Brief Summary

Competition — Merger Approval — Competition Tribunal approving merger between Main Street 333 (Pty) Ltd and Kumba Resources Limited without conditions — Merger involves complex restructuring of Kumba into two companies with significant black ownership — Competition Commission raised concerns regarding potential lessening of competition but parties refused proposed conditions — Tribunal found no substantial lessening of competition and approved the merger, emphasizing the creation of a significant black-owned resource company as a public interest benefit.

Comprehensive Summary

Summary of Judgment

1. Introduction

These proceedings concerned an application for approval of a large merger before the Competition Tribunal of South Africa in terms of the Competition Act 89 of 1998. The matter was heard by a Tribunal panel consisting of N Manoim (Presiding Member), Y Carrim (Tribunal Member), and U Bhoola (Tribunal Member).

The merging parties were Main Street 333 (Pty) Ltd (described as the acquiring firm, also referred to as “BEE Holdco”, controlled by a newly formed entity Eyesizwe SPV) and Kumba Resources Limited (the target firm, a publicly traded South African company controlled by Anglo American). Although the transaction involved multiple steps and entities, the Tribunal treated it as a single integrated restructuring transaction commonly referred to as “Project Pangolin.”

The Competition Commission recommended approval subject to conditions aimed at preventing interlocking directorships (specifically, prohibiting Anglo American from appointing directors to the boards of Exxaro and Eyesizwe SPV). The merging parties were unwilling to accept those conditions, which led to a formal hearing on 24–25 July 2006, following a pre-hearing held on 21 June 2006. The Commission called no witnesses, while the merging parties led evidence from an economist and a senior Anglo American executive.

The general subject matter of the dispute was whether the merger, particularly through board representation and structural links among competitors, was likely to substantially prevent or lessen competition in the relevant market through co-ordinated effects (tacit or explicit coordination), notwithstanding an absence of a unilateral market power concern. The Tribunal issued an order approving the merger without conditions on 15 August 2006, with reasons delivered on 14 September 2006.

2. Material Facts

The acquiring firm, Main Street 333 (Pty) Ltd (BEE Holdco), was a special purpose vehicle directly controlled by Eyesizwe SPV, a newly formed entity whose shareholding (as described by the parties) included Anglo American (11%), an Employee Trust (10%), PWC (4%), and BHP (9%), with the remaining 66% held by Eyesizwe Mining, a black empowerment company with interests in coal. The parties described the post-merger shareholding of Eyesizwe SPV as mirroring the then-current shareholding of Eyesizwe Coal, in which Anglo American and BHP already held minority interests coupled with board representation rights.

The target firm, Kumba Resources Limited, was a publicly traded company formed in 2001 following the unbundling of Iscor’s mining division. Kumba conducted iron ore operations through Sishen Iron Ore, coal through Kumba Coal, base metals through Kumba Base Metals, and heavy minerals through Ticor. It was controlled by Anglo American, which nominated five out of fifteen directors on Kumba’s board. Anglo American also owned Anglo Coal, which competed in coal markets relevant to the assessment.

Project Pangolin entailed a series of steps that resulted in Kumba being transformed into two companies, namely Kumba Iron Ore Limited and Exxaro Resources Limited. The Tribunal accepted that, in substance, Kumba’s coal, heavy minerals and base metals operations and assets would be combined with the coal operations and assets of Eyesizwe Coal within a newly created company, Exxaro, which would be controlled by BEE Holdco holding approximately 55%. Anglo American would initially hold approximately 17% of Exxaro shares and would have one representative on Exxaro’s board.

Kumba’s iron ore assets housed in Sishen Iron Ore Company (SIOC) would be sold to a wholly owned subsidiary to be called Kumba Iron Ore, which would hold 74% of SIOC, with Exxaro retaining 20%. Employee and community vehicles would hold 3% each. Anglo American would continue to exercise control over SIOC through its 66% shareholding in Kumba Iron Ore and would also hold an indirect economic interest through Exxaro’s 20% stake.

