---

 

COMPETITION TRIBUNAL OF SOUTH AFRICA

 

Case No.: IM139Dec21

 

n the matter between
DRASLOVKA HOLDINGS A.S.	First Applicant
SASOL SOUTH AFRICA LIMITED	Second Applicant
And
THE COMPETITION COMMISSION OF SOUTH AFRICA	Respondent
HARMONY GOLD MINING COMPANY LIMITED	First Intervenor
SIBANYE STILLWATER LIMITED	Second Intervenor
PAN AFRICAN RESOURCES PLC	Third Intervenor
In re the intermediate merger between:
DRASLOVKA HOLDINGS A.S.	Primary Acquiring Firm
And
THE SODIUM CYANIDE BUSINESS OF SASOL	Primary Target Firm
SOUTH AFRICA LIMITED

 

Panel:	J Wilson (Presiding Member)
	L Mncube (Tribunal Member)
	AW Wessels (Tribunal Member)
Heard on:	24, 25, 26 April 2023
	4, 23, 25, 26 May 2023 and;
	20 July 2023
Last submission on:	02 October 2023
Order issued on:	11 October 2023
Reasons issued on:	01 February 2024

 

REASONS FOR DECISION

 

INTRODUCTION

 

[1]             On 11 October 2023, the Competition Tribunal (“Tribunal”) prohibited the intermediate merger in terms of which Draslovka Holding a.s. (“Draslovka”), through Draslovka (South Africa) Proprietary Limited (“Draslovka SA”), proposes to acquire the assets and liabilities of the sodium cyanide (“NaCN”) business of Sasol South Africa Limited (“Sasol”) as a going concern. The reasons for the Tribunal’s decision are set out below.

 

PROCEDURAL BACKGROUND

 

[2]             On 2 September 2021, the merger parties notified the Competition Commission (“Commission”) of the proposed merger. The Commission found that the proposed transaction would likely result in an increase in the price of liquid NaCN to South African customers, and concluded that the transaction was likely to result in a substantial prevention or lessening of competition in the market for liquid NaCN in South Africa. The Commission also concluded, for the same reasons, that the proposed merger would have a negative effect on the South African gold mining sector given that liquid NaCN is an important input in the gold mining production process in South Africa. The merger parties tendered a set of remedies to address the Commission’s concerns but the Commission concluded that the proposed conditions did not adequately do so, and accordingly, on 26 November 2021, it prohibited the proposed merger.

 

[3]             Following the Commission’s decision, the merger parties, on 10 December 2021, filed a request for consideration of the merger with the Tribunal in terms of section 16(1)(a) of the Competition Act, 89 of 1998 as amended (the “Act”), in which they requested an unconditional approval of the merger.

 

[4]             Three gold mining companies – Sibanye Stillwater Ltd (“SSW”), Harmony Gold Mining Company Ltd (“Harmony”), and Pan African Resources PLC (“PAR”) – sought, and were granted, leave to intervene in the Tribunal proceedings. Prior to the Tribunal hearing, and after discussions with the merger parties, Harmony and PAR accepted revised term sheets offered by Draslovka regarding the supply of liquid NaCN. SSW, however, did not accept the term sheet that was offered by Draslovka, and persisted with its intervention in the proceedings, calling factual and expert witnesses. The Tribunal also requested representatives of Harmony and PAR to appear at the hearing to explain the reasons for their revised stance on the proposed merger.

 

[5]             The following factual witnesses gave evidence at the Tribunal hearing:

 

5.1. For the merger parties:

 

5.1.1.     Mr Pavel Bruzek, the chief executive officer of Draslovka;

 

5.1.2.     Mr David Mokomela, the Vice President of Sasol’s Base Chemicals Business; and

 

5.1.3.     Ms Nicole Wainer, the Corporate Transaction Lead for Mergers and Acquisitions (“M&A”) at Sasol.

 

5.2. For the Commission, Mr Philip Hayward, a unit manager in the procurement division of the South Deep Mine, which is owned by Gold Fields Ltd (“Gold Fields”).

 

5.3. For SSW:

 

5.3.1.     Mr Mpho Mochekela, the Unit Manager: Commodities at SSW;

 

5.3.2.     Mr Jaco Schoeman, an Operations Director at DRDGold Ltd (“DRD”), which is majority-held by Sibanye Gold Ltd (“Sibanye Gold”), a subsidiary of SSW; and

 

5.3.3.     Mr Ruan Vorster, the Vice President: Surface Operations at SSW.

 

5.4. For Harmony, Mr Elias Pobe, the Executive Head: Metallurgy South Africa at Harmony.

 

5.5. For PAR, Mr Jonathan Irons, the Group Metallurgist and Tailing Executive at PAR.

 

[6]       In addition, the following witnesses were called as economic experts:

 

6.1. For the merger parties: Mr Patrick Smith, a partner at RBB Economics;

 

6.2. For the Commission: Mr Yongama Njisane, a Principal Economist in the Economic Research Bureau of the Commission; and

 

6.3. For SSW: Mr Greg Harman, a Managing Director at Berkeley Research Group (UK) Ltd.

 

THE PROPOSED TRANSACTION

 

[7]             The proposed transaction involves the acquisition by Draslovka, through Draslovka SA, of the assets and liabilities of the NaCN business of Sasol as a going concern (the “Target Business” or the “NaCN business”). The land on which the Target Business’s assets are located would be leased to Draslovka SA in terms of a long-term lease agreement.

 

[8]             Draslovka would hold 75%, less one share, in Draslovka SA, and the remaining shares would be held by Navuka Investment Holdings Pty Ltd (“Navuka”), Draslovka’s broad- based Black Economic Empowerment (“B-BBEE”) partner, through Navuka’s subsidiary Sirtan Pty Ltd (“Sirtan”). Pursuant to the proposed transaction, Draslovka would, through Draslovka SA, exercise sole control over the Target Business.

 

[9]             An important component of the proposed transaction for purposes of the merger assessment is the conclusion of supply agreements between Draslovka SA and certain entities within the Sasol group of companies (the “Sasol Group”) for the supply of key inputs into the NaCN operations of the Target Business post-merger – in particular, caustic soda, ammonia, natural gas and electricity. We discuss these supply agreements in greater detail below.

 

THE MERGER PARTIES AND THEIR ACTIVITIES

 

[10]         The primary acquiring firm, Draslovka SA, is a newly-created entity for purposes of the proposed transaction, and does not currently conduct any activities. Draslovka, the controlling shareholder of Draslovka SA, is a company incorporated in the Czech Republic. Draslovka and its various subsidiaries (“the Draslovka Group”) are ultimately jointly controlled by NP Finance s.r.o., a joint venture arrangement between two Liechtenstein Anstalts, namely Nelson Anstalt and Petronia Anstalt.

 

[11]         The Draslovka Group is primarily involved in the manufacture and distribution of cyanide-based compounds internationally. Draslovka produces hydrogen cyanide (“HCN”) and HCN-based derivatives, including liquid and solid NaCN.[1] Draslovka sells solid NaCN globally, with its biggest exposure in the Americas, Europe and Turkey. Draslovka sells liquid NaCN in Europe from its manufacturing facilities in the Czech Republic. However, Draslovka currently has minimal sales of solid NaCN in Africa, including South Africa. Draslovka does not sell any liquid NaCN in South Africa.

 

[12]         The Target Business is a business unit within the chemicals division of Sasol, which is in turn, controlled by Sasol Limited. Sasol Limited is not controlled by any firm. The Target Business is the only manufacturer and supplier of liquid NaCN in South Africa, which it sells exclusively to customers in the gold mining industry. NaCN solution is generally used to leach gold from ore.

 

RATIONALE FOR THE TRANSACTION

 

Draslovka

 

[13]         Draslovka submitted in its merger filing that the Target Business presents Draslovka with a platform to accelerate its international diversification objectives, and to improve its strategic position in the cyanide-based chemicals market globally. Draslovka submitted further that the Target Business will enable it to bring its production capacities closer to new and potential customers in South Africa.

 

[14]         As regards its intentions for the Target Business, Draslovka stated that it intended to introduce technological improvements to the Target Business and thereby ensure its long-term sustainability. In addition, Draslovka indicated that it intended, subject to approval processes and the carrying out of future financial assessments, to expand the production capacity of Sasol’s existing NaCN plant.[2]

 

Sasol

 

[15]         Sasol, for its part, submitted that the Target Business forms only a small part of Sasol´s chemical business portfolio in South Africa, and that the proposed transaction is part of Sasol´s asset disposal programme executed in line with its balance sheet, shareholder value and strategic objectives, with the aim of divesting businesses not aligned to Sasol’s future strategic growth objectives. Mr Mokomela elaborated in his evidence that cyanide technology is not a speciality of Sasol; that Sasol has not invested in significant research and development in this regard; and that it currently (at the time of the merger hearing) has no plans or intentions to invest in the expansion of the Target Business to meet the growing demand for NaCN in South Africa.[3]

 

THE RELEVANT MARKET

 

[16]         NaCN is a chemical compound commonly used in the extraction of precious metals like gold and silver. NaCN can be produced in solid or liquid form.

 

[17]         In South Africa, NaCN is used almost entirely for the leaching of gold by gold mining companies. It was common cause amongst all the parties that there are currently no viable alternatives to NaCN for gold mining companies in South Africa.[4]

 

[18]         In its solid form, NaCN is a white, water-soluble powder or crystalline briquette produced through the evaporation of water from liquid NaCN in special solidification plants. As a solid, NaCN is easily preserved, and easy to transport and store.

 

[19]         Liquid NaCN, on the other hand, is a solution of concentrated solid NaCN and water. Liquid NaCN is not easily transported and requires purpose-built isotanks that are suitable for road or rail transport. The active ingredient in liquid NaCN also tends to decay over time through hydrolysis. As a result, liquid NaCN is typically used by customers located close to an NaCN production facility and relatively quickly.

 

[20]         The Target Business is a manufacturer of liquid NaCN at two adjacent plants in Sasolburg. For the sake of convenience, they are referred to below as Sasol’s “NaCN plant”.

 

[21]         The Commission considered whether solid and liquid NaCN are substitutable from a demand perspective (i.e., from the perspective of gold mining customers in South Africa), and found that they are not. The Commission determined that gold mining customers in South Africa generally only use liquid NaCN in their gold-mining operations and have limited, if any, infrastructure to dissolve solid NaCN for use in their operations. Furthermore, the construction of a mixing plant to convert solid briquettes into liquid form would require significant capital investment. Of the gold mining customers, only PAR currently has a small dissolving facility. In addition, because no solid NaCN is produced in South Africa, it would have to be imported, and the landed cost of briquetted cyanide is considerably higher than the domestic price of liquid NaCN purchased from Sasol. Solid NaCN is generally more expensive than liquid NaCN because its production involves an additional processing step to evaporate water, crystallise the solution and briquette the material.

 

[22]         There is also doubt regarding the ability of South African ports to handle imports of solid NaCN with the level of safety now required by international standards. Mr Mokomela testified that, while Sasol historically imported (and dissolved) limited quantities of solid NaCN to supplement its liquid NaCN inventory when necessary (e.g., during plant shutdowns), the rules of the International Cyanide Management Institute (“ICMI”) for the safe importation of NaCN have changed, and the Durban port is currently unable to guarantee adherence to these international safety standards. Sasol is therefore no longer able to import solid NaCN and has accordingly mothballed its dissolving facility.[5]

 

[23]         Based on its investigation, the Commission concluded that the relevant market in which the Target Business operates is the market for the supply of liquid NaCN.

 

[24]         As regards the geographic ambit of the relevant market, the Commission found that, because of the hazardous and unstable nature of liquid NaCN and the associated logistics costs, it is generally only transported by road or rail, within specific circumstances and across relatively short distances. The Commission determined that most of the purchasers of liquid NaCN from the Target Business are located within a 700km radius of its Sasolburg plant. The Commission accordingly assessed the supply of liquid NaCN at a national level.

 

[25]         The merger parties and the intervenors all agree that the relevant market is the market for the supply of liquid NaCN in South Africa, and that the Target Business occupies a monopoly position in this market.

 

[26]         For the above reasons, we consider the relevant market to be the supply of liquid NaCN n South Africa

 

THE COMMISSION’S FINDINGS ON THE COMPETITION AND PUBLIC INTEREST EFFECTS OF THE MERGER

 

[27]         In its merger investigation, the Commission found that, while both Draslovka and the Target Business are involved in the manufacture and supply of liquid NaCN, there is no horizontal overlap between the parties because Draslovka does not supply liquid NaCN in South Africa. Draslovka does not have any NaCN facilities in South Africa, and does not supply any liquid NaCN to the South African market from its production facilities in the Czech Republic.

 

[28]         Draslovka also does not sell any significant quantities of solid NaCN in South Africa. The Target Business, for its part, does not manufacture solid NaCN,[6] and historically only imported small quantities of solid NaCN when it was necessary to supplement its local supply of liquid NaCN, for example during plant shutdowns.

 

[29]         Having regard to this market structure, the Commission found that the proposed transaction does not give rise to any material input foreclosure or customer foreclosure concerns.

 

[30]         However, during its investigation, the Commission received concerns from various gold mining companies (including the three intervenors referred to above), and from the Department of Trade, Industry and Competition (the “DTIC”), that the merged entity is likely (i) to increase the price it charges domestic customers for liquid NaCN, potentially up to import parity levels (reflected, in this case, by the costs associated with having to import and dissolve solid NaCN for use in local gold mining companies’ operations), and/or (ii) to reduce the supply of liquid NaCN to South African customers in favour of exports. The concerned customers stated that either of these effects would have a materially negative impact on their operations because they are dependent on the supply of liquid NaCN for their gold mining operations in South Africa, and liquid NaCN constitutes a material portion of their operational costs.

 

[31]         As regards the local supply concern, the Commission found that exports of liquid NaCN would not be feasible for the merged entity. The Commission also found that any manufacture and export of solid NaCN by the merged entity was unlikely to be at the expense of the supply of liquid NaCN to local customers.

 

[32]         As regards the pricing concern, the Commission found that the Target Business has historically priced NaCN significantly below import parity levels, partly because it has the benefit of a cheap supply of key inputs (caustic soda in particular) as part of the vertically-integrated Sasol Group. However, post-merger, the Target Business would no longer receive this benefit because it would have to purchase these key inputs from Sasol at market-related prices that are currently considerably higher than those incurred by the Target Business for these inputs as a vertically-integrated firm. The arms-length supply agreements between the Sasol Group and Draslovka SA for these inputs would therefore result in increased input costs for the Target Business post-merger which would necessitate price increases by the Target Business for NaCN. The Commission found that this was a merger-specific price increase and that it constituted a substantial prevention or lessening of competition within the meaning of section 12A(1) of the Act.

 

[33]         As regards the public interest, the Commission found that the proposed transaction would not have any negative effect on employment, because the merger parties had indicated that the merger would not result in any retrenchments. In addition, no employees of the merger parties expressed any concerns regarding the proposed transaction. However, the Commission found that the pricing effects of the proposed merger would (for the reasons outlined above) have a substantial negative impact on the South African gold mining industry, and that this amounted to a substantial negative public interest effect in terms of section 12A(3)(a) of the Act.

 

[34]         The Commission engaged with the merger parties on a remedy to address this pricing concern, but concluded that the conditions offered by the merger parties did not adequately do so.

 

[35]         Based on the above findings, and the absence of remedies that it regarded as adequate to address the likely price effects of the proposed merger, the Commission prohibited the merger.

 

THE INTERVENORS’ CONCERNS

 

[36]         In their intervention applications before the Tribunal, SSW, Harmony and PAR elaborated on the price increase and local supply concerns they had expressed during the Commission’s merger investigation.

 

[37]         As indicated above, SSW persisted in its opposition to the proposed merger at the Tribunal hearing, and its witnesses testified in support of the concerns it had raised. However, Harmony and PAR accepted revised term sheets proposed by Draslovka in the period preceding the Tribunal hearing, and therefore did not persist in their opposition to the merger. The Tribunal nevertheless requested Harmony and PAR to attend at the Tribunal hearing to explain the reasons for their revised stance on the merger. The evidence provided by the intervenors’ witnesses is discussed in more detail below.

 

CENTRAL ISSUES

 

[38]         Having regard to the merger assessment required by section 12A of the Act,[7] the centra issues raised by the proposed merger are the following:

 

38.1   Whether the proposed merger is likely to give rise to an increase in the price of liquid NaCN sold in South Africa;

 

38.2   Whether the proposed merger is likely to give rise to a reduced supply of liquid NaCN for South African consumers;

 

38.3   Whether, on the basis of either or both of these factors, the merger is likely to result in a substantial prevention or lessening of competition in the relevant market, within the meaning of section 12A(1) of the Act;

 

38.4   If so, whether any such anti-competitive effects are outweighed by pro-competitive benefits of the merger;

 

38.5   Whether the proposed merger can or cannot be justified on substantial public interest grounds; and

 

38.6   Whether any negative competition or public interest effects of the proposed merger are adequately addressed by the conditions proposed by the merger parties.

 

[39]         We consider all these issues below.

