on the evidence, we could not conclusively find a substantial prevention or lessening of competition.

[248]                We were satisfied that the proposed remedies would address any competition concerns and the public interest concerns arising from the merger. Whether or not Suidwes meets all the legal requirements to be regarded a failing firm in terms of the Act, does not alter this conclusion.

Conclusion on the counterfactual

[249]                In light of the above, we are of the view that absent the proposed transaction, Suidwes’ financial position is likely to deteriorate; and that it is unlikely that the Suidwes business can timeously be turned around through business rescue, or that liquidation is a viable alternative, or that re-opening the bid for a potential purchase by OVK or any alternative purchasers is viable.

[250]                The consequences of such failure will have a negative impact on the public interest, as discussed later. Whether or not Suidwes’ silo assets would exit the relevant market if it failed is an issue that we do not have to decide given our conclusion regarding remedies. Put differently, whether or not Suidwes meets all the legal requirements for a failing firm does not alter our conclusion that the merging parties’ final set of tendered remedies address any likely competition concerns as well as the public interest considerations.

EFFICIENCIES

[251]                The merging parties alleged that the proposed transaction would lead to a number of efficiency gains related to cost savings for the integration of the merging parties retail businesses. The merging parties however did not demonstrate how these savings were likely to accrue as benefits to customers.

[252]                The Commission argued that the efficiencies claimed by the merging parties were speculative in nature and that there was no evidence of any benefits accruing to customers.

[253]                Given the inconclusive evidence on whether the Commission has established a substantial prevention or lessening of competition, and taking into account the public interest, it is not necessary for us to deal with efficiencies.

PUBLIC INTEREST

[254]                The merging parties claimed that there would be a number of positive public interest benefits arising from the proposed transaction. The Commission took no issue with these alleged benefits. However, it was of the view that these could be achieved by an alternative purchaser. As mentioned, we were not persuaded by the Commission’s assessment of the counterfactual.

[255]                We were persuaded by the following public interest benefits which will arise from the transaction.

Employees and local communities

[256]                As indicated, the proposed transaction will secure the employment of approximately 934 out of 1246 employees. Although it will result in 136 merger- specific retrenchments mainly due to head office relocation, and 171 retrenchments for operational reasons as Suidwes is loss-making, the net effect on employment will be positive. The merger will have a beneficial impact not only on the employees themselves, but also on the communities in which they live and the indeterminate number of people which they support. The Suidwes operations are located in rural areas. It is unlikely that they would be able to find alternative employment absent this transaction.

Effect on relevant regions

[257]                We also took note of the impact the failure of Senwes would have on the region. As was the case in Iscor Ltd/Saldanha, prohibiting the merger could not be justified given the substantial adverse public interest effects that would ensue and more particularly the adverse effect such a prohibition would have in the Saldanha region.

[258]                Suidwes submitted that it paid approximately R44 million to municipalities and R109 million in taxes and these payments would end with the collapse of Suidwes, resulting in a negative impact on the region.

[259]                Suidwes’ further submitted that suppliers that are currently owed R141 million will be paid in the event that the proposed transaction is approved, whereas they would only recover “cents in the Rand” in the event that the company were to go into liquidation.

[260]                Based on the testimony before us, Suidwes has numerous retail outlets and storage facilities in various small towns in the Free State and North-West. The consequences of it going into liquidation, in our view, would have negative consequences for the small rural towns and communities in which it operates. In addition, Suidwes also provides income for other local businesses that rely on Suidwes to continue to operate.

[261]                Prof Kirsten reinforced this point by providing his experience as an agricultural economist: “some experience I have is some rural towns, the whole town survives on the back of the so-called co-operative or agri business. They invest in the retail store. They invest in the fuelling depot. They invest in the coffee shop and in some way they provide life to the town and without that major anchor in the town, nothing will happen. You know, we had for years this problem of urbanisation and de-ruralisation of … die ontvolking van die platteland is die korrekte Afrikaanse woord. So, it is something that is probably the last opportunity for rural towns to have some activity going.”[70]

Negative implications for farmers

[262]                In relation to the provision of production loans to farmers in the Suidwes area, the farmers would not be able to obtain funding and this would have a very significant impact on the production of white maize in South Africa. This would be particularly more detrimental given that the Land Bank is in financial difficulty and in the current circumstances is unlikely to extend loans to farmers either.