It was common cause that the transaction created a horizontal overlap in the exploration and extraction of coal, and that coal is an internationally traded commodity, with significant South African exports and negligible imports. The Tribunal agreed with the Commission that the relevant geographic market for purposes of this merger was national, consistent with earlier Tribunal findings.

Within product market delineation, the Tribunal accepted the distinction between thermal (steam) coal and metallurgical (coking) coal, and confined its competitive assessment primarily to thermal coal, noting that Kumba alone was active in the metallurgical coal segment. The Tribunal accepted evidence that Eskom and Sasol accounted for approximately 87% of South African thermal coal consumption, with Eskom sourcing most coal through long-term contracts from so-called tied mines, and that competition for Eskom supply occurred mainly in relation to new power plants or shortfalls beyond contractual volumes.

The Commission’s primary factual concern was not unilateral market power but the risk that, post-merger, Anglo American’s board representation at Exxaro and Eyesizwe SPV, while Anglo remained the owner of Anglo Coal, would create a conduit for exchange of competitively sensitive information facilitating co-ordinated conduct among rivals in the thermal coal market. The merging parties’ central factual response was that cross-directorships existed pre-merger, that the merger did not worsen any coordination risk, and that structural changes post-merger (including changes in who met where and at what level) reduced any such risk.

The Tribunal’s analysis accepted, for purposes of assessing the merger’s effect, that there were features of the record suggesting a level of information flow and cooperation in the industry that made complacency inappropriate, but it ultimately relied on structural differences pre- and post-merger to decide whether the merger strengthened any existing coordination mechanism.

3. Legal Issues

The central legal question was whether the merger was likely to substantially prevent or lessen competition in the relevant market, as assessed under section 12A(2) of the Competition Act 89 of 1998, including the probability that firms post-merger would behave competitively or co-operatively, taking into account relevant factors such as a history of collusion.

Within that overall inquiry, the key contested issue was whether the merger raised co-ordinated effects concerns through interlocking directorships, and whether those links would materially facilitate information exchange and monitoring among competitors, thereby increasing the probability of tacit or explicit coordination.

The dispute was primarily one of the application of law to fact informed by economic theory: the parties did not materially contest the Tribunal’s power to assess co-ordinated effects, but differed on whether the structural changes created by the transaction increased the likelihood of coordination or strengthened any existing coordination, and whether the Commission’s proposed board-prohibition conditions were justified.

A further legal point was raised by the Commission in argument, namely whether conditions could be justified where a merger does not itself create new harm but perpetuates or sustains an allegedly anticompetitive structure. The Tribunal noted this as a point of law but did not decide it definitively, because it resolved the matter on the basis that the merger did not strengthen coordination and appeared to weaken it.

Public interest issues were raised in submissions by trade union representatives concerning employee participation in share schemes, but the Tribunal treated these as not giving rise to significant public interest obstacles on the record before it.

4. Court’s Reasoning

The Tribunal approached the competitive assessment by distinguishing unilateral effects from co-ordinated effects. On unilateral effects, it considered market share evidence and the structure of demand in thermal coal, including the dominant role of Eskom and Sasol and the prevalence of long-term supply arrangements. The Tribunal accepted that, even though the Commission had described the market as highly concentrated on certain measures, the Commission itself did not contend that the merged entity would have the ability to exercise market power unilaterally. The Tribunal also accepted the merging parties’ criticism that a Commission table included broader bituminous coal data (including metallurgical coal) and preferred the CRA thermal coal market share approach directed at thermal coal available to smaller customers, excluding Eskom and synthetic fuel sales. On that basis, the Tribunal found that the incremental increase in market share did not confer market power and that unilateral competitive harm was unlikely.

The Tribunal then focused on the Commission’s principal concern: whether the merger increased the likelihood of co-ordinated effects. It situated the analysis within section 12A(2), emphasising the statutory direction to assess whether post-merger firms would behave competitively or co-operatively. It also referred to international approaches describing co-ordinated effects as arising either where a merger strengthens existing coordination or where it increases the likelihood of new coordination. On the facts, the Tribunal regarded the first scenario—strengthening an existing coordination mechanism—as the relevant one.