 

THE PRICING CONCERN

 

Introduction

 

[40]         There is no dispute that the basis upon which the merged entity would charge gold mining customers for liquid NaCN post-merger would be different than the basis upon which Sasol has historically priced liquid NaCN. It is further common cause that the proposed merger would result in the “vertical de-integration” of the Target Business from the Sasol Group. However, there is a significant dispute between the parties as to whether this price increase was attributable to the merger and should therefore be included in or excluded from the relevant counterfactual. This is because, after the conclusion of the merger agreement in mid-2021 (and also after the Commission’s prohibition of the proposed merger in November 2021 and the filing of the merger parties’ request for consideration in December 2021), the pricing methodology used by Sasol for pricing liquid NaCN in South Africa changed significantly over the period 2022 to January 2023.

 

[41]         Because the changes in Sasol’s pricing methodology took place after the merger agreement and after the Commission’s prohibition of the merger and the filing of the merger parties’ request for consideration, the Commission contended that these changes in Sasol’s pricing should be regarded as attributable to the merger, and should accordingly be excluded from the relevant counterfactual for purposes of the assessment of pricing effects.

 

[42]         The merger parties, on the other hand, argue that the changes in the pricing approach of the Target Business in 2022 and 2023 had nothing to do with the proposed merger, and everything to do with the unprecedented increase in the international prices of certain input costs in the production of NaCN as a result of Russia’s invasion of Ukraine in early 2022.

 

[43]         Sasol’s position at the Tribunal hearing was furthermore that, irrespective of the extent to which the proposed merger might have contributed to the recent changes in its approach to the pricing of NaCN, its current methodology is “here to stay”, with the result that, even if the merger did not proceed, Sasol would continue to apply this new methodology. Sasol therefore contends that, as a matter of commercial reality, the prices charged by the merged entity for liquid NaCN post-merger would not be materially different from those that Sasol would charge if the merger did not proceed. Sasol accordingly argues that this is the “real world” counterfactual that should be applied in considering the competition and public interest effects of the merger. The Commission disagrees with this and contends that, if the merger does not proceed, it is more likely that Sasol will revert to the pricing approach it followed prior to the proposed merger. The Commission further argues that one of the reasons Sasol would do so is that it would otherwise face a real risk that its pricing of NaCN would be found to be excessive under the Act, especially given its monopoly position in the South African liquid NaCN market, and the margins the NaCN business would be making based on its low cost base as a vertically-integrated entity within the Sasol Group.

 

[44]         The Commission argues further that, in addition to the price increase brought about by Sasol’s alignment of its pricing mechanism with the pricing mechanism that Draslovka has indicated it would apply post-merger, Draslovka would have a merger-specific ability and incentive to increase the NaCN price even higher – up to, or perhaps even above, import parity levels. As discussed further below, Sasol has historically priced liquid NaCN to customers in South Africa at a significant discount to IPP, and even its current prices are below IPP. Draslovka denies that it would be commercially rational for it to price liquid NaCN up to IPP levels in South Africa.

 

Changes in Sasol’s pricing methodology

 

Sasol’s historical NaCN pricing mechanism

 

[45]         Mr Mokomela explained that Sasol’s historical pricing mechanism (prior to 2022) generally utilised an agreed “base price” that was negotiated annually with each of Sasol’s major gold mining customers. The negotiation of this base price involved the consideration of various data points, including international NaCN prices, year-on-year changes in the producer price index and consumer price index, and annual changes in electricity prices. However, Mr Mokomela stressed that these data points were used by Sasol simply as a “reference point” for negotiations with customers, and that the base price for each customer was not adjusted directly in accordance with international prices.[8]

 

[46]         The base price agreed with each customer was then adjusted every quarter with reference to formulas that tracked changes in the prices for inputs on global indices. For instance, the adjustment formulas had regard to changes in the international raw material prices for caustic soda and ammonia (excluding sea freight and landing costs); changes in natural gas prices (with reference to the NERSA-regulated price); and changes in local transport indices. Logistics costs were charged separately based on actual delivery costs.[9]

 

[47]         A general formula reflecting the weighting of all these factors is set out in an internal Sasol memorandum dated 22 November 2022 (the “November 2022 memorandum”).[10] This formula indicates that changes in the specified international caustic soda price index had by far the heaviest weighting […] %) in the quarterly adjustment of the NaCN price.

 

[48]         It is important to note that the global indices applied in the quarterly price adjustments did not reflect the actual prices paid, or costs incurred, by the Target Business for the various inputs. Because the Target Business forms part of the vertically-integrated Sasol Group, the key inputs (caustic soda, ammonia, electricity and natural gas) are “self-supplied” to the Target Business – in certain cases, at prices or costs materially lower than the international price of such inputs, and materially lower than the prices at which Sasol sells these inputs to third parties.

 

[49]         Of particular significance in this regard is caustic soda, which (as indicated above) has the heaviest raw material weighting in the quarterly adjustment pricing formula. Sasol is a significant domestic producer of caustic soda, and it has historically supplied caustic soda to the Target Business, as a vertically-integrated entity, at an internal transfer cost. This internal cost is much lower than the international price of caustic soda, and also much lower than the price at which Sasol sells caustic soda to third parties in South Africa.[11] However, post-merger, Sasol will charge the Target Business a full import parity price for caustic soda, as discussed further below.

 

[50]         Sasol explained in this regard that caustic soda is in short supply in South Africa, and that it is therefore able to (and does) sell caustic soda to third party customers at a price based on import parity.[12] Mr Harman calculated that the differential between the internal cost incurred by the Target Business for caustic soda and the price charged by Sasol to third parties for caustic soda (on a Rand/ton basis) was approximately […]% over the period July 2021 – June 2022, and approximately […]% over the period July 2022 – February 2023.[13] Mr Harman’s calculations are consistent with Mr Mokomela’s rough estimate of a differential of […]%.[14]

 

[51]         Sasol Chemical Industries Ltd (“SCI”), another entity in the Sasol Group, is the only producer of ammonia in South Africa. According to Sasol, ammonia is also in short supply in South Africa and can therefore also be sold by it to third parties at import parity prices which are considerably higher than its actual cost of production. However, arising from a settlement agreement concluded by SCI with the competition authorities in 2010, we understand that SCI currently charges the Target Business the same price for ammonia as it charges third party customers in South Africa (although the undertakings in terms of the settlement agreement have lapsed).[15]

 

[52]         Sasol Gas (Pty) Ltd (“Sasol Gas”), another entity in the Sasol Group, is the only supplier of natural gas in South Africa, which it supplies both internally to businesses within the Sasol Group (including the Target Business) and also to third party customers. We understand that Sasol Gas charges the Target Business for natural gas on the same basis that it charges third parties, in accordance with the maximum pricing regulations issued by the National Energy Regulator of South Africa (“NERSA”).[16]

 

[53]         We note that an excessive pricing complaint against Sasol Gas has been referred to the Tribunal based, inter alia, on Sasol Gas’s price/cost margin in respect of natural gas.[17] However, the question whether the prices charged by Sasol Gas for natural gas are excessive or not is not relevant to our analysis of the proposed merger. Rather, what is relevant in the present proceedings is whether there would be any change in the price that the Target Business charges customers for NaCN as a result of the proposed merger. One of the relevant factors in this regard is whether the Target Business is likely to pay higher prices for the inputs referred to above as a stand-alone firm post- merger than it does as a vertically-integrated firm within the Sasol Group.

 

[54]         In terms of Sasol’s historical pricing methodology, the quarterly adjustments generated by the above pricing mechanism were then subjected to a so-called “cap-and-collar” mechanism, the purpose and effect of which was to restrict the quarterly movement or variations in the NaCN price charged to any customer to approximately 5% (i.e., even where the prices reflected in the adjustment formula fluctuated more significantly). As a consequence, the NaCN price changes did not necessarily fully track changes in the underlying prices captured in the adjustment formula, at least in the short term.

 

[55]         It does not appear that the above pricing methodology, and its application each quarter, was typically shared with the customers themselves. The quarterly pricing letters included in the record simply refer generally to changes in the prices of the various input prices and specify a net percentage NaCN price change.

 

Changes to Sasol’s NaCN pricing mechanism

 

[56]         The evidence indicates that, during the course of 2022 to January 2023, Sasol changed its pricing mechanism – first to a transitional pricing mechanism in 2022 and then, with effect from January 2023, to a stoichiometry-based pricing mechanism (i.e. a mechanism directly based on the prices of the key variable inputs into the production of NaCN weighted according to the volumes of each input consumed in the production of one ton of NaCN).

 

[57]         Critically, however, as discussed further below, the input price used for caustic soda in the new formula reflects a full import parity price for caustic soda and not the (much lower) cost currently incurred by NaCN business for caustic soda as a vertically- integrated entity within the Sasol Group.

 

[58]         Sasol argues that these changes were a response to unprecedented increases in the prices of the key inputs used to make liquid NaCN as a result of Russia’s invasion of Ukraine in early 2022. Mr Bruzek explained in this regard that, from early 2022, the international prices of these inputs increased dramatically, and were extremely volatile, as a result of the Russian invasion of Ukraine, the consequent energy crisis in Europe, and congestion in international freight. As a result, most NaCN producers internationally increased NaCN prices by 30-50% in 2022.[18]

 

[59]         However, the Commission disputes this. It contends that, because Sasol currently self- supplies caustic soda to its NaCN business at an internal input cost that is based on its cost of production, its actual costs are unaffected by fluctuations in the international price of caustic soda. Therefore, the Commission argues, Russia’s invasion of Ukraine in early 2022 did not have any impact on the Target Business’ actual input costs. Hence, absent the merger, the NaCN business would not have had any incentive to change its NaCN pricing methodology to recover the international cost of caustic soda in its NaCN price.

 

[60]         The Commission’s position is therefore that, absent the merger, Sasol’s NaCN prices in 2022 and early 2023 would have risen only to the limited extent that international prices affected pricing under Sasol’s historical pricing mechanism, and that Sasol would not have changed the NaCN pricing mechanism itself to make it directly responsive to the international prices of caustic soda. The Commission argues further that the changes made by Sasol to its NaCN pricing mechanism in 2022 and January 2023 resulted in significantly higher prices over that period than would have been charged under Sasol’s historical pricing mechanism, and therefore do not represent the relevant counterfactual for purposes of assessing the price effects of the proposed merger. SSW joined cause with the Commission in this contention.

 

[61]         We proceed to consider the relevant evidence on this issue below.

 

[62]         In his witness statement, Mr Mokomela stated that:

 

In the past 12 – 18 months, the margins achieved by the NaCN business have not reflected a sustainable position. The margins have been declining as a result of the increasing feedstock prices and energy costs. For example, the caustic soda, ammonia, and natural gas (index) prices between 2021: Q4 and 2022 Q1 increased by 23%, 38% and 12% respectively. [19]

 

[63]         Mr Mokomela testified that this led Sasol to engage in a review of its pricing mechanism because “it was slow to respond to the input prices changes and/or did not capture the magnitude of the impact of the input costs/prices [20] Mr Mokomela stated further that, as a result of this review, a new, monthly, stoichiometry-based pricing mechanism, without any “cap and collar” mechanism, was adopted for NaCN pricing going forward.[21] In terms of this new model, the price charged for NaCN is adjusted on a monthly basis in accordance with changes in the international prices of the relevant inputs used in the manufacture of NaCN. Mr Mokomela explained that the purpose of this new mechanism was “to ensure responsiveness to market changes”.[22]

 

[64]         Further insight into the rationale for this change can be gleaned from the November 2022 memorandum.[23] Given its significance, we set out its contents in some detail. The November 2022 memorandum appears to have been prepared initially by the Sasol Base Chemicals division on 22 November 2022, and subsequently to have been approved by Mr Mokomela and others in early 2023. The memorandum bears the heading “Pre-Negotiation Document (PND) for Negotiations with Mining Houses on the Sodium Cyanide Contract And Price Range Pricing”, and states the following under the heading of “Summary

 

“Historical financial performance of the Sodium Cyanide producing asset has been good with average gross margin per ton approximately R[…] and EBITDA of approx. R[…] million per year. In the past 12–18 months, unit margins have however been reducing on the back of increasing feedstock and energy costs despite efforts from marketing and sales to address the trend. This has prompted a review of the current pricing mechanism which was slow to respond to these changes and/or did not capture the magnitude of the impacts. Several pricing mechanisms were reviewed and formulated to derive a mechanism that returns the financial performance of the asset to economically viable historical levels. Of the options reviewed a new stoichiometry-based pricing mechanism was selected as the most suitable. The new model safeguards the generation of the required gross margin per dry metric ton (dmt). To ensure a more responsive approach, the pricing mechanism would also change from quarterly to a monthly stochiometric based model where the cap and collar mechanism are removed to ensure full exposure to market changes.”[24]

(our emphasis)

 

[65]         The “key expectations of the new mechanism are then identified as:

 

·  To return the unit gross margin of the Sodium Cyanide Asset to historical levels and ensure its economic viability.

 

·  To formulate a new and transparent cost-price formula which reflects Sasol’s input costs, while considering the reference of global sodium cyanide trends.

 

·  To arrive at a pricing formula that is simple, responsive, and robust even during volatile periods.”[25]

 

(our emphasis)

 

[66]         Mr Mokomela confirmed in his evidence that the purpose of the new pricing mechanism was to “restore the business back to historical performance”.[26]

 

[67]         As regards Sasol’s historical NaCN pricing mechanism, the memorandum states:

 

Whilst this mechanism has served Sasol well in the past, recent, volatile swings in the input pricing saw the mechanism responding poorly to macro changes owing to the time lag and the cap and collar mechanism. This eventually resulted in economically unsustainable unit gross margins.”[27]

 

[68]         The memorandum goes on to state that interim price increases were effected by Sasol in 2022 pending the implementation of the new pricing mechanism as follows:

 

“CY2022: Q1- Increased prices by […]% as per mechanism cap and collar.

CY2022: Q2 – Increased Prices by […]%

CY2022: Q3 - Increased Prices by […]%

CY2022: Q4 - Increased prices by […]%”[28]

 

[69]         These price increases were confirmed by Mr Mokomela in his evidence.[29] The merger parties suggest that the quarters referred to above refer to Sasol’s 2022 financial year (which commenced in July 2021) and not calendar year 2022 as indicated in the November 2022 memorandum, with the result that the […]% increase (outside the cap and collar mechanism) took place in October 2021. However, whilst it appears that SSW[30] and DRD[31] began experiencing significant NaCN price increases from that date, PAR, Harmony, Gold Fields all indicated that their companies only began experiencing very significant increases in Sasol’s pricing in the first half of 2022,[32] and the pricing analysis conducted by Mr Harman suggests that Sasol began pricing above the 5% cap from approximately the second quarter of 2022.[33] We therefore refer to this change in Sasol’s pricing mechanism as a 2022 change in line with the November 2022 memorandum. However, the precise date on which Sasol first began exceeding the 5% cap does not affect our assessment below. What is more relevant is the date on which Sasol introduced its stoichiometry-based pricing mechanism which, it is common cause, was January 2023.

 

[70]         As to why the stoichiometry-based pricing mechanism was selected by Sasol, the memorandum states the following under the heading “Alternative Solutions :

 

Several options were considered in evaluating Sasol’s approach to cyanide pricing. These include the import parity pricing mechanism (IPP mechanism), refining the current model (move quarterly to monthly pricing, removal of the cap and collar mechanism) and/or introducing new stoichiometry-based pricing mechanism. While the IPP model was found to be indicative of global trends, finding an appropriate, publicly available benchmark was challenging and would make the likelihood of acceptance of IPP by our customers low, it was decided the IPP price model will only be used as a reference point. The change of the current model from quarterly to monthly pricing was easy to implement as it contained items (including but not restricted to differences in the standard weightings of components in the bill of materials) which needed to be remedied. To rectify these items, a new mechanism based on the stoichiometric bill of materials was formulated. The new model will ensure the generation of the required gross margin (>R[…] per drey metric ton (dmt)) on a sustainable basis. To ensure a more responsive approach, the pricing mechanism will change from quarterly to a monthly stochiometric based model where the cap and collar mechanism will be removed for full exposure to market changes."[34]

 

[71]         As regards the details of the stoichiometry-based pricing mechanism, the memorandum explains that Sasol conducted an analysis of the volumes of inputs consumed in producing one ton of NaCN over a period of 5 years (FY17- FY22), and then formulated an equation for the selling price that would produce "the required returns".[35]

 

[72]    The memorandum elaborates that the variable portion of the formula is made up of pricing for caustic soda, ammonia, and natural gas pricing where "the caustic [soda] and ammonia prices are brought back to a Sasolburg import parity equivalent” and natural gas is based on the NERSA-approved price.[36]In particular, the caustic soda component of the formula is defined as the […][37]

 

[73]    The purpose of the fixed component of the formula comprises, inter alia, an "asset service fee" which is the […]. The purpose of the fixed component is to “ensure a low risk to the margin which can be sustained an ongoing basis”.[38]

 

[74]    The memorandum notes that “[t]he Enterprise Value (R[…]b) used is consistent with the Enterprise Value calculated for the cyanide business during Project Silver”. “Project Silver” was the code name given to the proposed sale of the NaCN business.