[263]                It is undisputed that the Land Bank is currently in severe financial difficulties and has total debt of approximately R45 billion. Suidwes established a debtors’ book with the Land Bank to the value of R3.7 billion in total of which R1.7 billion is classified as “non-performing”. As a result of its financial difficulties, the Land Bank has indicated that it may not be in a position to honour various loan arrangements which it had previously concluded with Suidwes. The merging parties submitted that the proposed transaction will therefore ensure that these financing arrangements are honoured.

[264]                Further, Senwes has undertaken to assist the Land Bank in recovering the amounts owing to assist the Land Bank in its current financial difficulties. The Senwes loan also enabled Suidwes to repay its significant debt to the Land Bank.

[265]                Professor Kirsten also noted in his expert report that the failure of Suidwes would have very detrimental consequences – “As a result, the support and services network for farmers in the Suidwes production regions will disappear and could potentially destroy the productive agricultural landscape in those districts”.

[266]                The proposed transaction would therefore play an important role in ensuring food security (through ensuring the replacement of the Land Bank’s production loans to farmers in the Suidwes area).

Retention of South African agricultural assets in South African hands

[267]                The proposed transaction will result in the ongoing operation of a South African agricultural business to ensure that the country maintains local South African agricultural infrastructure, given the importance of food security at a national level.

Conclusion

[268]                Based on the above, we were of the view that the merger would result in substantial public interest benefits in the form of the retention of employment in rural areas where this is required as well as the retention of investment and infrastructure in rural communities where it is most needed.

REMEDIES

[269]                As mentioned above, a number of remedies were put before the Tribunal during the course of the proceedings. While the Commission initially agreed to the remedies tendered by the merging parties in its recommendation before us, it later rejected the remedies on the basis that once it tested the remedies with third parties, the remedies were found to be inadequate to address the competition concerns.

[270]                We pause to mention that the third parties which the Commission tested the remedies with were competitors (NWK and OVK) of the merging parties. Both were also unsuccessful bidders for Suidwes. The other third party was West Street Capital who was also an unsuccessful bidder for Suidwes.

[271]                The remedies were not tested with the farmers as customers of the merging parties, and as indicated the Commission did not call any farmer as a witness on either the relevant markets or remedies.

[272]                The conditions underwent a number of iterations which we briefly discuss below. The final offer covered divestiture, pricing, employment and an agricultural development programme. The latter was a proposed remedy by Professor Kirsten to promote small farmers.

Divestiture

[273]                In its initial recommendation to the Tribunal, the Commission and merging parties had agreed on the divestiture of the Senwes silo (Jan Kempdorp) in the Christiana geographic market and two Suidwes silos (i.e. Suidwes Amalia and Suidwes Vryburg concrete silos) falling outside of the identified overlap area of 40km of the Ottosdal and Leeuringsstad geographic markets.

[274]                The final offer proposed was the divestiture of the Suidwes Strydpoort silo located in the Ottosdal area, Suidwes Wolmaransstad silo (as well as the Africum Mill located in Wolmaransstad) located in the Leeudoringstad area, and the Senwes Jan Kemp Dorp silo located in the Christiana area. These silos have a combined silo storage capacity of 178 000 tons.

[275]                During the hearing, the Commission questioned the viability of the silos to provide a new entrant with the scale to become an efficient competitor. According to its witness, Nhlapo, a new entrant would require at least 500 000 tons to be able to enter and compete effectively. Nhlapo’s evidence became the basis for the Commission’s contentions on the issue of the minimum efficient scale.