To analyse the coordination theory of harm, the Tribunal used economic literature and comparative guidance to explain how coordination typically depends on the ability of would-be coordinators to reach agreement, monitor compliance, punish deviation, and ensure feasibility. It accepted that cross-directorships can, in theory, facilitate coordination by providing opportunities for information exchange and efficient monitoring, and it treated this as a legitimate competition concern warranting scrutiny.

However, the Tribunal’s decisive reasoning was that the post-merger structure did not strengthen, and indeed likely weakened, any coordination mechanism that might have existed pre-merger. It compared pre-merger and post-merger governance and ownership arrangements. Pre-merger, Anglo American held a minority stake in Eyesizwe and board appointment rights, controlled Kumba and nominated multiple directors, and a forum existed at the Eyesizwe board where Anglo, Ingwe (BHP via its coal subsidiary), and Eyesizwe-related directors could meet. Post-merger, Anglo’s holding in the coal assets formerly in Kumba was reduced to a minority investment in Exxaro; Anglo would have only one director at Exxaro level and one at Eyesizwe SPV level; and Ingwe would no longer be represented on Exxaro’s operational board and would instead be represented only at the shareholder investment-company level (Eyesizwe SPV).

The Tribunal treated the removal of Ingwe from the operating-company board forum as particularly significant, reasoning that successful coordination in this market would at a minimum have needed to involve such major players, and that relegating Ingwe to a more remote shareholder level reduced the immediacy and detail of information exchange and monitoring. The Tribunal also reasoned that Exxaro, unlike Eyesizwe pre-merger, would be a diversified mineral company in which coal constituted a relatively small portion of enterprise value, making coal-specific information flows on the Exxaro board less detailed and less central, and changing the incentives of participants.

In addressing the merging parties’ submission that there was no evidence of existing coordination, the Tribunal stated it was not in a position to accept that contention on the record, noting “curious” features in documentation and board materials concerning the type of operational and strategic information presented, and broader patterns of cooperation in the industry, including joint ventures and shared infrastructure arrangements (such as participation in the Richards Bay Coal Terminal). The Tribunal nevertheless chose not to decide whether coordination actually existed, because it proceeded on an assumption (favourable to the Commission) that coordination may have been possible pre-merger and held that, even on that assumption, the merger did not strengthen it.

The Tribunal also considered the transaction’s rationale and Anglo American’s reasons for seeking board representation, but cautioned against undue reliance on subjective intent in merger analysis. It nevertheless used the rationale to test whether the board representation was plausibly driven by considerations unrelated to coordination, and concluded that Anglo’s interest in board representation was explicable by investment protection and reputational concerns tied to the success of the new BEE-controlled enterprise, rather than an intention to coordinate market conduct.

Because the Tribunal found that co-ordinated effects harm was not established on the applicable test, it held that the Commission’s proposed conditions (barring Anglo board representation) were not warranted. It noted that undertakings might have enabled a more expedited process, and it recognised the Commission’s concern with interlocking directorships as legitimate, including comparative reference to jurisdictions where such interlocks are treated more strictly. Nonetheless, the Tribunal concluded that proper scrutiny did not translate into a basis for imposing the conditions sought in this case.

On public interest, the Tribunal recorded submissions from Solidarity and NUM on employee share option participation and accepted the merging parties’ confirmation that negotiations and undertakings regarding employee share schemes were ongoing and being respected, concluding that there were no significant public interest issues requiring intervention.

5. Outcome and Relief

The Competition Tribunal approved the merger without conditions, concluding that the transaction was unlikely to substantially prevent or lessen competition in the relevant market and that there were no significant public interest issues arising on the evidence before it.

The Tribunal refused to impose the conditions recommended by the Competition Commission that would have prohibited Anglo American from appointing representatives to the boards of Exxaro or Eyesizwe SPV.

No costs order is recorded in the reasons.