 

[75]    The memorandum then identifies three permutations of the new mechanism for purposes of pricing negotiations with customers, namely:

 

75.1.   a “going-in” position, […];

 

75.2.   a “fall-back” position, […] and

 

75.3.   a “walk-away” position, […][39]

 

[76]    Figure 1 of the memorandum then shows a historical simulation based on these options.[40] Significantly, Figure 1 shows that the “historical” gross margin being tracked is a “non-integrated” gross margin based on “simulated non-integrated [variable costs]”. This also indicates that the “required” average gross margin of R[…] referred to in the November 2022 memorandum (at which the new pricing mechanism is targeted) reflects a gross margin for the Target Business on a stand-alone basis (with non-integrated variable costs”), By the same token, Figure 1 suggests that this “required” average gross margin does not reflect the actual average gross margin historically generated by the Target Business calculated on the basis of the actual costs it incurred for inputs as a vertically-integrated business within the Sasol group. Rather, it is calculated on the basis of the “non-integrated variable costs  (i.e. “market” prices) that the Target Business would pay for these inputs as a third party customer. In the case of caustic soda and ammonia, the simulated prices are, as indicated above, “Sasolburg import parity equivalent” prices.

 

[77]         Ms Wainer confirmed in her evidence that the “historical” average gross margin and EBITDA figures referred to in the November 2022 memorandum approximated those of the Target Business on a pro forma un-integrated basis, i.e., a stand-alone business paying the Sasol Group “market” prices for all its key inputs.[41]

 

[78]         The November 2022 memorandum concludes by noting that Sasol was successful in secured the “going-in” position with all of its gold-mining customers with effect from 7 January 2023.[42] It appears from the appendix to the November 2022 memorandum that the new stoichiometry-based pricing mechanism was presented to the gold mining customers for the first time in the period November/December 2022 and thereafter implemented on short notice in January 2023.[43]

 

[79]         What the November 2022 memorandum indicates, therefore, is the following:

 

79.1       The purpose of the new pricing mechanism was to “restore” the financial performance of the Target Business to its “historical” average gross margin per ton of approximately R[…], which was described as the “required return”.

 

79.2       However, this “historical” performance does not reflect the actual performance of the Target Business based on the costs it actually incurred over the relevant period; rather, it is a “simulated” margin that reflects the margin the Target Business would have earned as an un-integrated entity paying the Sasol Group third party “market” prices for all its key inputs, including import parity prices for caustic soda and ammonia.

 

79.3       The Target Business’s financial performance on this simulated basis was negatively affected by the historical pricing mechanism, principally because that mechanism did not capture the magnitude of the increases in the international prices of caustic soda and ammonia in 2022.

 

79.4       In order to restore the simulated “historical” performance of the Target Business, the new stoichiometry-based pricing mechanism differs from the historical NaCN pricing mechanism in three key respects – (i) adjustments now take place monthly rather than only quarterly; (ii) the “cap and collar” mechanism no longer applies; and (iii) the NaCN price is now tied directly to the market price of inputs (including, in the case of caustic soda and ammonia, the import parity prices of those products) plus a fixed margin. As explained above, under the historical pricing mechanism, international raw material prices were relevant only to the extent that changes in the relevant international price indexes informed the quarterly adjustments to the base price negotiated by Sasol with each of its customers annually.

 

79.5       During the course of 2022, Sasol effected quarterly increases in the price of NaCN by amounts exceeding those previously permitted by the “cap and collar” mechanism pending the introduction of the new, stoichiometry-based pricing mechanism

 

79.6       The latter mechanism was finalised and approved in November 2022, and implemented with effect from January 2023, in a form that will enable Sasol to generate a higher return than the simulated, non-integrated target margin referred to above.

 

Differences between the prices generated by Sasol’s historical pricing mechanism and the prices generated by its current pricing mechanism

 

[80]         As appears from the November 2022 memorandum, Sasol increased its NaCN prices by significantly more than the price increases permitted under the historical “cap and collar” pricing mechanism over the second to fourth quarters of 2022 ([…]%, […]% and […]%, respectively).

 

[81]         This is consistent with the other evidence at the hearing.

 

81.1.   All the gold mining company witnesses testified about significant NaCN prices rises, well over what was permissible under the historical 5% “cap and collar” pricing mechanism, over the period 2022 to 2023. For example, Mr Hayward testified that Gold Fields experienced a cumulative increase of approximately […]% in the price of NaCN from Sasol over the course CY 2022, and of over […]% in CY 2023 (to date).[44]

 

81.2.   The Commission’s expert, Mr Njisane, calculated that Sasol implemented an average Nacn price increase of […]% over FY 2022 (ending in June 2022), and a further […]% increase over the period July 2022 to February 2023 (which included the introduction of the stoichiometric pricing mechanism in January 2023).[45]

 

81.3. Mr Njisane calculated further that, had the stoichiometric pricing mechanism been in place over the period July 2022 to January 2023, it would have generated prices that cumulatively were, on average, […]% higher than the prices in fact charged by Sasol over that period (which were themselves higher than those permissible under the historical pricing mechanism, as indicated above).[46] For the month of July 2022 alone, Mr Njisane calculated a differential of […]% for SSW, […]% for Gold Fields, and […]% for Harmony and DRD.[47] Mr Harman calculated a similar differential of […]-[…]% for DRD for July 2022.[48]

 

[82]    Mr Harman also compared the relationship between Sasol’s NaCN selling prices and the unit variable costs of the NaCN business over the period July 2021 to March 2023, and found that, whilst that relationship was relatively stable until about March 2022, it then began to break down, and disappeared completely from January 2023 when the new stoichiometric pricing mechanism was introduced. Mr Harman concluded that changes in the unit variable costs in fact incurred by the NaCN business over the period 2022 to January 2023 did not account for the price increases experienced by DRD over that period.[49]

 

[83]         All of this evidence is consistent with the stated objective of the price mechanism changes implemented by Sasol over the period 2022 to January 2023, namely to achieve prices that would secure the profit margin that was being targeted for the NaCN business on an un-integrated business. As explained above, these price increases were a function of a new pricing approach that was not only more responsive to changes in the costs of the underlying inputs into the NaCN business, but which also directly reflected such input costs on an arms-length unintegrated basis.

 

[84]         In the case of caustic soda in particular, this represented a significant change to the historical pricing mechanism, which had not costed caustic soda on a “market price” basis. As noted above, Mr Harman compared the annual average price for caustic soda reflected in the variable cost reports of the NaCN business over the period July 2022 to February 2023 with the price at which Sasol sold caustic soda to external customers over that period, and found that the latter was […]% higher than the former.[50]

 

[85]         Mr Smith did not materially dispute the extent of these price rises. Mr Smith’s position was rather that (i) it had always been economically rational for Sasol to set the price of NaCN with reference to the opportunity cost (and not the actual cost) of caustic soda – which, in the “short” markets that exist for these products in South Africa, approximates an import parity price – and there was accordingly nothing merger-specific about the price rises in 2022 and January 2023. We consider the question of merger-specificity and the relevant counterfactual further below.

 

Sasol’s future pricing of NaCN absent the merger

 

[86]         Mr Mokomela stated that, if the proposed merger does not proceed, it will maintain the new stoichiometric pricing mechanism it adopted in January 2023 because of the benefits referred to above – in particular, that it will better reflect changes in the international prices of feedstock inputs, and thus ensure that the NaCN business is able to operate on a “commercially viable” basis.[51]

 

[87]         The merger parties therefore contend that the appropriate counterfactual for purposes of assessing whether the proposed merger would have any pricing effects is the prices determined by the new, stoichiometry-based pricing mechanism and not the prices that would have been determined by Sasol’s historical pricing mechanism. We return to this contention further below.

 

The proposed pricing of NaCN by Draslovka

 

[88]         In his witness statement, Mr Bruzek stated that Draslovka prices NaCN on a “cost plus” basis, which allows Draslovka to recover the actual cost of key inputs and an appropriate margin.[52]

 

[89]         Mr Bruzek elaborated that Draslovka applies an ex-works “cost plus” pricing mechanism comprised of three components, namely:

 

89.1   the unit price of each of the four key inputs (caustic soda, ammonia, natural gas and electricity) multiplied by the relevant consumption factor to produce one ton of NaCN;

 

89.2    other processing and cash fixed costs; and

 

89.3           an EBITDA margin of […]-[…]%, which allows producers of NaCN to invest and reinvest in keeping their assets well-maintained, reliable and optimised from an environmental, health and safety point of view.[53]

 

[90]         Mr Bruzek explained further that, as regards the Target Business, all of the four key inputs (caustic soda, ammonia, natural gas and electricity) will continue to be supplied by the Sasol Group to the business post-merger. In terms of supply agreements agreed between the Sasol Group and Draslovka, the formulae for caustic soda and ammonia are based on international ammonia and caustic soda indexes, whilst the formulae for natural gas and electricity are based on the regulated prices of these products.[54] These costs will then be passed on to consumers (together with other costs and a margin) in terms of the “cost plus” pricing mechanism referred to above.[55] Insofar as the Target Business is concerned, Mr Bruzek stated that Draslovka is willing to commit to charge a base margin of […]% on its costs.[56]

 

[91]         The precise formulae for the determination of the caustic soda and ammonia prices payable by the NaCN business post-merger are reflected in the relevant supply agreements concluded between Sasol and Draslovka.[57] The ammonia price formula reflects an average of various international prices for ammonia, plus all transport, insurance, storage and other related costs to Secunda and Sasolburg.[58] The caustic soda price formula reflects an average of various international prices for caustic soda, plus logistics, port handling and overland transport costs, a working capital fee and a distributor fee.[59]

 

[92]         This is the basis of the Commission’s argument that the proposed merger would result in “double marginalisation” flowing from the “vertical de-integration” of the NaCN business from the broader Sasol Group. Whereas currently Sasol supplies caustic soda to the Target Business at an internal transfer cost rather than a third party “market” price, post-merger Sasol would charge the NaCN business, as a third party customer, a full import parity price for caustic soda, and the NaCN business would then pass on that increased cost (and the other costs of the business), plus a full margin of its own, by way of an increased downstream price.

 

[93]         We highlight at this juncture the similarity between the nature of Draslovka’s “cost plus” pricing mechanism, and the stoichiometry-based pricing mechanism that has been implemented by Sasol with effect from January 2023. In particular, both mechanisms are “cost plus” models in which the costs of all the inputs provided by Sasol to the NaCN business reflect arms-length prices for those inputs – including, in the case of caustic soda and ammonia, prices equivalent to a full import parity price.

 

[94]         It appears to be common cause that, as a result of the similarities in these pricing models, there is no longer a material difference between the NaCN prices charged by Sasol, and the prices that, according to Mr Bruzek, Draslovka would charge post- merger. This is illustrated in slide 54 of Mr Smith’s “hot tub” presentation to the Tribunal.[60]

 

The test for the relevant counterfactual

 

[95]         Relying on the Tribunal’s decision in BB Investment[61] (which was endorsed by the Competition Appeal Court (“CAC”) in Coca-Cola[62]), the merger parties argued that, in order to determine whether or not the changes in Sasol’s NaCN pricing in 2022 and January 2023 should be included in the relevant counterfactual for purposes of assessing the pricing effects of the merger, it is necessary to “consider the incentives of the new controller and whether the effect in question has a nexus with and aligns with the incentives of the new controller [63]

 

[96]         In BB Investment, the Tribunal noted that the above approach must take into account that “firms are dynamic institutions and that “[n]ot every change that results post-merger is necessarily attributable to the merger” . There may, for instance, be changes in a firm’s behaviour post-merger that would have happened in any event and which therefore would not be regarded as merger-specific.[64]

 

[97]         In Coca-Cola the CAC endorsed the BB Investment test as “objective and sound because the focus is on demonstrable outcomes [effects] rather than the subjective attitude or intention of the merging parties[65]

 

[98]         Both of the above cases were concerned with the question whether post-merger retrenchments should be regarded as merger-specific for purposes of the public interest analysis under section 12A(3) of the Act. They were not concerned with the question at issue in this merger, namely whether or not price increases implemented after the agreement, but before the implementation, of the proposed merger should be included in the relevant counterfactual for purposes of assessing the pricing effects of the merger. In the present context, it may be more appropriate simply to apply standard counterfactual analysis, namely to compare the likely market outcomes with, and absent, the merger.[66]

 

[99]         In the present case, however, nothing turns on this question because our conclusion regarding the relevant counterfactual would be the same on either test.

 

[100]      Only events that would have happened in the absence of the proposed merger under review, and are not a consequence of it, can be incorporated into the counterfactual. We consider below the relevant evidence regarding the likely pricing approach of the NaCN business absent the merger. The central question in this regard is whether Sasol would have made the changes it did to its pricing methodology in 2022 and January 2023 absent the proposed merger. As discussed above, those changes resulted in Sasol’s current pricing mechanism being effectively the same (from January 2023) as the pricing mechanism that Draslovka would apply post-merger. Therefore, if it is not likely that Sasol would have made those changes absent the merger, they should not be taken into account in assessing the pricing effects of the merger.

 

Were the changes in Sasol’s NaCN pricing mechanism in 2022 and January 2023 attributable to the merger?

 

Sasol’s pricing of caustic soda and NaCN prior to its engagements with Draslovka

 

[101]      One of the central propositions advanced by Mr Smith on behalf of the merger parties was that, even pre-merger, Sasol faces an opportunity cost to supply the NaCN business with its key inputs (caustic soda, ammonia, natural gas and electricity), and that, in the case of caustic soda and ammonia, that opportunity cost is represented by an import parity price, because both of the caustic soda and ammonia markets are structurally “short” in South Africa. On this basis, Mr Smith testified that the proposed merger would not bring about any change in Sasol’s pricing of caustic soda (or ammonia).[67]

 

[102]      Mr Smith also testified that downstream firms that face a binding capacity constraint (such as that facing the Target Business) can no longer set profit-maximizing prices with reference to the interaction between customers’ price-sensitivity and some measure of marginal costs (or the wholesale prices of inputs). Mr Smith stated that, for this reason too, the merger would not change the pricing incentives of the NaCN business.[68]

 

[103]      In our view, however, these arguments do not take into account the relevant facts in this case, in particular the manner in which caustic soda and NaCN have in fact historically been priced by Sasol as a vertically-integrated entity. In particular, the evidence discussed below indicates that, prior to its engagements with Draslovka in relation to the proposed merger, Sasol did not in fact follow an opportunity cost approach in supplying caustic soda to the Target Business, nor did the Target Business price NaCN having regard to the opportunity cost of caustic soda.

 

[104]      We have set out above the significant difference between the cost at which Sasol has historically supplied (and continues to supply)[69] caustic soda to the Target Business as a vertically-integrated business, and the price at which Sasol sells caustic soda to third parties. Ms Wainer explained that this is an internal transfer “cost” rather than a transfer “price” because caustic soda is produced by the same legal entity that houses the NaCN business, and accordingly there is no “sale” of caustic soda to the NaCN business.[70] However, nothing turns on this distinction for our purposes. What is relevant is that the NaCN business, as a vertically-integrated business within the Sasol Group, did not historically (and does not currently) incur an actual cost for caustic soda that is equivalent to the “opportunity cost”-based import parity price at which Sasol sells caustic soda to third parties. It incurs a much lower internal cost.

 

[105]      Furthermore, there is no evidence that, prior to its engagements with Draslovka in relation to the proposed merger, Sasol sought to recover the “opportunity cost” of caustic soda in its pricing of NaCN, or assessed the financial performance of the Target Business on that basis. Ms Wainer expressed the view that, from her perspective, this is how the NaCN business “should have been operated, but she acknowledged that she was not personally involved in the operations of the NaCN business, and did not have any insight into how it was in fact operated prior to the proposed merger.[71]

 

[106]      As set out above, under Sasol’s historical NaCN pricing model, the annual “base price” of NaCN involved the consideration of international NaCN prices, but simply as one of various “data points”  used by Sasol as a “reference point” for negotiations with customers. Similarly, the quarterly adjustments to that base price had regard, inter alia to changes in the international raw material price for caustic soda (excluding sea freight and landing costs).[72] However, this is obviously very different to recovering the full import parity price of caustic soda in the price of NaCN as that import parity price fluctuates from time to time.

 

[107]      The fact that Sasol did not historically price NaCN based on the opportunity cost of caustic soda is also reflected in the documentary evidence in the record.

 

[108]      The first category of documents is internal Sasol documents dealing with the proposed sale of the Target Business.

 

108.1.    An asset review presentation dated 8 April 2020 (which resulted in the recommendation to sell the NaCN business) recorded that the profitability of the NaCN business “has been […]” but that this was based on “the current cost for caustic soda vs market related prices” that would be payable if Sasol […].[73]

 

108.2.    Then, dealing with the question whether the NaCN business could function as a “stand-alone” business, the presentation forecasted a “decline … for standalone EBIT based on impact of sourcing caustic soda at market price vs integrated cost model used in historical EBIT calcs”.[74] The presentation states that the gross margin of the Target Business could be adjusted “to allow for full cost pass through to end customer, however this remains at risk based on end consumer alternative options and current gap in market”[75] The summary at the end of the presentation states “sodium cyanide standalone economics (EBIT) shows decline, this is mainly linked to imported caustic soda prices […]”.[76]

 

108.3.    The accompanying Asset Review Detail document (dated January 2020) noted that, […] the NaCN business had to import caustic soda, that would impact on the price of caustic soda payable by the NaCN business,[77] with a knock-on effect on the price of NaCN if the margins of the business were to be protected.[78] The document reflects a “normalised” budget income statement to show “the [reduced] standalone value if caustic soda is purchased at a market related price (vs the current cost of production methodology currently used)”.[79]

 

108.4.      A subsequent presentation to the Project Silver Steering Committee (“Steercom”) headed “Project Silver Transaction Feasibility”, and dated 2 June 2020, compared the “internal transfer price” of caustic soda against the (higher) price at which Sasol sold to external parties in South Africa, and against various international caustic soda prices over the period July 2017 to September 2019[80]. The “base case valuation (R[…]bn) and the “upside” valuation (R[…]bn) of the Target Business both assumed that Sasol would, post-transaction, sell caustic soda to the Target Business at “IPP + inland logistics”. The “base case” valuation assumed that the Target Business would, post-transaction, price NaCN at “IPP + logistics – historical average discount of […]%” valuation (R[…]bn) assumed that the NaCN business would price NaCN at “IPP + logistics (no discount)”[81]The presentation concluded with a request to the Steercom to recommend that the sale process commence for the NaCN business on the basis that it “ remains a viable business for divestment even at a caustic soda import parity pricing scenario”.[82]

 

108.5.    A Project Silver Steercom presentation dated 13 October 2020 confirmed that the price at which Sasol would sell caustic soda to the NaCN business post- transaction “will be based on import (incl. logistics) prices, […]” [83] This was repeated in various subsequent Project Silver Steercom presentations.