[276]                The merging parties challenged this evidence and pointed out that there were a number of players in the market who were operating successfully with less than 178 000 tons. In particular, the merging parties put up Exhibit NH7, reproduced below which showed that there were a number of players operating in the market with less than 500 000 ton capacity.

Table 14: SAFEX registered storage capacity of relevant role players in the market

 

Relevant Role Players	SAFEX Registered Storage capacity (tonnes)	WOL+STRYDP+JAN KEMP Storage Capacity	% of Operational Businesses
Obaro	164 621	178 000	108,1%
AllemBrothers	160 000	178 000	111,3%
Bester Feed & Grain	35 000	178 000	508,6%
BKB	292 740	178 000	60,8%
GWK	165 000	178 000	107,9%
Kaap Agri	350 506	178 000	50,8%
Keystone Milling	11 550	178 000	1541,1%
OVK	412 253	178 000	43,2%
Schoeman Boerdery	58 000	178 000	306,9%
Standerton Oil Mills	80 000	178 000	222,5%
SSK	230 500	178 000	77,2%
TWK	145 000	178 000	122,8%

Source: Exhibit NH7 submitted 3 July 2020

[277]                While the Commission challenged the alleged minimum scale referred to by Kruger, it also did not put up any evidence to contradict Exhibit NH7, save to say that 500 000 tons was the minimum viable scale for a new entrant. Absent this scale, the Commission contended, the merger should be prohibited.

[278]                We note that the capacity of the revised silos proposed by the Commission after its initial recommendation to us (which were not accepted by the merger parties) had a capacity of 249 923 instead. Table 12 below shows the silos proposed for divestiture by the parties and those proposed by the Commission, while Table 13 shows the respective capacities of these silos.

Table 12: Proposed divestiture package of the Commission and the merging parties

Silos to be divested	Geographic Market 1 (Ottosdal)	Geographic Market 2 (Leeudoringsta d)	Geographic Market 3 (Christiana)
Merging parties proposed divestiture package	Suidwes Strydpoort	Suidwes Wolmaransstad	Senwes Jan Kemp Dorp
Commission’s Proposed divestiture package	Suidwes Bamboesspruit and/or Senwes Bothaville	Suidwes Wolmaransstad	Suidwes Christiana

Letter from the Merging parties dated 19 June 2020 and letter from the Commission dated 30 June 2020.

[279]                The merging parties’ proposed silos have the following capacity capabilities:

Table 13: […]

[…]

[280]                We now turn to consider the silos to be divested in each of the relevant geographic areas.

Ottosdal

[281]                The merging parties proposed the divestiture of Strydpoort, which the Commission rejected. It submitted that the Strydpoort silo offered, on the merging parties’ own submissions had low utilization rates and was seen more as an overflow silo to that of the Bamboesspruit silo. The Commission proposed to the merging parties that they instead divest of the Suidwes Bamboesspruit and/or Senwes Bothaville silo located instead.

[282]                The merging parties submitted that the Strydpoort silo and (Wolmaranstad silo) provided a guaranteed flow of grain to the Africum Mill (which forms part of the divestiture) as the Africum mill is an end-processor and a well-located producer of maize-meal. Both Strydpoort (and Wolmaransstad) silos fall within a 40km radius of the Bamboesspruit silo. Strydpoort also has road and rail access. Finally, they submitted that Strydpoort had been profitable over the past five years as shown in Table 15 below.

[283]                As to Senwes Bothaville, they submitted that it fell outside of the geographic overlaps identified by the Commission 60km. The Suidwes Bamboesspruit silo is located approximately 28km from the Suidwes Wolmaransstad silo, and 6km from the Strydpoort silo.

Leeuringdoringstad

[284]                Although the Commission also had concerns regarding the Suidwes Wolmaransstad silo in this geographic market it remained part of the Commission’s preferred silos for divestiture. The Commission’s concern was that this silo was in an area with declining wheat production.

