Imerys South Africa (Pty) Ltd and Andalusite Resources (Pty) Ltd v The Competition Commission

In January 2015, Imerys South Africa (Pty) Ltd ('ISA') and Andalusite Resources (Pty) Ltd ('AR') notified the Competition Commission of an intermediate merger whereby ISA would acquire all shares in AR. Both parties mine andalusite, a mineral used to make refractories (materials that line furnaces and kilns). ISA and AR are the only two andalusite producers in South Africa, making this a two-to-one merger creating a domestic monopoly. The Commission prohibited the merger in April 2015. The appellants referred the matter to the Competition Tribunal, which confirmed the prohibition on 21 September 2016. The appellants proposed conditional approval with: (1) five-year supply agreements at current prices plus PPI increases; and (2) post-five-year domestic price cap at export parity price ('EPP'). Economic experts agreed that if both parties became capacity-constrained within five years, the proposed conditions would adequately address anti-competitive effects; if not, the conditions would be inadequate. The central issue became when both parties would become capacity-constrained.

• Whether a merger likely to substantially lessen competition should be conditionally approved or prohibited where future market conditions are uncertain
• The meaning of 'likely' in the phrase 'likely to substantially prevent or lessen competition' in section 12A(1) of the Competition Act
• The incidence of the burden of proof regarding the adequacy of proposed merger conditions
• The standard of appellate review applicable to the Competition Tribunal's choice between prohibition and conditional approval
• The appropriate method for assessing future demand growth and production capacity constraints in merger analysis
• Whether public interest considerations support prohibition where a merger would divert domestic supply to export markets

Appeal dismissed with costs, including costs of two counsel. The Competition Tribunal's prohibition of the merger was upheld.

Where a merger would cause a substantial lessening of competition by creating a monopoly, and the merging parties propose conditional approval based on the contention that future market developments will eliminate the anti-competitive effects within a specified period, the Competition Tribunal may legitimately prohibit the merger if: (1) there is a reasonable possibility (not just a probability) that the proposed conditions will fail to adequately remedy the substantial lessening of competition because market conditions may not develop as the parties predict; (2) the merger would irreversibly change market structure; and (3) there are no countervailing pro-competitive gains or public interest considerations justifying the merger. In assessing whether proposed conditions adequately address anti-competitive effects, the Tribunal is entitled to consider all reasonably possible future scenarios, including scenarios at the conservative end of plausible ranges for demand growth and production capacity, as well as possibilities such as economic shocks and capacity expansion. The choice between prohibition and conditional approval involves an exercise of discretion by the Tribunal, which appellate courts should be slow to interfere with, particularly given the Tribunal's specialist composition and expertise.

Rogers JA made several important non-binding observations: (1) On the meaning of 'likely' in section 12A(1): There is much to be said for interpreting 'likely' as meaning 'reasonably probable' rather than 'more probable than not', particularly given that sections 4 and 5 of the Act refer to conduct that 'has the effect' of substantially lessening competition (suggesting a higher standard), while section 12A uses 'likely to' (suggesting a lower standard). However, absent full argument, the Court declined to express a final view. (2) On burden of proof: The distinction between the burden of proof and that which must be proved is important. Facts from which conclusions are drawn about future risks must be established on a balance of probability, but the conclusion itself may be that there is (for example) a 20% risk of something occurring. (3) On the nature of the Tribunal's discretion when choosing between prohibition and conditional approval: There is much to be said for characterizing this as a 'true discretion' (choice between permissible alternatives) rather than a discretion in the loose sense (only one right answer), though this was not definitively decided. (4) On capacity expansion: The Court expressed skepticism about the parties' claim that demand growth would not attract investment in capacity expansion, noting that capacity expansion by either party would create additional years of surplus capacity and potential competitive effects not addressed by the proposed conditions. (5) On expert evidence: The Court cautioned that expert witnesses should be confined to testimony within the scope of their expertise, noting that economists were questioned extensively on matters of demand growth and production capacity where they had no particular expertise.

This case establishes important principles for South African competition law on merger assessment in conditions of uncertainty: (1) It clarifies that when choosing between prohibition and conditional approval of a merger that would cause SLC, the Tribunal may consider all reasonably possible future scenarios, not just the most probable scenario. (2) It confirms that the Tribunal may prohibit a merger where there is a reasonable possibility (not just a probability) that proposed conditions will be inadequate, particularly where the merger irreversibly changes market structure (duopoly to monopoly). (3) The judgment provides guidance on interpreting 'likely' in section 12A(1), suggesting it may mean 'reasonably probable' rather than 'more probable than not', though this was not definitively decided. (4) It demonstrates the weight given to internal business documents and projections in assessing parties' claims about future capacity constraints. (5) The case shows how public interest considerations (particularly effects on domestic customers and downstream industries) can legitimately be considered in choosing the appropriate remedy. (6) It establishes that prohibition is not necessarily final - if market conditions change, parties can re-notify the transaction. This case is significant for two-to-one mergers and situations where parties claim future market developments will eliminate anti-competitive effects. It demonstrates the courts' willingness to uphold prohibitions where long-term market predictions are uncertain and conditions are inadequate to address all reasonably possible scenarios of competitive harm.