 

108.6.    A Project Silver Steercom presentation dated 9 March 2021 stated that the Sasol valuation model was “conservative” because “[…] “historic prices are well below import parity”.[84] Mr Mokomela confirmed this in his evidence.[85] However, the presentation went on to note that prospective buyers had factored in “upside potential” that assumed a “[…]% increase in sodium cyanide pricing going forward from FY24”[86]

 

[109]      A second category of relevant documents is the due diligence reports prepared by PricewaterhouseCoopers Advisory Services (Pty) Ltd (“PWC”), Draslovka’s transaction advisors, during the bidding process.

 

109.1.    In an initial “Limited Financial Due Diligence” report dated 17 February 2021,[87] PWC indicated that it considered “ pro forma carved out historical results for the three years ended June 2020, prepared by Sasol’s management.[88] PWC explained that this “carved out” information was “based on pro forma theoretically developed information” because the Target Business was part of the Base Chemicals segment of Sasol and therefore “has not been reported on historically on a fully arm’s length (commercial) standalone basis”.[89]

 

109.2.    As regards the pricing of caustic soda and NaCN, the report stated the following:

 

“The Business may have been relatively protected from significant market price fluctuations in feedstock costs due to the majority of inputs being sourced directly from Sasol at internal transfer prices. Therefore, customer pricing may not have always followed an approach that commercially aims to drive real margins. Prices are typically negotiated with larger customers on an annual basis and adjusted each quarter based on market prices, however market price fluctuations have not been consistently passed on to customers in the past.  [90] (our emphasis)

 

109.3.    One of the adjustments contained in the pro forma statements was, therefore, an adjustments of caustic soda prices (which were stated to represent […]%, on average, of various costs) “to represent the market value”.[91] The report explained that the variable costs of the business “have increased significantly as a result of the carve-out process as the market prices are higher than the internal Sasol transfer prices”. It explained further that “[…]”.[92]

 

109.4.    The report elaborated that the formula used by Sasol to calculate the market value of caustic soda reflected Sasol's approximation of the globally available cost of caustic soda plus an assumed logistics fee. PWC noted that this logistics fee was lower than the fee included in the formula in terms of which Sasol proposed to supply caustic soda to the merged entity post-merger.[93]

 

109.5.    PWC prepared an updated “Limied Financial Due Diligence (Final Phase)” report on 17 May 2021.[94] In that report, it stated the following:

 

“[…]. The standard pricing mechanisms and quarterly adjustment mechanisms have had limited effectiveness due to negotiated deviations, mostly in favour of customers    It will be key to ensure that there is a smooth transition of the sales function and customer knowledge for negotiation purposes. It should be considered whether the price mechanisms can be refined post transaction, to better align with Cyanide’s input costs, in order to protect margins.[95] (our emphasis)

 

109.6.       PWC also stated that:

 

Cyanide’s pricing has historically been low due to favourable internal transfer prices from Sasol’s vertically integrated business, and due to political’ reasons. This is reflected in the lower margins in the pro formas, as compared to those Sasol reported historically.[96] (our emphasis)

 

109.7.    PWC elaborated on this view in two bullet points. The first bullet point was that customers leverage Sasol’s internal transfer pricing in the negotiation process – resulting in lower sales prices [97] PWC expanded on this bullet point as follows:

 

Sasol’s integrated business provides certain feedstock items to Cyanide at a lower than market price. The customers are aware that they are receiving a discount to the international import parity price, due to Sasol being able to afford this based on the cost of feedstock being lower.

 

It is important to consider the position of the standalone Cyanide business in this context – which won’t have access to the feedstock items at transfer prices. Customers are aware of the sale process and that contracts will need to be revisited going forward. Customers may be alert about price increases, which will need to be managed.[98]

 

109.8.    The second bullet point was that “Sasol has historically been willing to accept lower sales prices to customers in the gold sector when it has come under pressure[99] PWC expanded on this bullet point as follows:

 

There is considerable political interest in the gold mining industry, as a result of its importance to the South African economy and employment. Sasol has also received scrutiny and legal challenges on the pricing of other products in its portfolio in the past. The gold industry is volatile – with profits fluctuating based on the USD gold price and the ZAR/USD exchange rate.  Given the above and reliance on cyanide as a key gold mining input, Sasol has been wary of taking a hard line on pricing with the Cyanide customers, and has been willing to reduce margin and absorb costs to support the industry and provide assistance to its top customers, especially in challenging times. . . .

 

However, when the industry recovers, the realised cyanide price comes off a low base and does not seem to increase accordingly.

 

Considering the margin benefits of Sasol’s integrated operations and Cyanide being a relatively small part of the Sasol Group, this was not considered to be an issue for Sasol, but will be for the standalone Cyanide business.[100] (our emphasis)

 

[110]      This understanding of Sasol’s previous pricing philosophy is also reflected in the Final Binding Offer that Draslovka submitted to Sasol on 6 April 2021,[101] where Draslovka stated the following:

 

Draslovka recognises the role played by Sasol in the broader gold mining ecosystem and the pricing support Sasol has provided to the gold mining industry in the past. Whilst we have identified material risk for a new buyer to freely negotiate prices in the long-term, we have not made any downward adjustments to the Base EV [Enterprise Value] to account for this risk. ’[102] (our emphasis)

 

[111]      That Draslovka anticipated the need to raise NaCN prices post-merger to account for the fact that it would be paying higher prices for caustic soda as a third party customer is also reflected in correspondence sent by Draslovka’s attorneys to the Commission on 29 October 2021, in which they stated the following regarding Draslovka’s future pricing strategy:

 

Draslovka’s anticipated input and production cost base for the target business post-merger will differ from Sasol’s existing cost base. Draslovka will be purchasing raw materials at the competitive market price which will be based on internationally accepted price mechanisms that will take into consideration international price indexes and will be expressed in USD. As such, Draslovka’s future pricing strategy must of necessity (and reasonably) be informed by these costs and Draslovka can therefore not be required to apply the Sasol pricing formulas as these would not reflect the commercial reality in which Draslovka will be operating and could lead into a situation where the facility would not be able to sustainably operate the business.’[103] (our emphasis)

 

[112]          Draslovka’s attorneys accordingly informed the Commission that:

 

As the existing supply agreements draw to a close, it is Draslovka’s intention to renegotiate the terms and conditions of its supply on an arms-length basis having regard to its own input and production costs (and achieving suitable returns).’[104]

 

[113]      In our view, all of this evidence indicates that, prior to its decision to sell the NaCN business to Draslovka:

 

113.1.    Sasol supplied caustic soda to the NaCN business at an internal transfer cost because the business formed part of a vertically-integrated business;

 

113.2.    Sasol did not seek to recover the opportunity cost of caustic soda in its pricing of NaCN to customers;

 

113.3.    As a result of the low cost it incurred for caustic soda, the NaCN business did not have to, and did not, maximize the price at which it could sell NaCN. This decision may also have been informed by the profits the Sasol Group was making elsewhere in its overall operations; broader Sasol Group objectives and priorities; a “political” commitment to supporting the South African gold mining industry; and/or legal concerns, based on previous challenges to Sasol’s pricing as a vertically-integrated entity, that, given its low cost base, the NaCN business might be accused of excessive pricing if it sought to profit-maximize.[105]

 

113.4.    We note in the latter regard that, in the Sasol Chemical Industries ase, the CAC held that the feedstock price that should be used for the excessive pricing calculation was the “actual cost” at which Sasol Synfuels sold such feedstock to Sasol Chemical Industries.’[106]

 

[114]      However, whatever these reasons may have been for what Sasol refers to in its internal documents as the NaCN business’s “defensive pricing, the above documents suggest that, prior to Sasol’s engagements with Draslovka, they afforded gold mining customers a limited degree of countervailing power in their pricing negotiations with Sasol, and had the effect that Sasol did not in fact price caustic soda, or NaCN, on a profit-maximising basis as suggested by Mr Smith as a matter of general economic theory. Mr Schoeman described Sasol’s change in attitude since 2022 as follows:

 

“[W]here it was previously a negotiation and almost a partnership and they would understand where we’re coming from when we say to them we can’t take this price increase because the gold price is at this level and it hasn’t moved. None of that is – that seems to be – to have dissipated in the negotiation phase.’[107]

 

[115]      The above documents suggest that Sasol did not even track the financial performance of the NaCN business on an opportunity cost basis historically – the first time it prepared pro forma stand-alone financial statements for the NaCN business which reflected third party pricing of caustic soda (and of the other key inputs) was in 2020, when it was considering the sale of the NaCN business, and wished to assess whether the business would be commercially viable on a standalone basis. Updated versions of these pro forma statements were then provided to Draslovka for purposes of conducting its due diligence on the NaCN business.

 

[116]      Ms Wainer confirmed that these pro forma financial statements, and the “third party” input prices for the stand-alone business reflected in those statements, had been prepared by the Sasol’s M&A team (not by the management of the NaCN business) specifically for purposes of Project Silver. Ms Wainer elaborated that the key adjustment related to the caustic soda price, which had been increased from an internal transfer cost to a “market price  calculated by the M&A team, and which was subsequently revised in negotiations between Sasol’s deal team and Draslovka.[108]

 

[117]      It is furthermore clear from the documents referred to above that Sasol intended throughout to sell caustic soda to the NaCN business at a third party (import parity) price post-transaction; and that both Sasol and Draslovka appreciated that this would require Draslovka to increase the price of NaCN in order protect the margins of the Target Business as a stand-alone firm.

 

Sasol’s pricing of caustic soda and NaCN after the commencement of the merger negotiation process

 

[118]      Central to the assessment of the relevant counterfactual are the changes that took place in Sasol’s NaCN pricing after it agreed to sell the Target Business to Draslovka in mid- 2021. As noted above, the proposed merger was notified to the Commission for investigation on 2 September 2021, and was prohibited by the Commission on 26 November 2021. The merger parties filed their request for consideration of the Commission’s prohibition before the Tribunal on 10 December 2021.

 

[119]      As set out above, the November 2022 memorandum states that the primary objective of the new pricing mechanism was “to return the unit gross margin of the Sodium Cyanide Asset to historical levels and ensure its economic viability.[109]

 

[120]      In particular, the “  equired gross margin  was an average gross margin of at least R[…] per ton, which was said to reflect the “historical financial performance  of the NaCN business. The November 2022 memorandum stated that this margin had dropped significantly as a result of sharp increases in the prices of key inputs, including caustic soda, which threatened the “economic viability of the NaCN business.

 

[121]      However, as set out above, it appears from the November 2022 memorandum (and was confirmed by Ms Wainer) that the “historical financial performance” of the NaCN business, and the “historical average gross margin” , referred to therein, reflected the performance of the NaCN business on a pro forma un-integrated basis. Ms Wainer also explained that these pro forma financial statements, and the input prices for the stand-alone business reflected in those statements, had initially been prepared by the Sasol’s M&A team specifically for purposes of Project Silver.

 

[122]      Having regard to all the above facts, it does not appear to us that the NaCN business would likely have made the changes to its pricing mechanism that it did in 2022 and 2023 absent Project Silver. On the evidence before us, it does not appear likely that it would have done so, because the actual internal cost of caustic soda was unaffected by the changes in the international price of caustic soda during that period; and, accordingly, those changes would not have threatened the “economic viability  of the NaCN business on an integrated basis.

 

[123]      Even if, absent the merger, the indirect exposure that the historical pricing mechanism had to international input prices might have caused the NaCN business to make some changes to that mechanism to reflect the sharp spike in international prices (e.g., by removing the “cap and collar” mechanism, as Sasol did with effect from the second quarter of 2022), there is no evidence that it would likely have gone so far as to introduce a stoichiometric pricing mechanism based on the recovery of a full import parity price for caustic soda (and ammonia), as it did with effect from January 2023. As set out in the November 2022 memorandum, that mechanism is clearly tied to the notion of a stand- alone business that is paying international “market prices” for caustic soda (and other key inputs) – which the NaCN business would not have done absent Project Silver, and indeed has not done to date.

 

[124]      This is supported by correspondence between Draslovka and Sasol commencing in August 2022, which suggests that Sasol’s introduction of an international input price- based stoichiometric pricing mechanism in January 2023 may have been a direct response to concerns raised by Draslovka regarding the decline in the performance of the NaCN business on a pro forma stand-alone basis over the first half of 2022.

 

[125]      On 1 August 2022, PWC (on behalf of Draslovka) wrote an email to Ms Wainer and Mr Cornelius Erasmus (another member of Sasol’s M&A team) in which it requested, inter alia a call “to understand the sequence of how you would restore the EBITDA margin […]% that we acquired the business on 9 July 2021, prior to the transfer of the company ”.[110]

 

[126]      This email, and the call referred to therein, followed a report by Sasol to Draslovka regarding the pro forma performance of the NaCN business over FY2022 (ended 29 July 2022), which showed that the EBITDA and gross margin of the pro forma business had declined dramatically as a result in the increase in the international prices of key inputs (in particular, caustic soda) over that period.[111]

 

[127]      The clear implication of the above email – which was confirmed by Mr Bruzek and Ms Wainer in their evidence – was that Draslovka had agreed to purchase the NaCN business on the basis of the pro forma financial statements that reflected an EBITDA margin of over […]% (expressed in later correspondence from Draslovka as a gross margin of […]%)[112] for the business on a stand-alone basis; that this pro forma margin had dropped sharply during the course of 2022 because of the sharp rise in international prices for key inputs (to which the stand-alone business was directly exposed because it would buy such inputs at import parity prices post-merger); and that Draslovka accordingly expected Sasol to take the necessary steps to “restore” this pro forma margin.

 

[128]      Mr Erasmus responded to this email on 2 August 2023, proposing a discussion the following week, and informing PWC that:

 

I will in the meantime engage with the CFO of the business in this regard. They are aware of the business situation and is [sic] managing this based [on the] short and long-term requirements of the business [113]

 

[129]      Mr Erasmus also queried whether Draslovka contended that the restoration of the business’s performance was a “closing condition” and, if so, on what “legal clause this was based.

 

[130]      On 2 August 2022, Mr Bruzek responded to Mr Erasmus agreeing to a call the following week and noting that “all the information about the business performance [is] very concerning”.[114]

 

[131]      The next correspondence in the record is an email from PWC to Ms Wainer and Mr Erasmus dated 31 August 2022 in which PWC referred to “ arious calls between the parties regarding “the problems of the performance of the target business and also to a “promise  by Sasol to “[restore] the performance of the business …. between October and December 2022 (Q2 FY 23) through the appropriate price increases” . PWC repeated later in the email that “[ f]irst and foremost, we would like Sasol to continue with its plan of restoring profitability in the target business on a monthly pro forma basis by Q2 FY23.’[115]

 

[132]      PWC also proposed two options for a reduction in the purchase price of the business as a result of the “damage  that had already been done to the NaCN business, one of which was based on the difference between (i) the actual gross margin of the business in the 12 months before the closing date and (ii) “ he gross margin that our purchase price was based on” (referred to as a “the gross margin underpin”).[116]

 

[133]      Ms Wainer responded on 1 September 2022 suggesting that the parties “d iscuss these and some other potential options”.[117]

 

[134]      The discussions between the parties led to a presentation prepared by Ms Wainer and distributed to Steercom on 30 September 2022 under cover of an email in which Steercom was requested to agree to an adjustment of the purchase price of the NaCN business on the basis that “Business performance has indirectly created difficulty for Draslovka’s structure and funding , and that Draslovka was unlikely to agree to an extension of the long-stop date for the transaction (31 December 2022) absent an adjustment of the purchase price.[118]

 

[135]      The presentation recorded that the “ EBIDTA of the stand-alone business has declined in the last 12 months due to significantly escalating input costs . . . and cyanide pricing formulas which were unable to pass on these escalations to customers rapidly enough”[119] The presentation also noted that “pricing is under review by Sasol, and FY23Q2 pricing has been increased . However, the presentation went on to explain that Draslovka was nevertheless demanding a reduction in the purchase price of the Target Business to compensate it for the effects of the declining pro forma performance of the business on its funding arrangements, and unanticipated transaction costs.[120] Ms Wainer therefore requested Steercom to approve a renegotiation of the terms of the transaction on the basis of a reduction in the purchase price and a recovery mechanism based on the future performance of the NaCN business.[121]

 

[136]      Ms Wainer obtained this approval and proceeded to negotiate revised transaction terms with Draslovka over the period October – December 2022.