[285]                The merging parties disagreed with the Commission’s contentions. They submitted that there was no evidence to support the Commission’s contentions that grain in the Wolmaransstad area is on the decline. They submitted that, like Strydpoort, it is also well located on the N12 and adjacent to several producing areas. The merging parties submitted further that the Commission also did not take account of the existence of alternative storage capacity in these areas or the Silostrat facilities located in the areas proximate to the Leeudoringstad silo which would constrain the merging parties’ behaviour.

Christiana

[286]                Finally, with respect to the Senwes Jan Kemp Dorp silo in this geographic market located within 40km of Suidwes’ Christiana silo, the Commission submitted that this silo is in a wheat and barley area, with maize production declining due to changes in weather patterns. Further that the merging parties had also indicated that they were going to mothball this silo in the near future as it was not profitable. Instead of the Jan Kemp Dorp silo the Commission sought the divestiture of Suidwes Christiana.

[287]                In oral evidence Kruger said plans to mothball the Jan Kemp Dorp silo had been contemplated some five to six years ago due to an aeration issue they had been experiencing at the time. Since then, Suidwes had taken the decision not to mothball it and instead invested in the silo by introducing six aeration facilities. Kruger submitted that Jan Kemp Dorp was profitable.[71]

[288]                The merging parties further submitted that the Commission’s suggested remedy package sought to introduce a significantly more costly and onerous remedy package which would require the merging parties to divest of the most significant assets of Suidwes (the Christiana and Bamboesspruit silos) as well as a large Senwes asset (the Bothaville silo), in circumstances where Senwes is attempting to save a failing firm in Suidwes. Importantly, the Bothaville silo did not even fall within the overlapping areas identified by the Commission.

[289]                The Commission’s other issue with the proposed silos, apart from scale, was that the silos tendered by the parties for divestiture were not profitable.

Table 15: […]

[…]

[290]                The merging parties disputed this. They put up the profitability figures for all the silos to be divested for the preceding five years. Reproduced below are the profitability figures for the year ended 2019/2020, which showed each of the silos were profitable.

[291]                The Commission disputed that the silos were profitable. However, it was unable to provide a detailed assessment of the profitability of the silos in the time available.

[292]                We were not persuaded, on the evidence, that the Commission had made out a case for prohibition. This is because the Commission’s conclusions regarding the proposed merger resulting in a substantial prevention or lessening of competition were premised on relevant product and geographic markets which were not adequately tested. The evidence in this regard, was mixed and inconclusive.

[293]                In light of this evidence, the financial position of Suidwes which was not disputed, and the the counterfactual, i.e. that business rescue or the liquidation of Suidwes would not be in the public interest, we concluded that the divestiture of the silos tendered by the merging parties with a capacity of 178 000, in the absence of data to the contrary, combined with the other conditions in their totality, would ameliorate any competition and public interest concerns that arise.

The pricing condition

[294]                The proposed pricing condition was as follows:

“following the implementation date, Senwes must ensure that for a period of 5 years, the differential between (i) the handling and storage tariffs applicable at the Leeudoringstad, Christiana and Bamboesspruit silos and (ii) those applicable at the Regina, Werda and Melliodora silos must remain the same as the differential applicable on the day before the Implementation Date, unless Senwes invests into the Leeudoringstad, Christiana and Bamboesspruit silos to (i) increase the loading and outloading speed by 10% from the existing rate or (ii) increase the ability to handle higher moisture grain from the existing rate by 10%, or (iii) a 10% increase in the efficiency of the silo by way of installing 10% more stock measurement equipment (i.e. Crux laser technology), replacing the existing grading machinery or installing temperature strings and CO2 meters for more than 10% of the bins at the silo in which case this provision must no longer apply to the silo in respect of which the investment has been made.”

[295]                This condition sought to address the concern raised by the Commission that Senwes charged significantly more than Suidwes for storage and handling. According to the Commission, this differential was a direct function of the savings and efficiencies benefitting those farmers using those facilities at which Senwes had implemented technological improvements. It is for this reason that the pricing condition contained a carve out in respect of improvements made by Senwes.