Cases Cited

Anglo American Holdings and Kumba Resources Limited with the Industrial Development Corporation intervening, Competition Tribunal of South Africa, Case No. 46/LM/Jun02; Anglo South Africa Capital (Pty) Ltd, Eyesizwe Coal (Pty) Ltd, Mafube Coal Mining (Pty) Ltd and Arnot North Mining Business, Competition Tribunal of South Africa, Case No. 44/LM/May05; BHP Steel Southern Africa, BHP Minerals International Exploration Inc, BHP World Exploration Inc and Billiton SA Limited and Mine & Smelter Investments (Pty) Ltd, Competition Tribunal of South Africa, Case No. 32/LM/Jun01; Sasol Limited and Others and Engen Limited and Others, Competition Tribunal of South Africa, Case No. 101/LM/Dec04; Airtours plc v Commission of the European Communities (Case T-342/99) [2002] All ER (EC) 783; Independent Music Publishers and Labels Association (Impala) v European Commission (Case T-464/04), judgment of 13 July 2006; Federal Trade Commission v H.J. Heinz Co, 246 F.3d 708; Hospital Corporation of America v Federal Trade Commission, 807 F.2d 1381 (United States Court of Appeals for the Seventh Circuit, 1986); United States v Sears, Roebuck & Co., 111 F. Supp. 614 (United States District Court for the Southern District of New York, 1953).

Legislation Cited

Competition Act 89 of 1998, including section 12A(2) and section 4(2); Clayton Act of the United States, section 8.

Rules of Court Cited

No rules of court are cited in the reasons.

Held

The Tribunal held that the merger, properly characterised as a restructuring that created Exxaro Resources Limited as a BEE-controlled diversified resource company and separated out Kumba’s iron ore assets into Kumba Iron Ore, was unlikely to substantially prevent or lessen competition in the national thermal coal market.

The Tribunal held that the merger did not raise unilateral effects concerns of market power, and that the Commission’s co-ordinated effects theory—based on the risk of information exchange through Anglo American’s interlocking directorships—was not established on the relevant test because the merger did not strengthen any existing coordination mechanism and appeared, in structural terms, to weaken it.

The Tribunal further held that there were no significant public interest concerns requiring conditions, and it accordingly approved the merger without conditions.

LEGAL PRINCIPLES

Section 12A(2) of the Competition Act 89 of 1998 requires merger authorities to assess the probability that firms in the market after the merger will behave competitively or co-operatively, taking into account relevant competitive factors, including any history of collusion.

Merger analysis distinguishes between unilateral effects (market power exercised by the merged entity acting alone) and co-ordinated effects (enhanced scope for coordination among rivals, whether explicit or tacit). A co-ordinated effects assessment focuses on whether a merger materially increases the likelihood of successful coordination or strengthens existing coordination, evaluated as a probabilistic assessment on the evidence.

Structural links, including cross-directorships and related governance arrangements, may in theory facilitate coordination by enabling exchange of commercially sensitive information and by improving the ability to monitor adherence to coordinated strategies. Whether such links warrant merger conditions depends on whether the merger materially changes market structure and incentives in a way that increases coordination risk, rather than merely the presence of such links in the abstract.

Where coordination concerns are advanced, structural comparisons between pre-merger and post-merger governance and incentives may be central to evaluating whether a merger strengthens a coordination mechanism. If the post-merger structure makes information exchange and monitoring less immediate or less likely (for example by removing key competitors from operational-level forums, altering board composition, or changing economic incentives), the merger may be found not to strengthen coordination even where coordination is assumed to be possible pre-merger.

The Tribunal indicated that merger assessment is generally objective and does not ordinarily depend on parties’ stated subjective rationale, but that rationale may be used as a contextual check when evaluating whether a contested feature of a transaction (such as board representation) plausibly serves non-collusive commercial purposes on the record.

Main Street 333 (Pty) Ltd and Kumba Resources Limited (14/LM/Feb06) [2006] ZACT 77; [2006] 2 CPLR 601 (CT) (14 September 2006)