 

[137]      During that period, in a report submitted by Sasol to Draslovka in October 2022 in respect of the first quarter of FY23 (01 July to 30 September 2022), Sasol informed Draslovka that:

 

Marketing & Sales have begun the process to evaluate mechanisms to protect margins against future market factors which should allow the recovery of variable costs and improve margins onwards in FY23. FY23 Q2 prices have already been communicated (provided in the VDR in 2.6.1), and price / methodology changes for FY23 Q3 and 4 will be communicated once finalised internally.[122](our emphasis)

 

[138]      This is of course consistent with the change in the NaCN’s business’s NaCN pricing model to one based on international market prices of key inputs, as referred to in the November 2022 memorandum and subsequently introduced with effect from January 2023.

 

[139]      The negotiations between Sasol and Draslovka on revised transaction terms culminated in an agreement between the parties’ negotiation teams in early November 2022 that the purchase price of the NaCN business would be permanently reduced by R […] million (approximately […]%), and would include a contingent payment calculated as the difference between (i) the gross margin actually achieved by the NaCN business in the 12 months leading up to the closing date and (ii) the “average historical gross margin reflected in the pro forma statements, subject to a cap of R […] million[123].

 

[140]      The implications of this contingent payment for Sasol were spelled out by PWC in an email to Ms Wainer and Mr Erasmus on 9 November 2022 in the following terms:

 

Please note that the quantum of the Draslovka Contingent Payment is completely in the hands of Sasol. If Sasol restores the performance over the next year, like planned, the Draslovka Contingent Payment should be zero.[124]

 

[141]      Ms Wainer expressed the same view in an email to Steercom (including Mr Mokomela) on the same day:

 

“What this means is:

 

·                         If the business is able to restore gross margin up to historical average Gross Margin of R […] million per annum – there will be little to zero contingent payment component . . .;

 

·                         If the business is not able to restore margin, Sasol risks R […] million of the purchase price dependent on Draslovka’s achievement of earnings…”[125]

 

[142]      Ms Wainer concluded the email by stating that “the transaction team is comfortable to recommend this as a reasonable outcome, as the business has committed to restore margins  [126]

 

[143]      Steercom approved this recommendation, and an amendment to the sale agreement was subsequently signed in early December 2022.[127]

 

[144]      What the above correspondence indicates is the following:

 

144.1.    Prior to Project Silver, the NaCN business did not have stand-alone financial accounts, and its profitability was assessed based on the actual costs it incurred for key inputs, including an internal transfer cost for caustic soda that was significantly lower than the market price of that product.[128]

 

144.2.    For purposes of the proposed sale of the NaCN business pursuant to Project Silver, Sasol’s M&A team created pro forma financial accounts for the NaCN business that indicated to potential bidders the profitability the business would have as a stand-alone business paying Sasol market prices for key inputs (caustic soda and ammonia) post-transaction.[129] As explained above, the market price for caustic soda is significantly higher than the internal transfer cost actually incurred by the NaCN business as a vertically-integrated firm within the Sasol Group.

 

144.3.    Based on the decline in the stand-alone financial performance of the pro forma NaCN business as a result of the sharp increase in the international prices of inputs (including caustic soda) that Draslovka would be paying Sasol post-merger, Draslovka demanded a reduction in the purchase price.

 

144.4.    Sasol agreed to such a reduction, part of which was permanent, and R […] million of which was contingent on Sasol not being able to “restore” the gross margin of the business to the levels reflected in the pro forma financial statements of the business.

 

144.5.    Sasol was willing to agree to this reduction because it was confident that it would be able to “restore” the performance of the pro forma NaCN business based on commitments that the NaCN business had made to increase the price of NaCN.

 

144.6.    The Sasol deal team was aware of steps being taken by the management of the NaCN business to increase the pricing of NaCN in a manner that would improve the performance of the pro forma business (which included a change of its pricing mechanism), and communicated these steps to Draslovka.

 

144.7.    According to both Draslovka and the Sasol deal team, the management of the NaCN business had “committed” / “promised” to take the necessary steps to improve the performance of the pro forma business – in particular, by raising the price of NaCN – and that it was doing so.

 

[145]      In her evidence, Ms Wainer laid considerable emphasis on the fact that the negotiations between Draslovka and Sasol regarding the adjustment of the purchase price only involved the Sasol “deal team” and not the management of the NaCN business. She also stressed that she never directly instructed the NaCN business to increase its NaCN prices to meet Draslovka’s demands, and stated that the steps taken by the NaCN business to increase the price of NaCN were not a response to those demands.

 

[146]      However, Ms Wainer acknowledged that Mr Mokomela was a member of Steercom, and was therefore fully aware of Draslovka’s concerns regarding the pro forma performance of the NaCN business. Mr Mokomela confirmed that he became aware of Draslovka’s concerns in mid-2022[130]. Mr Mokomela also confirmed that he was aware that the renegotiation of the sale agreement between the parties meant that Sasol would receive up to R […] million less for the Target Business if the management of the business did not “restore” its performance to its historical pro forma levels.[131] It is also clear from the correspondence above that the Sasol deal team engaged with the management of the Target Business during the period August to December 2022 regarding its pro forma performance, and that the management committed to increase the price of NaCN to restore the business’s historical pro forma performance.

 

[147]      Having regard to all these facts, it is, in our view, implausible that the decision of the NaCN business to change its pricing mechanism in 2022 and January 2023 was not attributable to Project Silver, and, in particular, that its decision to introduce a pricing mechanism based on IPP prices for key inputs (in particular, caustic soda) with effect from January 2023 was not a function of:

 

147.1.    the declining pro forma performance of the NaCN business as a carved-out business pursuant to Project Silver;

 

147.2.    Draslovka’s demands that Sasol address this decline in the pro forma p erformance of the carved-out NaCN business it had agreed to buy; and

 

147.3.    the incentives created by the renegotiated purchase price for the NaCN business for Sasol to increase the price charged for NaCN pending the transfer of the business to Draslovka.

 

[148]      This is also consistent with the reference to the target margin in the November 2022 memorandum as a “required return” . The return was certainly not “required  in terms of NaCN business’s historical pricing model, which (as the documents reflect and Mr Mokomela acknowledged) did not include any “direct margin protection”.[132] Even less did the Target Business, as a vertically-integrated entity, historically “require” a return based on the pro forma performance of the business. Rather, the notion that a return of this nature was “required” strongly suggests that this return was a specific function of Draslovka’s demands, or at least of Project Silver (which is expressly referenced in the November 2022 memorandum, as noted above).

 

[149]      As set out above, there is no evidence that, prior to Project Silver, Sasol ever priced caustic soda to the NaCN business on an opportunity cost basis, or that the NaCN business ever priced NaCN based on the opportunity cost of caustic soda.

 

[150]      Even with the spike in the international prices of inputs during the course of 2022, it does not appear to us to be likely that the NaCN business would have had an incentive to price NaCN on this basis, because those changes did not affect the internal transfer cost it incurred for caustic soda as a vertically-integrated entity. In our view, a stoichiometry- based pricing mechanism based on IPP prices for key inputs (in particular, caustic soda) only makes sense if the NaCN business is regarded as a stand-alone business buying its key inputs (including caustic soda) at import parity prices – and that was indeed the basis on which the pricing changes were explained in the November 2022 memorandum. And, in our view, the most plausible reason why the NaCN business began pricing on that basis in January 2023 was its appreciation that doing so was necessary to preserve the stand-alone value of the NaCN business it had sold to Draslovka, and thereby to avoid a potential further reduction of up to R […] million in the purchase price for the NaCN business.

 

[151]      Having regard to the evidence referred to above, we are of the view that it is unlikely that the changes to Sasol’s pricing methodology in 2022 and January 2023 would have taken place absent the proposed merger, and we accordingly conclude that they should be excluded from consideration in assessing the pricing effects of the proposed merger

 

[152]      In their heads of argument, the merger parties appear to accept that the renegotiation of the purchase price for the NaCN business provided Sasol with an incentive to increase the NaCN price, but they argue that this means the price change was unrelated to the incentives of Draslovka and hence not merger-specific on the BB Investment test. Even assuming that this is the relevant test,[133] we disagree. The fact that the changes Sasol made to the NaCN pricing methodology mitigated the purchase price reduction it would otherwise have suffered does not mean that those changes were not associated with the interests of Draslovka on the BB Investment test. Indeed, the evidence shows that the primary purpose of the price increases was to “restore” the pro forma value of the NaCN business as demanded by Draslovka, as the purchaser of the business, for the reasons explained by Mr Bruzek. Clearly, none of this would have happened absent the merger. Therefore, even on the BB Investment test, we believe that the changes to Sasol’s pricing methodology in 2022 and January 2023 should be excluded from consideration in assessing the pricing effects of the proposed merger.

 

[153]      The merger parties also argued that the changes made by the NaCN business to its pricing mechanism were necessitated by Sasol’s contractual obligation under clause 16.3 of the sale agreement dated July 2021 to manage the NaCN business prudently pending the implementation of the transaction, and that this required Sasol to ensure that the pro forma margin of the business was maintained during the interim period.

 

[154]      However, clause 16.3 of the sale agreement provides that Sasol must carry on the NaCN business “in the Usual and Ordinary Course to the standard of a Prudent Operator” where “Usual and Ordinary Course is defined as “the manner in which [Sasol] conducted the Cyanide Business in the 12 months prior to the Signature Date”. Given the evidence above that Sasol did not conduct the NaCN business on an opportunity cost, or pro forma “carved-out”, basis prior to July 2021, it does not appear to us that clause 16.3 obliged Sasol to do so after that date.

 

[155]      This notwithstanding, even if Sasol’s price changes were motivated by a perceived contractual obligation under clause 16.3 of the sale agreement to maintain the pro forma value of the NaCN business, it appears to us that that would itself be a merger-specific effect (in both senses referred to above) given that this obligation would not have been in place absent the proposed merger, and also would be aligned with the interests of Draslovka as the proposed purchaser.

 

[156]      In our view, therefore, nothing much turns on the question whether the Sasol deal team instructed the management of the NaCN business to change the pricing mechanism of the business to increase the NaCN price; or whether Mr Mokomela understood that the commercial imperative under the renegotiated sale agreement was to ensure that the historical pro forma performance of the business was “restored” without expressly being requested to do so; or whether the NaCN business understood this to be a contractual obligation under the sale agreement; or whether Sasol of its own accord wished to preserve the pro forma value of the NaCN business as a stand-alone business for purposes of Project Silver. Whichever it was, it appears to us that the changes made by the management of the Target Business to the NaCN pricing mechanism in 2022 and 2023 were likely a specific effect of the proposed merger because they were triggered by Project Silver; they were aligned with the interests of Draslovka as the proposed purchaser of the Target Business; and there is no evidence that Sasol would otherwise have changed its historical pricing mechanism to a pricing model based on IPP prices for key inputs (in particular, caustic soda).

 

[157]      For all these reasons, we are satisfied that the change made by the NaCN business to its NaCN pricing mechanism in 2022, and particularly its change to a pricing mechanism based on IPP prices for key inputs in January 2023 (which effectively mirrored Draslovka’s proposed post-merger pricing mechanism) would not likely have taken place absent the merger (and that such changes were also merger-specific on the BB Investment test), and should accordingly be disregarded in assessing the pricing effects of the proposed merger.

 

[158]      The Commission argued that the changes made by the NaCN business to its pricing mechanism in 2022 and 2023 were merger-specific for an additional reason, namely that they represented a deliberate attempt by Sasol, after the prohibition of the proposed merger by the Commission on 26 November 2021, to “game the system” by creating a false counterfactual for purposes of the merger hearing before the Tribunal.

 

[159]      We do not believe there is a sufficient evidential basis for this contention. Having regard to the evidence referred to above, we believe that the most plausible explanation for the changes Sasol made to its NaCN pricing methodology is the one set out above. This, in our view, is a sufficient basis upon which to conclude that the changes Sasol made to its pricing methodology were attributable to the proposed merger and should accordingly be excluded from the relevant counterfactual in assessing the pricing effects of the merger.

 

Sasol’s pricing approach if the merger does not proceed

 

[160]      What do we make of Mr Mokomela’s position at the hearing that, even if the merger does not proceed, Sasol does not intend to change the pricing mechanism it introduced in January 2023?

 

[161]      Mr Mokomela testified that it is necessary to keep this new mechanism in place because Sasol’s NaCN business remains under “severe pressure” as a result of the increase in the “market prices  of feedstocks, which it is unable fully to pass onto consumers in terms of Sasol’s historical pricing formula; and therefore that Sasol’s NaCN prices need to increase at least in line with such market price increases in order for the business to operate on a “c ommercially viable or sustainable basis [134]

 

[162]      However, as discussed above, this view of commercial viability is based on the notional premise that the prices paid by the NaCN business for its inputs are not the actual costs incurred by it as a vertically-integrated entity, but the market prices payable by third parties for such inputs (and, in particular, an import parity price for caustic soda).

 

[163]      If the merger does not proceed, the NaCN business would remain vertically-integrated within the Sasol Group, and there is no suggestion that it would not continue to source feedstock (in particular, caustic soda) at a low internal transfer cost rather than an external market or import parity price. No cogent evidence was presented by Sasol to explain why, notwithstanding the above, Sasol would, if the merger does not proceed, persist with a pricing model based on import parity prices for key inputs. As explained above, that approach was developed for, and only makes sense in the context of, a de- integrated NaCN business purchasing feedstock (in particular, caustic soda) at international market prices.

 

[164]      Whatever the reasons are why, as a vertically-integrated entity, Sasol has historically priced caustic soda on an internal transfer basis, and priced NaCN at a significant discount to IPP, Sasol did not provide any evidence at the hearing that any of those reasons have changed since early 2022. We note that, even as at the date of the hearing, there was no evidence that Sasol had (for independent reasons) decided to start charging its NaCN business an import parity price for caustic soda, let alone that it would continue to do so if the merger does not proceed. Therefore, no evidential basis has been provided by Sasol to support its assertion that, if the merger does not proceed, it would persist with its current NaCN pricing model based on international prices of key inputs (in particular, caustic soda).

 

[165]      There are also principles of merger control that are at issue here. If we were too readily to accept a self-serving counterfactual advanced without any obvious evidential basis at a Tribunal hearing, that would raise a considerable risk that merger parties would be able to “game the system” merely by asserting that merger-specific conduct would persist even if the proposed merger did not proceed.

 

[166]      We do not suggest that this is what Sasol has done in this case. However, this risk highlights why we are unwilling to accept the position advanced by Mr. Mokomela in the absence of persuasive evidence explaining why Sasol would, if the merger does not proceed, likely persist in a pricing practice that is inconsistent with the historical pricing conduct of the NaCN business as a vertically integrated entity, and that appears to us only to be appropriate for, and to have been designed for purposes of, a stand-alone business purchasing all its inputs (including caustic soda) at import parity prices.

 

[167]      We therefore maintain our view that the changes made by Sasol to its NaCN pricing mechanism in 2022 and 2023 should be regarded as merger-related, and that they should on that basis be excluded from the assessment of the effects of the merger on the price of NaCN.

 

Is Draslovka likely, as a result of the merger, to increase NaCN prices to IPP levels or higher?

 

[168]      Apart from the pricing changes reflected in Draslovka’s proposed “cost plus” pricing model, there is a separate question whether it is likely that the merged entity would, post-merger, increase the price of NaCN even further – up to IPP levels (or possibly beyond) and, if so, whether this too would be a merger-specific effect.

 

[169]      Given that the Target Business is a monopoly supplier of liquid NaCN in South Africa, and there are no alternatives for the use of liquid NaCN by gold mining companies in South Africa, it would be expected that, as a profit-maximizing firm, it would have the ability and incentive to price liquid NaCN up to the level of any import constraint. In this case, that hypothetical pricing constraint is represented by the costs that would have to be incurred by South African gold mining customers in importing solid NaCN and then converting it to liquid NaCN for use in their respective mining operations.

 

[170]      However, as discussed above, under its historical pricing mechanism, Sasol has in fact priced liquid NaCN at a significant discount to IPP. The Steercom presentation dated 2 June 2020 stated that Sasol historically sold NaCN at an average discount of […]%.[135] The Steercom presentation dated 13 October 2020 stated that the sales price of NaCN in the previous year was […]% below import parity.[136] In his expert report, Mr Njisane calculated that the landed price of solid NaCN was, on average, […]% more expensive than the cost of locally produced liquid NaCN over the period 2018 – 2021.[137]

 

[171]      Sasol’s internal documents recognise that this means that there is scope for the merged entity to increase the price of NaCN without inducing imports of solid NaCN. As noted above, the Steercom presentation dated 9 March 2021 recorded that prospective buyers had factored in “upside potential that assumed a “[…]% increase in sodium cyanide pricing going forward from FY24  [138]

 

[172]      As noted above, there is even a question whether IPP itself will be an effective constraint on the pricing of liquid NaCN in the South Africa in the future. Sasol explained in this regard that it is not currently feasible to import solid NaCN into South Africa at all given the current difficulties faced by South African ports in handling solid NaCN in accordance with required ICMI safety standards. Mr Mokomela testified that:

 

“Sasol is no longer able to import solid cyanide into South Africa and has not done so for the past two years. This is because the international safety rules for the importation of cyanide have changed, and the Durban port is unable to guarantee adherence to these international safety standards.[139]

 

[173]      Mr Hayward of Gold Fields[140] and Mr Schoeman of DRD[141] confirmed the difficulties currently associated with importing solid NaCN into South Africa. As Mr Schoeman explained, this means that customers of the NaCN business have “no feasible alternative sources of supply and thus no credible means of disciplining the post-merger pricing of NaCN”[142]

 

[174]      The merger parties again disputed that this is a merger-specific concern. They argued that the merged entity does not have any ability or incentive to increase prices post- merger to IPP levels (or beyond) that Sasol itself would not have absent the merger.

