South African Airways (SAA) entered into override incentive agreements and trust payment agreements with various travel agents between 1 June 2001 and 31 March 2005. These agreements provided travel agents with financial incentives based on achieving sales targets calculated on a 'back to rand one' principle for override commissions. Comair and Nationwide Airlines complained that these agreements constituted prohibited exclusionary practices. The Competition Tribunal had previously found against SAA in earlier proceedings (the Nationwide matter) for substantially similar conduct occurring between October 1999 and 31 May 2001. SAA changed its agreement structure in 2001 (introducing '3rd generation' contracts and trust payments) and again in 2005/2006 (4th generation contracts) following the Tribunal's adverse findings. The agreements in question provided for: (1) flat override payments for achieving base revenues calculated 'back to rand one'; (2) differentiated override payments depending on ticket class; (3) trust payments (lump sum payments for achieving revenue and market share targets); and (4) targets based on flown revenue rather than BSP figures, which only airlines could calculate.
Appeal dismissed with costs, including costs of two counsel. The Competition Tribunal's order was upheld, declaring that: (1) SAA's override incentive agreements with travel agents from 1 June 2001 to 31 March 2005 constituted prohibited practices in contravention of section 8(d)(i) of the Competition Act; and (2) SAA's trust agreements/payments with travel agents from 1 June 2001 to 31 March 2005 constituted prohibited practices in contravention of section 8(d)(i) of the Competition Act.
A dominant firm contravenes section 8(d)(i) of the Competition Act when it implements incentive agreements with intermediaries that: (1) provide financial inducements for the intermediaries to prefer the dominant firm's products over those of competitors; and (2) have a substantial or significant effect in foreclosing the market to rivals, even if rivals are not completely excluded. Such foreclosure can be established either through evidence of actual competitive harm or through evidence that the exclusionary practice has the potential to foreclose competition. The form and design of the incentive structure is relevant evidence of exclusionary effect, particularly where it includes: 'back to rand one' retroactive commissions; individualized targets tailored to each intermediary; rewards based on overall volume rather than incremental sales; and targets calculated on metrics only known to the dominant firm. A dominant firm cannot avoid section 8(d)(i) liability by arguing that intermediaries have only limited ability to influence customer choices; it is sufficient that they have some meaningful ability to do so. For section 67(2) purposes, conduct occurring in a different time period, even if substantially similar in nature, constitutes 'new conduct' that is not protected by the double jeopardy provision.
The court made important observations about procedural reform needed in competition proceedings. It noted that the repetitive nature of voluminous records, coupled with much irrelevant evidence (particularly expert economists opining on legal interpretation issues which are properly matters for the court), compels reform. The court suggested: (1) Greater use of the 'hot tub' method for expert evidence, whereby experts meet to determine common ground and identify genuine disputes, with proceedings confined to disputed issues. This would allow relaxation of the adversarial system and enable the Tribunal to better assess differences in expert testimony. (2) More effective use of pre-trial conferences to define issues to be determined, avoiding expenditure of time and resources on matters not genuinely in dispute (such as market power where market share exceeds the statutory threshold). (3) Greater intervention to ensure crisper and more nuanced definition of issues requiring determination. The court also observed that while the European Court jurisprudence on abuse of dominance reflects a somewhat different legal framework (Articles 101 and 102 of the European Treaty), the approach is not dissimilar to that mandated by the South African Competition Act, particularly given the Act's objectives in section 2 to protect the competitive process rather than merely competitors.
This case is significant for several reasons: (1) It clarifies the scope of section 67(2) protection against double jeopardy, establishing that the prohibition applies only to substantially the same conduct in the same time period, not to similar conduct occurring at different times or involving modified agreements. (2) It confirms that section 8(d)(i) does not require proof of actual consumer harm; it is sufficient to show that the exclusionary practice is substantial or significant or has the potential to foreclose the market to competition. (3) It establishes principles for assessing exclusionary incentive schemes by dominant firms, particularly the ability of such schemes to foreclose rivals even where the dominant firm's competitors are not completely excluded from the market. (4) It recognizes the importance of 'directional selling' by intermediaries (travel agents) and the ability of dominant firms to leverage this through tailored incentive schemes. (5) It provides guidance on market definition in multi-sided markets involving intermediaries. (6) The judgment contains important obiter dicta regarding the need for procedural reform in competition proceedings, including the use of 'hot tubbing' for expert evidence and more effective pre-trial case management to control the length and cost of proceedings.