[296]                Throughout the hearing we heard evidence on the growth of alternative storage In particular, alternative storage facilities have been placing an increasing competitive constraint on concrete silos especially in the last three to five years. Further there still remained the issue that because of the history of concrete silos and the fact that these have been built in regions where grain production may be declining there was a possibility that this could affect the pricing at these silos through increased storage and handling fees.

[297]                Absent clear evidence of a substantial prevention or lessening of competition, in order to address the potential of Senwes increasing its handling fees charged at silos without any concomitant technological benefit to the farmers, we were satisfied that the condition tendered was sufficient to allay any concerns at least for a period of five years subject to there being no technological investment which may improve the silos for customers in this period.

The employment condition

[298]                With the exception of the 136 positions which Senwes has identified as being potentially duplicative, the parties have undertaken not to retrench any employees as a result of the merger for a period of 24 months from the date on which the transaction is implemented (the “implementation date”).

[299]                We are satisfied that this addresses any employment concerns arising from the proposed transaction.

Agricultural development programme

[300]                According to Professor Kirsten, a more effective remedy would be one that facilitates an agricultural development programme for black farmers in the region as such a condition could have a much more long-term impact on agriculture in the region from the perspective of inclusive growth and transformation. He was of the view that this would have more of a long-term impact on agriculture from the perspective of inclusive growth and transformation in the region.

[301]                The Commission disagreed with Prof Kristen’s proposed remedy as it alleged it would not address any competition concerns resulting from the structural change to the market. The Commission was of the view that emerging black farmers were likely to be confronted with the anti-competitive effects from the merger in the relevant markets as they would have to participate within a more concentrated industry and would not have viable alternative storage for their grain.

[302]                However, for the reasons discussed, the Commission’s theories of harm were not sufficiently supported by the evidence before us.

[303]                Regarding the establishment of a fund as proposed by Professor Kirsten, the merging parties tendered a condition in which they committed an amount of R20 million per annum for a period of three years in the form of production loans to emerging black farmers.

[304]                In light of the evidence regarding access to funding for emerging farmers, the financial difficulties experienced by the Land Bank on whom farmers rely for funding, the precarious financial position of Suidwes which further reduces funding opportunities for farmers, we considered this condition taken overall with the other conditions, to be in the public interest.

[305]                We strengthened this condition through the monitoring provision which requires the parties to report to the Commission, on an annual basis, inter alia, the number of loans to emerging black farmers (from the pre-merger baseline), whether such farmers are new or existing customers of the merging parties as well as confirmation that an aggregate amount of at least R20 million has been advanced to emerging black farmers for the preceding year.

Conclusion

[306]                We found that the Commission’s theories of harm were not sufficiently supported by the evidence before us. We concluded that the public interest benefits outweigh the counterfactual which is the likely failure of Suidwes.

[307]                The consequences of such failure will have a negative impact on the public interest. Whether or not Suidwes’ silo assets would exit the relevant market if it failed is not an issue that we have to decide given our conclusion regarding remedies.

[308]                The final set of tendered remedies address any likely competition concerns as well as the public interest considerations.

[309]                We are satisfied that these conditions, when considered in their totality, are appropriate in the context of the proposed transaction.

[310]                We accordingly approved the proposed transaction subject to the conditions attached in Annexure “A” hereto.

                                                                                                                                                                                                                                                                          12 April 2021

Ms Mondo Mazwai                                                                              Date

Mr Enver Daniels and Prof. Imraan Valodia concurring

Tribunal Case Manager:                                        Busisiwe Masina

Tribunal Economists:                          Karissa Moothoo-Padayachie and Lumkisa Jordaan

For the Commission:                          Romeo Kariga and Grashum Mutizwa

For the Merging Parties                                          Anthony Norton and Anton Roets of Nortons Inc.

---