 

[175]      However, in our view, this again ignores the firm-specific reasons that appear to be behind Sasol’s historical practice of pricing its liquid NaCN to South African customers at a significant discount to IPP. As discussed above, these relate principally to the fact that the NaCN business is currently vertically integrated within the broader Sasol Group. As such, it already enjoys high profitability as a result of its low internal cost base and may fear complaints of excessive pricing if it seeks to profit maximize in the context of that low cost base. The NaCN business is also a relatively small business within the Sasol Group, which may have broader interests and priorities than short-term profit maximization in the NaCN business. It also appears from the evidence discussed above that the Sasol Group has a so-called “ political” interest in supporting the South African gold mining industry in a manner that is not based on purely commercial considerations.

 

[176]      None of these firm-specific factors would apply to the NaCN business in the hands of Draslovka. Mr Bruzek accepted in his evidence that the merged entity would be able to price up to import parity levels if it so wished (“the ability is there”).[143] He stated, however, that this was not consistent with Draslovka’s pricing philosophy, and that it would be “c ommercial suicide” for Draslovka to exploit South African gold mining customers in this manner.[144] Significantly, though, neither Mr Bruzek nor Mr Smith pointed to any economic factors that would constrain the merged entity from pricing up to IPP post-merger. As Mr Bruzek stated, “customers will always prefer to purchase liquid NaCN if they can source it from a closely located producer and at a price that is, once delivered, not higher than the price of self-dissolved liquid NaCN”[145]

 

[177]      Having regard to this evidence, it appears to us that Draslovka may well have a merger- specific ability and incentive to increase the price of NaCN up to an IPP level (or possibly beyond). For obvious reasons, competition authorities would be reluctant to rely on self- restraint alone to discipline the conduct of a merged entity post-merger in the absence of any other economic constraints.

 

[178]      In these circumstances, the evidence suggests to us that it is at least “reasonably possible”[146] that the Target Business would, as a result of the merger, be able to price up to IPP levels (or possibly higher).

 

[179]      Therefore, we are of the view that the NaCN business would likely be able to increase the NaCN price up to an IPP level (and possibly beyond) as a result of the merger.

 

Do the pricing effects of the merger amount to an anti-competitive effect?

 

[180]      The merger parties argued that, even if the NaCN business were to charge a higher price for NaCN post-merger than it would do absent the merger, that would not constitute an anti-competitive effect for purposes of the competition analysis under section 12A(1) of the Act. Put differently, the merger parties argued that, on the facts of this case, the price increase identified above would not constitute a “prevention or lessening of competition” as contemplated in section 12A(1),

 

[181]      The merger parties argued in this regard that the notion of a “s ubstantial lessening or prevention of competition” in section 12A(1) refers specifically to a lessening in the process of competitive rivalry and not merely to a consumer welfare effect such as a price rise that might result from a merger. Put differently, they argued that a merger- specific price rise should only be regarded as an anti-competitive effect if it is the product of reduced competitive rivalry in the market. Absent this nexus, a merger-specific price rise would (at best) be cognisable as a public interest consideration under section 12A(3) of the Act.

 

[182]      The merger parties argued that, in this case, the purchase of Sasol’s NaCN business by Draslovka would simply result in a substitution of one monopolist for another, with the same capacity constraints and the same ability and incentive to profit-maximise subject to the threat of imports of solid NaCN. The merger would therefore bring about no reduction in competitive rivalry, or relevant change in market structure, and therefore any price rise as a result of the vertical “de-integration” of the NaCN business from the Sasol Group should, at best, be assessed as a public interest effect under section 12A(3) of the Act and not as a competitive effect under section 12A(1).

 

[183]      We agree that the proposed merger would not give rise to any increased concentration in the relevant market – the NaCN business is currently a monopoly in the relevant market, and will remain so post-merger. However, we disagree that the merger will not bring about any change in market structure, or in the ability or incentive of the Target Business to raise the price of NaCN.

 

[184]      In our view, the vertical “de-integration” of the NaCN business as a result of the merger would clearly constitute a structural change in the liquid NaCN value chain. Furthermore, as a result of that vertical “de-integration”, the Target Business would pay a significantly higher cost for caustic soda, and would therefore be likely to increase the price of NaCN in order to pass on this higher input cost. We refer to the evidence above in this regard.

 

[185]      In addition, for the reasons discussed above, it appears that the constraints that have historically caused Sasol to price its liquid NaCN at a significant discount to IPP are firm- specific ones relating to the current position of the Target Business as a vertically- integrated entity within the broader Sasol Group, and which accordingly would not constrain the pricing conduct of the Target Business as a stand-alone entity in the hands of Draslovka. In this respect, the Target Business would enjoy greater market power post-merger than it would absent the merger.

 

[186]      We also agree with the Commission’s submission that the merger parties’ approach to what constitutes an anti-competitive effect appears to be inconsistent with the Constitutional Court’s judgment in Mediclinic[147] That case involved an appeal against a judgment of the CAC[148] concerning the proposed merger of two hospital groups. In its judgment, the CAC (contrary to the Tribunal) had found that the merger parties operated in different local hospital markets, and accordingly that the structure of the relevant markets, and the market power of the merger parties therein, would be unaffected by the proposed merger.

 

[187]      The CAC proceeded to state that, in these circumstances, the Commission’s concerns regarding post-merger price increases in the local markets should be appraised as a public interest consideration and not as an anti-competitive effect of the merger within the meaning of section 12A(1) of the Act.[149] The CAC stated that, because the pricing effects alleged by the Commission would flow simply from the fact that the acquiring firm would be imposing its higher (pre-merger) tariffs on the target hospitals, rather than from any enhancement in the acquiring firm’s market power as a result of the merger, these effects should be assessed in the context of the public interest assessment rather as part of the competition assessment under section 12A.[150]

 

[188]      However, the majority judgment in the Constitutional Court expressly rejected this approach, stating as follows:

 

[52]           [I]t is necessary that brief reflections be shared on the test for the assessment of harm or the substantial prevention or lessening of competition and the correct approach to constitutional interpretation.

 

[53]           Beginning with the test and the correct interpretive approach, the Appeal Court failed to give proper effect to the purpose of the Act set out in section 2(b). This is particularly so in relation to its assessment of the likely substantial prevention or lessening of competition and public interest considerations. It also misdirected itself in a material respect by construing section 12A(1)(a) and (2) of the Act as requiring that a price increase post-merger be shown to be the result of the market share changes, which it termed “enhancement of market power”. This is not the test required by the Act. And nothing in the language and context of section 12A(1)(a) and (2) allows for the assessment to be conducted with reference to the “enhancement of market power” which is not even among the factors listed in section 12A(2). There is no textual or contextual support for this new test … [T]he assessment of harm has to be conducted within the specific framework of the Act. This enquiry necessitates recourse to the Preamble to the Act and the purpose thereof as set out in section 2. The Appeal Court thus failed to follow this approach in circumstances where it was required to and innovatively laid down the “enhancement of market power” as the yardstick for the assessment to be conducted under section 12A(1)(a) and (2).

 

[54]           All that section 12A requires in this regard is that a determination be made whether there is a substantial prevention or lessening of competition. And this is ordinarily measured with reference to a potential increase in price. It does not lay down the “enhancement of market power” as the test or provide any basis for a court to do so. It follows that the majority departed from the wording of the Act which is the point of departure in statutory interpretation.

 

[55]           In its interpretation of section 12A(1)(a) and (2) of the Act, the majority overlooked sections 7(2) and 39(2) of the Constitution, thus failing to adopt the correct interpretive approach to statutes as set out in this Court’s judgments. Its approach fails to advance the purpose of the Act and to promote the spirit, purport and object of section 27 of the Constitution.”[151]

 

[189]      The merger parties argued that the Constitutional Court’s judgment in Mediclinic is not applicable to the present merger because it is distinguishable on two key grounds, namely:

 

189.1. In Mediclinic the fundamental reason why the Constitutional Court rejected the CAC’s interpretation was that it failed to give effect to the right of access to healthcare embodied in section 27 of the Constitution. The present case, by contrast, does not implicate a right in the Bill of Rights.

 

189.2.In Mediclinic the Constitutional Court found there was no basis to disturb the Tribunal’s finding that the merger parties’ hospitals were located in the same geographic market, and that the merger would therefore cause a structural change in the relevant hospital market. In the present case, by contrast, there is no suggestion of any structural change in the relevant market.

 

[190]      We agree that the facts in Mediclinic are distinguishable from those in the present case. However, it does not appear to us that these points of difference affect the application of the approach set out in the majority judgment to the present case. In particular, we do not understand the majority judgment’s approach as necessarily being limited to mergers implicating a right in the Bill of Rights, nor do we understand it as being premised on the Tribunal’s finding that the merger in that case involved a structural change in the local hospital market. On the contrary, the majority judgment states that the test in section 12A(1) does not require that a post-merger price increase be shown to be the result of increased market power on the part of the merged entity.

 

[191]      In any event, as discussed above, the merger in this case may be regarded as giving rise to a structural change in the market (albeit not one involving a reduction in competitive rivalry) in that it would involve the vertical de-integration of the NaCN business from the Sasol Group, with negative consequences for the pricing of NaCN as discussed above. The proposed merger may also be regarded as affording the Target Business greater market power than it enjoys pre-merger as a result of its de-integration from the broader Sasol Group and the NaCN pricing considerations associated therewith.

 

[192]      We therefore consider the pricing effects of the merger to be ognizable as an anti- competitive effect under section 12A(1) of the Act, and not merely as a public interest effect. However, even if we are incorrect in this regard, it would not affect our conclusion regarding the merger, for the reasons we discuss further below under the heading of Public Interest

 

THE substantiality OF THE PRICE INCREASE

 

[193]      Insofar as the pricing effect of the proposed merger is regarded as an anti-competitive effect, the question arises as to its substantiality under section 12A(1) of the Act. In Medicross, the CAC held that “substantially  in the context of section 12A of the Act mean “materially or considerably in amount or duration”[152]

 

[194]      Any negative price effect as a result of a merger arguably represents a substantial effect for purposes of section 12(1) of the Act. In this case, however, various of the gold mining company witnesses provided evidence regarding the material effects that a rise in the price of NaCN would have (and already has had) on their respective gold mining operations.

 

[195]      As set out above, Mr Harman calculated that the differential between the cost incurred by the NaCN business for caustic soda and the price charged by Sasol to third parties for caustic soda (on a Rand/ton basis) was approximately 95% over the period July 2021 – June 2022, and approximately 216% over the period July 2022 – February 2023.[153]

 

[196]      Mr Njisane calculated that, had the stoichiometric pricing mechanism been in place over the period July 2022 to January 2023, it would have generated NaCN prices that cumulatively were, on average, […]% higher than the prices in fact charged by Sasol over that period (which were themselves higher than those permissible under the historical pricing mechanism, as indicated above).[154] Mr Harman calculated a similar differential of […]-[…]% for DRD for July 2022.[155]

 

[197]      The various mining company witnesses all spoke to the significance of any increase in NaCN prices for their respective gold mining operations. They explained that, because of the materials that surface tailings operations process, NaCN accounts for a greater proportion of the working costs of such operators than of gold mining companies that extract and process virgin ore.[156] They also explained that the South African gold mining sector is a sunset industry that is experiencing an increasing trend away from underground mining operations towards surface tailings operations, with the result that NaCN will likely account for a greater percentage of South African gold mining companies’ working costs in future years than they have to date.[157] This is also reflected in Sasol’s October 2019 Debottlenecking Plan.[158]

 

[198]      Mr Bruzek agreed that:

 

“the importance of cyanide and its price effect for the mining industry will be growing because as the quality of the ore is going down and being more complex, they are expected to use more cyanide when they’re going to be applying current technology …”[159]

 

[199]      Mr Irons of PAR testified that NaCN makes […]% of PAR’s tailings operational costs.[160] Mr Pobe of Harmony explained that, in Harmony’s case, NaCN makes up approximately […]% of the monthly cash operating costs of its metallurgical plants,[161] with the result that “fluctuations in the price of an input of such significance as NaCN could have a significant financial impact on Harmony”[162]

 

[200]      As regards DRD, Mr Schoeman testified that DRD’s two subsidiaries - Ergo and Far West Gold Recoveries (“FWGR”) – are involved in the retreatment of surface tailings in the East and West Rand. In terms of DRD’s June 2022 Technical Report Summary (“TRS”), Ergo’s reserves represent a life of mine (“LOM”) of 19 years (2023 to 2041), and FWGR’s reserves represent a LOM of 20 years (2023 to 2042).

 

[201]      Mr Schoeman testified that NaCN currently represents 6% of DRD’s working costs, and that this is projected to increase to 15% over the next 10 years because of DRD’s LOM programme and the declining head grade of gold in its processing material.

 

[202]      Mr Schoeman then calculated the effect that increases in the price of NaCN would have on the LOM and net present value (“NPV”) of Ergo on (i) two different assumptions of the international gold price (one a consensus forecast of 15 international financial institutions, and the other Ergo’s own (flat-line) forecast reported in DRD’s 2022 TRS) over the LOM period; and (ii) four working cost assumptions based on different projections of the NaCN price over the LOM period:

 

202.1.    The first working cost assumption (“WCA1”) reflected the cost of delivery of NaCN to Ergo as at June 2022.

 

202.2.    WCA2 included the cost of delivery of NaCN to Ergo as at January 2023.

 

202.3.    WCA3 included a 20-25% increase in the price of NaCN above WCA2, but excluded the cost of delivery.

 

202.4.    WCA4 assumed an estimated import parity price for NaCN.

 

[203]      Mr Mochekela and Mr Vorster performed the same exercise for SSW’s Sibanye Gold and Cooke operations. They explained that NaCN currently represents […]-[…]% of Sibanye Gold’s working costs, and […]-[…]% of Cooke’s working costs.

 

[204]      WCA3 appears to have been based on a misunderstanding of Draslovka’s proposed pricing post-merger, and we therefore disregard it for purposes of our assessment. Furthermore, the above analyses cannot, in our view, be used to quantify with any precision either the pricing effects of the proposed merger, or their knock-on effects on the operations of DRD and SSW. This is because the analyses relate only to a limited number of mines, and also do not, as we understand them, contain an “apples and apples” comparison between NaCN prices that would have been charged under Sasol’s historical pricing mechanism, and those that would be charged under Draslovka’s proposed pricing mechanism, over the same time period.

 

[205]      However, the analyses are, in our view, broadly demonstrative of the sensitivity of gold mining companies to changes in the price of NaCN. For example, according to the analyses, even the increase in the price of NaCN from June 2022 to January 2023 would, all else equal, have reduced the NPV of Ergo by more than R […] million (on either gold price assumption) and, on the consensus gold price assumption, it would have reduced the LOM of Ergo by […] years. If the price of NaCN were to increase to IPP post-merger, the NPV of Ergo would, all else equal, decline by over R […] million, and its LOM would decline by at least […] years (on either gold price assumption).

 

[206]      Mr Irons confirmed in his evidence that, if PAR was required to pay IPP for NaCN, this would have a “significant effect on its cost base.[163]

 

[207]      The significance of NaCN costs to the long-term sustainability of the South African gold mining industry is also acknowledged in Sasol’s October 2019 Debottlenecking Plan, where Sasol stated that the cost of NaCN is “  for surface gold tailings facilities in particular, and accordingly “as the only local producer of sodium cyanide, Sasol has an important role to play in the future of sustainable gold production in South Africa.[164]

 

[208]      In the same document, Sasol calculated that, if gold mining customers paid an import parity price for NaCN rather than the price Sasol was charging at the time, the profitability of surface tailings re-processing operators in South Africa would decline by […]-[…]%.[165]

 

[209]      Mr Schoeman explained that the LOM and financial effects of an increase in the price of NaCN would also have significant knock-on environmental and employment effects for SSW and DRD. For instance, a reduction in the LOM of any surface tailings site would reduce the ability of DRD and SSW to rehabilitate mine dumps, which pose significant environmental risks. Similarly, a reduction in the LOM of any mining operations would result in an earlier retrenchment of employees working in those operations. As at March 2023, Sibanye Gold had 24 100 employees and 8 600 contractors, 400 of whom are involved in the Cooke operations; whilst DRD had 920 employees and 2 600 contractors.[166]

 

[210]      In our view, the above evidence supports the contention of SSW and DRD that an increase in the price of NaCN as a result of the merger is likely to have a material impact on the LOM and financial performance of their respective mining operations, with the knock-on environmental and employment consequences explained by Mr Schoeman.

 

[211]      Therefore, on the basis of the above evidence, we find that, insofar as the pricing effect of the proposed merger is regarded as a competition effect, that effect amounts to a substantial” anti-competitive effect within the meaning of section 12(1) of the Act.

 

THE LOCAL SUPPLY CONCERN

 

[212]      The second theory of harm investigated by the Commission was whether the merged entity would likely reduce the supply of liquid NaCN to South African customers in favour of exports. As noted above, the Commission found that exports of liquid NaCN would not be feasible for the merged entity, and that any manufacture and export of solid NaCN by the merged entity was unlikely to be at the expense of the supply of liquid NaCN to local customers.

 

[213]      However, in their evidence before the Tribunal, Mr Schoeman of DRD and Mr Mochekela of SSW testified that exports of solid and even liquid NaCN from South Africa are potentially practicable and profitable (because the price of solid NaCN is higher than that of liquid NaCN). They contend that this creates an opportunity for Draslovka – which does not have any historical affiliation with South Africa – to establish a manufacturing base in South Africa with a view to supplying customers both inside and outside South Africa with NaCN, and that it would have an incentive to supply NaCN to whichever customers – within or outside South Africa – yielded it the highest profit.”[167]

 

[214]      Messrs Schoeman and Mochekela stressed that security of supply of liquid NaCN is crucial to the South African mining customers. Since Sasol’s NaCN plant is the sole supplier of liquid NaCN in South Africa, and imports of solid NaCN are not feasible, it would be highly prejudicial to local gold mining customers, and their respective operations, if liquid NaCN that was required for their local operations were instead to be solidified and exported to other countries.”[168]

 

[215]      Mr Mokomela testified that Sasol had previously exported liquid NaCN but ceased doing so mainly because there is now sufficient local demand for liquid NaCN to take up all of the capacity of Sasol’s NaCN plant.[169] Draslovka’s attorneys confirmed in their correspondence with the Commission that it is commercially viable for Draslovka to produce solid NaCN for export from South Africa, and that it intended to increase production capacity in order to meet this demand, but that this would only be after all local demand for liquid NaCN had been satisfied.[170] Mr Bruzek confirmed in his evidence that exports of solid NaCN are feasible, but stated that the merged entity would only be incentivised to construct a solidification plant and export solid NaCN if there was sufficient surplus capacity after local demand for liquid NaCN had been met to justify doing so.[171] Mr Smith likewise acknowledged that “exports are a real possibility if they build the new plant but referred to the evidence of Mr Bruzek that Draslovka would be ncentivised to “s erve local first”[172]

 

[216]      The evidence accordingly suggests that exports of at least solid NaCN post-merger would be feasible and potentially profitable if there was an expansion of the current NaCN plant post-merger. The question from a security of supply perspective is, however, whether any such exports would be at the expense of local supply of liquid NaCN to South African gold mining customers.

 

[217]      In our view, there is insufficient evidence before us regarding the relative profitability of liquid NaCN sales in South Africa and solid (or liquid) exports to other countries to make a finding on this question. Draslovka (the party best positioned to do so) did not put up any pricing evidence on the relative profitability of exports and local sales, nor did the Commission or the intervenors do so. Mr Harman confirmed in his evidence that there is insufficient evidence in the record to make a determination on this question.[173]

 

[218]      We therefore leave open the question whether the proposed merger would be likely to negatively affect the security of supply of liquid NaCN to South African gold mining customers. This is a concern that is potentially capable of being addressed by a supply condition, and (as discussed further below) Draslovka was willing to tender a condition in this regard. However, difficult questions would arise regarding the necessary duration of such a supply condition and it would also not address the anti-competitive pricing concern that we have identified in relation to the proposed merger.

 

EFFICIENCIES

 

[219]      Insofar as the pricing effect of the proposed merger is regarded as an anti-competitive effect, it is necessary to consider whether that effect is outweighed by any efficiencies or other pro-competitive benefits of the merger.

 

Draslovka’s claimed efficiencies

 

[220]      The merger parties do not advance any efficiency justification for the proposed merger. This notwithstanding, we consider below the efficiency and other pro-competitive benefits that the merger parties referred to in their evidence before the Tribunal.

 

[221]      Mr Mokomela testified that Sasol decided to sell the NaCN business because it is no longer regarded as a core business by Sasol and does not align with Sasol’s future. Mr Mokomela also stated that, if the NaCN business is not sold, […][174]

 

[222]      Mr Bruzek testified that, by contrast, Draslovka intended, if the merger were approved, to introduce technological improvements to the Target Business and thereby ensure its long-term sustainability. In addition, Draslovka intends, “subject to approval processes and the carrying out of future financial assessments”, to expand the production capacity of the NaCN plant to at least 50 ktpa of NaCN within a period of three to four years.[175]

 

[223]      Mr Bruzek elaborated that this would be achieved by the construction and development of a new HCN plant alongside the Target Business’s existing operations. This new plant would be based on Draslovka’s proprietary Andrussow process, which he claimed would produce higher quality NaCN, have greater energy and carbon efficiency, and generate more synergies, than the Shawinigan process currently utilised by the Target Business. The existing operations of the Target Business would be discontinued once the new HCN plant became operational.[176]

 

[224]      Mr Bruzek also testified that Draslovka has recently brought to market a new glycine leaching technology, GlyCat, for the leaching of gold and other metals from mining ore.[177] Mr Bruzek explained that GlyCat uses a […] system of glycine and sodium cyanide, and that tests conducted by Draslovka in other countries (such as Brazil, Australia and Zimbabwe) have indicated that GlyCat enables improved recovery of gold (and other by-metals) as well as significant cost savings for gold mining companies.[178]

 

[225]      Mr Bruzek stated that the GlyCat technology will not be introduced into the South African market unless the proposed merger is approved. This is because the sale of GlyCat in South Africa would not be technically or financially feasible unless Draslovka was already a supplier of liquid NaCN in South Africa, and was thus able to ensure production of NaCN of the necessary quality required by the GlyCat technology.[179]

 

Analysis of the claimed efficiencies

 

[226]      Referring to the Tribunal’s approach to efficiency claims set out in Trident Steel[180] the Commission disputed that the above claims are verifiable, quantifiable, substantial or merger-specific. It also disputed that they would be passed on to consumers (in this case, South African gold mining companies).

 

Upgrade of the NaCN plant

 

[227]      As regards the improvements that Draslovka stated it would make to Sasol’s existing plants, the Commission argued that these were not identified with any precision, and that the evidence demonstrated that Sasol’s existing plant remained effective and highly profitable notwithstanding its age. There was also no evidence that any benefits from these improvements would be passed on to consumers given the absence of any competitive constraint on the NaCN business as a monopoly supplier.

 

[228]      We agree that the proposed upgrades to the Sasol plant improvements have not been defined with any specificity, or sufficiently quantified. Draslovka acknowledged in subsequent correspondence regarding the proposed conditions that it “cannot determine or give particularity regarding the upgrades that are required”. The only detail Draslovka could give was that it “intends to guarantee long-term operational stability of the plant, improve the performance of the plant, ensure the environmental and social sustainability of the NaCN production plan and improve the quality of NaCN so it can be used with the GlyCat technology”[181]

 

[229]      Even to the extent that these benefits should be characterised as pro-competitive rather than public interest benefits of the merger, they are , in our view, not verifiable and have not been quantified.

 

[230]      The evidence in the record also provides some support for the Commission’s contentions regarding the state of the existing NaCN plants. For example, the “Project Silver Transaction Feasibility  document dated 2 June 2020 states that, while the plant is old, it is “well maintained and in a good operating condition”[182] There is no suggestion in that document, or in the 2019 Debottlenecking Plan, that the plant is in serious need of refurbishment.

 

[231]      Mr Pobe stated that Sasol had experienced a six week breakdown in 2022, but he subsequently indicated that this may have been a result of vis majeure (lightning) rather than any consistent problems with the existing NaCN plant.[183]

 

[232]      SSW also complained to Sasol in December 2022 about more frequent plant breakdowns in recent years. However, this was disputed at the time by Sasol, which explained that it had to conduct two statutory maintenance shutdowns of the NaCN plant per year; and that there had only been one incident in 2022, which related to technology and not to the performance of the plant.[184]

 

[233]      There is also no evidence of safety concerns with the plant. Mr Mokomela has confirmed that, even absent the merger, Sasol would ensure that the NaCN plant is safely operated.[185]

 

Capacity expansion

 

[234]      As regards the prospect of a new plant with increased production capacity, the Commission pointed out that there was no certainty that this would take place, as Draslovka was unwilling to commit to the establishment of the new plant until it had implemented the proposed merger and was able to ascertain the future demand for NaCN and GlyCat in South Africa. Nor is there any quantification of the price at which such NaCN from the new plant would be made available to South African gold mining customers, save that it would be significantly higher than the NaCN price historically charged by Sasol. We discuss this further under the section headed “P roposed Conditions below.

 

[235]      The Commission also argued that, if there was demand for more NaCN in South Africa, Sasol would (despite the position it adopted at the Tribunal hearing) be incentivized to increase the production capacity of its plant absent the proposed merger.

 

[236]      The Commission relied in this regard on Sasol’s October 2019 Debottlenecking Plan.[186] This was a “Pre-feasibility Business Plan” prepared by Sasol Base Chemicals which recommended further consideration of a project to increase the capacity of Sasol’s NaCN plant to meet the expected increase in demand for NaCN from the South African gold mining industry.

 

[237]      According to the Debottlenecking Plan, the NaCN business was (at the date of that document) one of Sasol Base Chemicals’ best performing products in terms of its contribution margin. The document also stated that the existing Sasol plant was operating […] capacity, and that the demand for NaCN is expected to increase in future years as the South African gold mining industry increasingly turns from the mining of virgin ore to the retreatment of tailings (which requires a higher use of NaCN, as discussed above). This was also the evidence of Sasol and the mining company witnesses at the merger hearing.

 

[238]      The Debottlenecking Plan therefore investigated the feasibility of expanding the capacity of the NaCN by approximately […]ktpa in order to meet this expected increased demand, and found that such an expansion was likely to be profitable. The Plan further noted that “[…]”.[187]

 

[239]          Sasol argued that no reliance could be placed on the Debottlenecking Plan document because it was prepared in 2019, and Sasol had subsequently decided to sell the NaCN business pursuant to its asset disposal programme, rather than to follow the recommendation contained in the Debottlenecking Plan. As noted above, Mr Mokomela stated that Sasol does not, given its current priorities, […]. Mr Mokomela stated further that he was not aware of any request to reconsider the classification of the Target Business as an asset for disposal, which would require a new determination by the relevant investment committees at Sasol.[188]

 

[240]      The Commission responded that, even if that is Sasol’s position currently, there is no reason why Sasol would not reconsider this absent the merger given all the factors mentioned in the Debottlenecking Plan, namely the highly profitable nature of the business; the increasing demand for NaCN in South Africa; and the risk that, if Sasol does not expand its operations to meet this increasing demand, it may induce entry from imports, which would be to the long-term detriment of the business.

 

[241]      We agree that, on the face of it, there appear to be good commercial reasons for Sasol to give further consideration to expanding the capacity of the NaCN plant absent the merger, despite Sasol’s stance at the Tribunal hearing that it currently has no intention to do so. However, it is not necessary for us to make a finding in this regard because, even assuming that a capacity expansion by Draslovka would be merger-specific, the likelihood of the expansion, and the pricing of any product therefrom, is, in our view, too uncertain for us to be able to attribute any significant weight to this benefit.

 

[242]      The merger parties argued that two of the initial intervenors, PAR and Harmony, decided to withdraw their opposition to the proposed merger on the grounds, inter alia that security of supply of liquid NaCN was more important to them than an increase in the price of NaCN. On the other hand, however, Gold Fields, SSW and DRD remained opposed to the proposed merger throughout, essentially on the grounds that it was preferable for them to have access to the existing output of the NaCN plant on Sasol’s historical pricing basis than potential access to a greater volume of NaCN at a significantly higher price.

 

[243]      Given that PAR and Harmony had initially opposed the proposed merger on similar grounds to SSW and DRD, we requested them to explain the reasons why they had decided to withdraw their opposition to the merger.

 

[244]      Mr Irons of PAR explained that its changed stance followed engagements with Draslovka during the latter part of 2022 in which Draslovka had indicated its intention to invest in increased capacity which would give PAR greater security of supply. PAR had also gained the impression that Sasol’s existing plant might not be reliable given certain shortages it had experienced. In addition, Mr Irons said that PAR valued the transparency in the term sheet offered by Draslovka, and that there was, in any event, no longer any material difference between Sasol’s NaCN pricing and that proposed by Draslovka.[189]

 

[245]      As regards the transparency of Draslovka’s term sheet, Mr Irons acknowledged that Sasol’s new stoichiometry-based pricing model has the same “ ook and feel”[190]

 

[246]      Mr Pobe’s evidence was to similar effect. He explained that Harmony also changed its stance on the merger after engagements with Draslovka. Mr Pobe elaborated that Harmony was ultimately more concerned about security of supply than pricing, especially because it will require increasing volumes of NaCN in the future, and does not regard importing solid NaCN (at a higher price than that currently proposed by Draslovka) as commercially viable. Draslovka informed Harmony that it intended to invest in greater, and more reliable, capacity than Sasol would provide absent the merger. Harmony also understands from Sasol’s public statements that Sasol regards the NaCN business as non-core and is unwilling to invest further in it.[191]

 

[247]      Mr Pobe also remarked that Sasol’s current pricing is no longer materially different from that proposed by Draslovka, although he expressed a concern that Sasol’s 2022/2023 price rises might have been merger-related.[192]

 

[248]      The decisions by PAR and Harmony to change their stance on the proposed merger therefore appear to have been motivated by two key factors, namely (i) an assumption that Draslovka would invest in increased capacity post-merger; and (ii) an acceptance that there is no longer any material difference between Sasol’s actual pricing and Draslovka’s proposed pricing.

 

[249]      However, for the reasons discussed above, we are of the view that Sasol’s more recent significant price increases are related to the proposed merger and therefore should not be taken into account when determining the relevant counterfactual to the proposed merger. In addition, Draslovka indicated at the Tribunal hearing that it was not able to make any commitment to invest in new capacity. It therefore appears to us that the decisions of PAR and Harmony to change their stance on the proposed merger were influenced by factual assumptions that are not supported by the evidence that emerged at the hearing.

 

[250]      It is also significant in this regard that Gold Fields, SSW and DRD all remained opposed to the proposed merger for reasons that are consistent with our understanding of the available evidence.

 

GlyCat

 

[251]      As regards the potential introduction of GlyCat into South Africa, this appears to be merger-specific on the evidence of Mr Bruzek, but it is unclear how valuable it would be for South African gold mining customers given that it is largely untested in the specific mining conditions that exist in South Africa.

 

[252]      Mr Schoeman of DRD explained that it would take millions of rand over a period of up to 2 years to test the cost-effectiveness of GlyCat for the distinct minerology of its various different tailings dams.[193]

 

[253]      Mr Mochekela stated that SSW required further information regarding the nature of the GlyCat technology in order to assess its potential value.[194]

 

[254]      Mr Irons of PAR stated that he was “not excessively excited” about the benefits of GlyCat for PAR’s operations, and that “extensive further test work and engagements” with Draslovka would be required in order to determine whether GlyCat would be of benefit for PAR. According to Mr Irons, Glycat currently is not the shining light that I would have hoped it might have been[195]

 

[255]      Mr Pobe of Harmony stated that GlyCat could potentially have benefits for Harmony, but that it was awaiting the results of samples it had sent to Draslovka for testing to assess the viability of using GlyCat in its South African operations.[196]

 

[256]      Mr Hayward of Gold Fields said that Glycat was an unproven technology in South African conditions, and that he was “unconvinced” that Glycat would have any efficiency benefits for Gold Fields’ operations.[197]

 

[257]      Mr Bruzek acknowledged in his evidence that Draslovka has never tested Sasol’s sodium cyanide so there is no certainty it would be effective in the production of GlyCat[198] He also confirmed that Draslovka has not tested GlyCat technology against South African ores, so “it’s not 100% guarantee or certainty that it’s going to work[199]

 

Conclusion on claimed efficiency benefits

 

[258]      Having regard to the above evidence, it appears to us that the main potential benefit of the proposed merger from an efficiency perspective would be an increase in the capacity of the NaCN plant. However, as discussed above, there is no certainty that this investment would take place post-merger nor has there been any quantification of the price at which NaCN from the new plant would be sold to South African customers. Therefore, an investment in additional capacity would not address the pricing concern referred to above, namely that any NaCN produced by the plant would be sold at an elevated price given inter alia the IPP cost that the business would be having to pay for caustic soda post-transaction. Indeed, it appears to us that any such investment would increase the costs on which the NaCN price is based.

 

[259]      Furthermore, as was explained at the hearing, any NaCN capacity constraint would simply reduce the extraction rate of a gold mining company, whereas a significant increase in the price of NaCN might reduce its entire life of mine, with all the operational, employment and financial consequences associated therewith.[200]

 

[260]      We therefore find that, on the evidence before us, the claimed efficiencies of the proposed merger would not outweigh the pricing effect to which it would likely give rise.

 

[261]      We therefore conclude that the anti-competitive effects we have identified in relation to the proposed merger are not outweighed by any efficiency or other pro-competitive gains associated with the merger.

 

PUBLIC INTEREST

 

[262]      There are two instances in section 12A of the Act where considerations of public interest arise.

 

262.1.    First, under section 12A(1), if the Tribunal determines that a merger is likely to have a substantial anti-competitive effect, it must determine “whether the merger can or cannot be justified on substantial public interest grounds by assessing the factors set out in subsection 3” (section 12A(1)(b)).

 

262.2.    Second, under section 12A(1A), irrespective of its determination in section 12A(1), the Tribunal must also make the above public interest determination.

 

[263]      Therefore, a merger that is anti-competitive may nevertheless be justified on substantial public interest grounds. Conversely, a merger that is not anti-competitive effect may nevertheless not be justified on substantial public interest grounds.[201]

 

[264]      As regards the public interest assessment under section 12A(3) of the Act, the Tribunal has previously explained that it is a holistic one, in terms of which the different public interest grounds listed in section 12A(3) must be separately assessed, and then, if necessary, weighed against each other in order to arrive at a net conclusion on the public interest effects of the merger.[202]

 

Section 12A(1) assessment

 

[265]      Dealing first with section 12A(1), if we are correct that the pricing effect we have identified in relation to the proposed merger should be regarded as a substantial anti-competitive effect, the enquiry is whether the merger is nevertheless justifiable on substantial public interest grounds.[203]

 

[266]      The merger parties argue that the proposed merger is justifiable in the public interest on three grounds, namely that it would:

 

266.1.    have a positive employment effect under section 12A(3)(b) because it would not result in any retrenchments (given undertakings made by Draslovka in this regard), and would create a further management position and improved exposure and training for employees of the Target Business;

 

266.2.    have a positive effect under section 12A(3)(c) because Draslovka would allocate 25%, plus one share, in Draslovka SA to Navuka, Draslovka’s B-BBEE partner; and

 

266.3.    have a positive effect on “a particular industrial sector or region  under 12A(3)(a) in that it would provide gold mining companies in South Africa with security of supply of liquid NaCN.

 

[267]      In our view, the first benefit is not substantial, and the second benefit is not sufficiently substantial to justify the approval of the proposed merger notwithstanding the significant merger-related pricing effect we have identified.

 

[268]      Regarding the third benefit, for the reasons we have discussed above under the heading of “Efficiencies”, we do not believe that this benefit is sufficiently likely to arise, or to outweigh the pricing effects of the merger, to justify the approval of the merger on public interest grounds.

 

Section 12A(1A) assessment

 

[269]      Moving to the section 12A(1A) assessment, the CAC stated the following in Mediclinic

 

“The fact that price effects must, in the present case, be assessed in the context of public interest rather than SLC has an important effect on the evidence dealing with the relative efficiency of the targets and Mediclinic. Where it is shown that a merger is likely to substantially prevent or lessen competition in a relevant market, it is for the merger parties to establish that the merger is likely to result in technological, efficiency or other pro-competitive games which will be greater than, and will offset, the effects of any prevention or lessening of competition (s 12A(1)(a)(i)). Because no SLC has been shown in the present case, the merger parties do not attract this onus. In the context of public interest, we are trying to ascertain what prices are likely to prevail at the targets if the merger is allowed, and whether (assuming such prices to be higher than they would otherwise have been) this is a sufficient basis to prohibit the merger on public interest grounds. The efficiencies which were the subject of factual and actuarial evidence are simply part of this predictive exercise.

 

We were not addressed on questions of onus and sufficiency of proof in relation to the prohibition of a merger on public interest grounds. It seems to me that in absence of evidence that a particular harm, which is substantial, may eventuate if the merger is approved, the prohibition of the merger cannot be ‘justified’ within the meaning of s 12A(1). I leave open the question whether this requires the likelihood of harm to be established on a balance of probability or whether it suffices that the danger of such harm is reasonably possible.[204] (our emphasis)

 

[270]      Therefore, if we are incorrect in our view that the pricing effect we have identified should be regarded as an anti-competitive effect of the merger, we need to consider whether that effect constitutes a significant adverse public interest effect and, if so, whether the proposed merger can or cannot be justified having regard to that effect and any other public interest effects of the merger.

 

[271]      The merger parties argue that a pricing effect is not cognisable in and of itself as a public interest effect under section 12A(3) of the Act because it is not one of the five public interest grounds that are listed in section 12A(3). They accept that price may potentially give rise to a substantial effect under one of those five listed grounds, but submit that this nexus must be established on a balance of probabilities.

 

[272]      In the present case, the Commission and SSW/ DRD contend that the pricing effect is cognisable as an effect on “a particular industrial sector , being the South African gold mining industry, within the meaning of section 12A(3)(a) of the Act. The merger parties argue that it must therefore be shown that the price effect will cause substantial harm to the gold mining industry as a whole, and that this onus has not been discharged on the available evidence.

 

[273]      n Mediclinic the CAC held that an effect for purposes of section 12A(3)(a) includes an effect on competitors or consumers within a particular sector or region,[205] and that a merger-specific price rise may potentially give rise to such an effect. It must however be shown to be a “substantial effect.[206]

 

[274]      For the reasons discussed above, we believe that the likely price effects of the merger will have a substantially negative effect on the South African gold mining sector. The price rises caused by Sasol’s changes to its NaCN pricing mechanism in 2022 and 2023 have themselves been significant, and there is no constraint on them rising even further post-merger towards (or perhaps beyond) import parity levels. We have discussed above the evidence regarding the substantiality of these price increases, and regarding the substantiality of their effect on the operations of South African gold mining companies. Furthermore, this effect is likely to increase in the future as the South African gold mining sector increasingly engages in (NaCN-intensive) surface tailings operations.[207]

 

[275]      This negative effect must be weighed against the potential benefit of greater NaCN capacity post-merger. However, for the reasons discussed above under the heading of “Efficiencies , we do not believe that this potential benefit outweighs the negative pricing effect of the proposed merger on the South African gold mining industry. In particular, there is no certainty that Draslovka would construct a new plant post-merger, and even if it did so, the price of product from the new plant would likely be higher because any such investment would increase the costs on which the NaCN price is based).

 

[276]      For all these reasons, it appears to us that the proposed merger will likely have a substantially negative effect on the South African gold mining sector. In reaching this conclusion, it is unnecessary for us to make any determination on whether this likelihood needs to be established on a balance of probabilities or merely as a reasonable possibility (a question left open by the CAC in Mediclinic) , because in our view the likelihood has been established on either test.

 

PROPOSED CONDITIONS

 

[277]      As regards the approach to be followed in circumstances where conditions are tendered to address the negative competition or public interest effects of a merger, the CAC stated the following in Imerys

 

[40]          Where, in the situation just mentioned, the Tribunal is asked to approve the merger with conditions rather than prohibit it, the choice of remedies is in the nature of a discretion. I reject the proposition that the Commission bears the burden of proving that the proposed conditions will not adequately address the likely SLC. The Tribunal has the power to prohibit the merger if it is not satisfied that the conditions will adequately remedy the likely SLC. And regardless of where the onus lies in respect of proposed conditions (if it is accurate to speak of onus at all), I do not think that 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. I think it is permissible for the Tribunal to reason thus: The merger will likely give rise to an SLC. Although the proposed conditions are more likely than not to remedy the likely SLC, there is a reasonable possibility that they will fail to do so. Therefore we prohibit the merger.’

 

[41]          Particularly where the uncertainty about the adequacy of the conditions concerns the likely duration of the SLC rather than the nature and content of the SLC, prohibition has this advantage over conditional approval: it does not necessarily represent the final word. If the merger is conditionally approved and the conditions turn out to be inadequate to neutralise the SLC, the harm cannot be reversed. If, on the other hand, the merger is prohibited and with the passing of time it becomes clear that the merger will no longer give rise to SLC, the transaction can be renewed.

 

[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.[208]

 

[278]      In this case, the conditions proposed by Draslovka underwent various iterations during the course of the Tribunal proceedings, including after argument had been completed, which necessitated a further series of submissions by all the parties.

 

[279]      The conditions finally proposed by Draslovka contained the following main commitments:

 

279.1.    the B-BBEE transaction with Navuka;

 

279.2.    that the proposed transaction will not result in any job losses;

 

279.3.    priority of supply of liquid NaCN to South African gold mining customers;

 

279.4.    the offer of supply agreements to all South African gold mining customers on the same terms mutatis mutandis as those contained in the term sheets agreed with PAR and Harmony (the “Term Sheets”);

 

279.5.    a commitment to make a capital expenditure investment of US$30 million in the introduction of Glycat technology, a new plant and/or upgrades of the existing NaCN plant subject to certain conditions (discussed further below); and

 

279.6.    extended supply agreements, subject to various revisions, depending on whether a new plant is established (also discussed further below).

 

[280]      As regards Draslovka’s investment commitment, this involves a commitment to incur capital expenditure of at least US$30 million over a 5-year period:

 

280.1.           to make available the Glycat technology to South African gold mining customers, subject to “sufficient demand” (10kt of NaCN consumption) for the Glycat technology, and on “commercially reasonable terms and conditions

 

280.2.           to undertake “necessary and appropriate” upgrades to Sasol’s existing NaCN plant; and/or

 

280.3.           to construct a new NaCN plant with a capacity of 50-55kt using the Andrussow process – “provided that the customers are willing to purchase the Product on commercially reasonable terms and conditions and that supply is economically and technically feasible in the ordinary course

 

[281]      Draslovka explained in response to subsequent queries from the Tribunal that it is willing to commit to invest $30 million but “ t wishes to be able to choose whether it should be done through an investment in a [new plant], upgrading the existing plant or introducing GlyCat technology”[209]

 

[282]      We have addressed above the untested nature of GlyCat in South African mining conditions, and the doubts expressed by the gold mining company witnesses regarding its potential value in their respective operations in South Africa. It is also unclear what is meant by “c ommercially reasonable terms and conditions  in relation to the price at which GlyCat would be made available to South African consumers.

 

[283]      As regards the possible plant upgrades, as noted above Draslovka stated that it “cannot determine or give particularity regarding the upgrades that are required , but that it intends to guarantee long-term operational stability of the plant, improve the performance of the plant, ensure the environmental and social sustainability of the NaCN production plan and improve the quality of NaCN so it can be used with the GlyCat technology[210]

 

[284]      As regards the new plant, Draslovka stated that its commitment to building a new plant of 50-55kt is subject to there being the requisite demand for more NaCN volumes, which could not be determined until the extent of adoption of GlyCat post-merger has been ascertained – if there was significant uptake of GlyCat, a new plant “may not be necessary or commercially viable in South Africa”[211]

 

[285]      As regards the nature and duration of the NaCN pricing remedy, we have noted above that the existing Term Sheets agreed with Harmony and PAR contain a 3-year ex-works “cost plus” pricing mechanism for NaCN with the following three components:

 

285.1.    the unit price of each of the four key inputs (caustic soda, ammonia, natural gas and electricity) multiplied by the relevant consumption factor to produce one ton of NaCN;

 

285.2.    other processing and cash fixed costs; and

 

285.3.    a margin of […]% on the above two components, that will initially be set at R[…] per tonne (subject to a discount depending on volumes purchased), and thereafter adjusted annually based on movements in the consumer price index.

 

[286]      We have explained above that the unit prices of the four key inputs are based on supply agreements concluded by Sasol with Draslovka, in terms of which caustic soda and ammonia are priced on a full import parity basis, whilst natural gas and electricity are based on the regulated prices of these products.

 

[287]      The merger parties have emphasised that customers have the option under the Term Sheets to source caustic soda of the same specification at cheaper prices than those provided by Sasol and to deliver it to the NaCN business. However, there was no evidence as to what amounts of caustic soda are available for purchase from other suppliers, or the prices at which they would be available. There was certainly no suggestion that any such volumes would be sold at or near the internal transfer cost incurred by Sasol’s NaCN business for caustic soda as an integrated entity within the Sasol Group. Therefore, as Mr Hayward remarked, this optionality is unlikely to make much difference to the price payable by gold mining companies for caustic soda post- merger.[212]

 

[288]      Under the proposed conditions, Draslovka has offered revised supply agreements as follows:

 

288.1.    Supply agreements for NaCN produced by the existing Sasol plant will retain the provisions of the Term Sheets save that they will be extended to apply until the earlier of 10 years and the operation of the new plant (if any), subject to a re-basing of the cost base in the pricing mechanism after 3 years to reflect the actual costs of the business at that date.

 

288.2.    Supply agreements for NaCN produced by the new plant (if any) will have a duration of at least 5 years and contain materially the same provisions as the Term Sheets, subject to (i) a rebasing of costs after 3 years as indicated above, (ii) a variation of consumption factors to reflect the usage of key inputs in the Andrussow process; and (iii) a change in margin if that change “will not result in the price per tonne of Product being higher in the first month of the New Supply Agreement than in the preceding final month of the Supply Agreement

 

[289]      Draslovka also undertakes to “negotiate in good faith  regarding any extensions of the above agreements.

 

[290]      The Tribunal afforded the parties a further opportunity to engage with each other on the conditions tendered by Draslovka after the merger hearing in order to determine whether mutually acceptable conditions could be identified. However, those engagements were not successful.

 

[291]      We have given careful consideration, in accordance with the guidance provided in Imerys  to the final conditions proposed by Draslovka, and we have concluded that they do not adequately address the pricing effect that will arise from the proposed merger.

 

[292]      Recall that the relevant pricing effect concerns a raising of NaCN prices to reflect, inter alia the purchase of caustic soda at an IPP rather than the much lower internal transfer cost the NaCN business has historically incurred for that input. We have also found that the Target Business is, as a result of the merger, likely to raise the price of liquid NaCN up towards an import parity level for solid NaCN (and possibly even above that depending on the feasibility of such imports). The pricing effect is therefore a structural one, and would be indefinite in its duration.

 

[293]   The pricing remedy offered by Draslovka does not address this pricing effect because (i) it is time-bound and (ii) it is premised on the very pricing effect that needs to be addressed, namely the purchase of caustic soda at IPP by Draslovka from Sasol. As set out above, both the supply agreements offered in relation to NaCN from the existing NaCN plant, and those offered in relation to NaCN from the new plant (if any), would enable Draslovka to recover the full IPP that Sasol would charge for caustic soda post- merger, plus other costs and a margin of […]% (or at least […]% in the case of product from the new plant).

 

[294]      Given the extent of this price effect, it also would not be addressed, insofar as NaCN from the new plant (if any) is concerned, by Draslovka’s post-hearing proposal to pass on to customers […]% of any efficiencies generated by the new plant.[213]

 

[295]      The proposed pricing remedy therefore does not address the pricing effect to which the merger is likely to give rise; on the contrary, it entrenches it.

 

[296]      Furthermore, after the expiry of the proposed supply agreements, there is (as discussed above) no evidence of any constraints (other than possible imports of solid NaCN) that would likely discipline Draslovka’s pricing of NaCN to South African gold mining customers post-merger. A commitment to “good faith negotiations” regarding the terms of any further supply agreements does not guarantee that even the existing terms offered by Draslovka would be extended to gold mining customers in the future.

 

[297]      It is important to bear in mind in this regard that the South African gold mining sector is a highly significant part of the South African economy, and responsible for the employment of thousands of people.

 

[298]      In the light of all of the above considerations, the US$30 million capital investment tendered by Draslovka, and its conditional undertaking to install a new plant in South Africa, do not, in our assessment, justify the approval of the merger. Indeed, in terms of the conditions proposed by Draslovka, the construction of a new plant would further increase the price payable by South African gold mining customers for NaCN from the new plant (which has not in any event been quantified by Draslovka).

 

[299]      A further difficulty raised by the Commission with the proposed conditions is that they are not readily capable of monitoring and enforcement. This applies not only to the terms of the pricing mechanisms themselves, but also the vaguely defined nature of, and conditions attached to, Draslovka’s proposed investment commitments. This further increases the risk that the proposed conditions will not be effective.[214]

 

[300]      We therefore conclude that the remedies proposed by Draslovka do not sufficiently address the adverse pricing effects of the proposed merger, are lacking in specificity and certainty, and would not be capable of effective monitoring and enforcement by the Commission.

 

CONCLUSION

 

[301]      For all the reasons set out above, we find that the proposed merger is likely to give rise to a pricing effect which amounts to a substantial anti-competitive or public interest effect, and which is not outweighed by any pro-competitive or public interest benefits associated with the proposed merger.

 

[302]      We are also of the view that the conditions proposed by Draslovka are unlikely to address the adverse pricing effect of the merger, are lacking in specificity and certainty, and would not be capable of effective monitoring and enforcement by the Commission.

 

[303]      We therefore prohibit the proposed merger.[215]

 

Signed by:Jerome Wilson

Signed at:2024-02-01 15:10:18 +02:00

Reason:Witnessing Jerome Wilson

 

1 February 2024

Date

Adv Jerome Wilson SC          

 

Mr Andreas Wessels and Prof Liberty Mncube concurring.

 

Tribunal case managers:	Kameel Pancham and Leila Raffee
For the merger parties:	Adv Michelle le Roux SC, Adv Kerry Williams and Adv Lebogang Phaladi instructed by Dentons and Bowmans.
For the Commission:	Adv Ngwako Maenetje SC, Adv Lerato Zikalala and Adv Silindile Mhlongo.
For Harmony Gold Mining Company Limited:	Adv Michael Mbikiwa instructed by Herbert Smith Freehills.
For Sibanye Stillwater Limited:	Adv Robin Pearse SC, Adv Katlego Monareng and Adv Craig Thomas instructed by Werksmans Attorneys.
For Pan African Resources PLC:	Mr Anton Roets of Nortons Inc.

---

---