---

COMPETITION TRIBUNAL OF SOUTH AFRICA

Case no.: 101/LM/Oct05

In the Large Merger between:

Murray & Roberts Limited Acquiring Firm

And

Concor Limited Target Firm

Panel: D Lewis (Presiding Member), Y Carrim (Tribunal Member) and

N Manoim (Tribunal Member)

Date of Hearing: 14 June 2006

Order issued on: 14 June 2006

Reasons issued on: 11 October 2006

Reasons for Decision

APPROVAL

1. On 14 June 2006, the Competition Tribunal issued a Merger Clearance Certificate approving the transaction between Murray & Roberts Limited and Concor Limited. The reasons for this decision follow.

THE TRANSACTION

1. In terms of the proposed transaction, Murray & Roberts Limited (“M&R”) will acquire the entire issued share capital of Concor Limited (“Concor”) from its shareholders.¹ The transaction will be effected either by a scheme of arrangement or a substitute offer.

2. M&R is the main operating subsidiary in the group of companies controlled by Murray & Roberts Holdings Limited (“M&R Holdings”), a public company listed on the JSE. The subsidiaries of M&R Holdings include the following firms: Murray & Roberts Construction, Murray & Roberts Engineering Solutions, Murray & Roberts Marine, Murray & Roberts Steel, Murray & Roberts Concessions and Murray & Roberts Cementation.

3. Concor is listed on the JSE and is controlled by Hochtief AG (“Hochtief”), a German company. Concor has the following operating divisions: Concor Building Division, Concor Civils, Concor Engineering, Underground Mining, Opencast Mining, Concor Roads, Concor Technicrete andConcor Facility Management.

4. M&R believes that the acquisition will inter alia be value enhancing for it and will give it access to capabilities that M&R do not currently have, namely the Technicrete and open cast mining division of Concor. For Concor, its controlling shareholder, Hochtief wishes to exit its investment in Concor for strategic business reasons. ²

THE MERGING PARTIES’ ACTIVITIES

5. Through its subsidiaries, M&R Holdings is involved in construction, engineering, construction materials and services and fabrication and manufacturing. Concor is involved in civil engineering, building, structural and mechanical engineering, roads construction, opencast mining contracting, underground mining contracting, manufacturing of building and mining products and facilities management.

COMMISSION’S FINDINGS

6. The proposed transaction involves a number of product markets both at a vertical and horizontal level. The investigation of the transaction conducted by the Commission was lengthy and comprehensive.

7. The Commission and the parties are in agreement that overlaps occur in the following areas of activity: Building/Construction, Civil Engineering, Engineering (general), Roads, Facilities Management/Concessions and Underground Mining.³

8. There is also a degree of vertical integration that occurs in that both parties supply products used in the downstream markets for building/construction, roads construction and civil engineering.

Vertical markets

9. The Commission identified several upstream markets in which the merging parties are active. However, an examination of the market shares of the merging parties in all the markets reveals that there are alternative suppliers and that foreclosure is not likely to result from the merger in any of the identified markets.⁴ Since these vertical markets raise no competition concerns we do not deal with these any further in these reasons.

Horizontal markets

10. The Commission identified and analysed the following product markets:

1. The market for building construction (focusing on the large project sub market);

2. The market for the provision of civil engineering services (focusing on the large project sub market);

3. The market for the provision of engineering (general) services;

4. The market for the provision of roads construction;

5. The market for the provision of facilities management; and The market for the provision of underground mining services.

11. The market shares for the merged entity and changes in the HHI calculations in the markets for the provision of engineering (general) services, underground mining, and facilities management were notably high thereby raising prima facie competition concerns. ⁵

12. In the course of the Commission’s investigation, an objection was received by the South African National Roads Agency Limited (“SANRAL”) in relation to the market for the provision of facilities management.

13. SANRAL is mandated under the National Roads Act 1998 to develop, maintain and manage South Africa's national road network of some 16000kms.⁶ SANRAL must also strategically plan, design, construct, operate, rehabilitate and maintain the roads.⁷ As part of its mandate, SANRAL is responsible for the awarding of contracts for toll road concessions and toll road operations.⁸ SANRAL’s submissions were limited to the market for toll road operations.

14. The Commission, after considering SANRAL’s objection, concluded that no competition concerns arose from any of the markets identified above and accordingly recommended that the transaction be approved unconditionally.

15. A hearing was held on 14 June 2006. The following witnesses testified at the hearing:

1. Mr. Nazir Ali;⁹

2. Mr. Wynand Dreyer,¹⁰

3. Mr. Mario Gericke,¹¹

4. Mr. Trevor Robinson,¹²

5. Mr. Henry Laas.¹³

16. Apart from SANRAL’s objection in relation to the market for the provision of facilities management, no objections or concerns from customers were received in relation to any of the other markets. The Tribunal had a residual concern about the market for underground mining, which was adequately addressed in the hearing. (See note below)

17. Accordingly, the only market in respect of which any objections were raised was the market for facilities management. Most of the hearing and witness testimony was focused on this market.

18. After hearing the witnesses and considering the evidence placed before the Tribunal approved the merger unconditionally and issued a certificate on 14 June 2006. These are the reasons for that order.

19. In these reasons, we deal only with the market for toll road operations (as a sub-market of facilities management). We do not deal with any of the other relevant markets identified by the Commission in its investigation because in our view the competition evaluation of the toll road operations market applies equally to any of the other relevant markets identified by the Commission.

RELEVANT GEOGRAPHIC MARKET

20. Both the Commission and the parties were in agreement that the relevant geographic market in respect of all the above markets is national. We will accept, for the purpose of analysing this transaction, that the relevant geographic market in respect of all the above markets is national.¹⁴

RELEVANT PRODUCT MARKET

The Market for Facilities Management

21. Facilities management involves the provision of management, administration and technical services with regard to property belonging to third parties. The merging parties’ activities in this market overlap only in respect of toll roads.

22. The parties and the Commission are in agreement that that this market can be divided into two sub markets viz. Toll road concessions and Toll road operations.

Toll Roads Concessions

23. A toll road concession is a contract that is awarded to a consortium, for the construction, operation and maintenance of a toll road. Firms who bid for such concessions usually enter into consortia or joint ventures with other private parties, commonly constituted by contractors, designers, engineers and financial institutions. These third parties undertake to fund the construction, operation and maintenance of a toll road. In return for their investment the successful consortia are entitled to collect tolls for a specified period. Hence, they usually conduct toll road operations as part of the concession contract. The toll road concession includes a number of relevant markets such as road construction, engineering services and toll road operations.¹⁵

Toll Road Operations

24. Toll road operations generally involve the collection of tolls and the management and maintenance of toll gates and roads.

25. The first toll road operations in South Africa commenced in 1983. Until 2001, government sponsored toll road operations were conducted under operations contracts in which most technical functions for a particular operation were carried out on a selected subcontract basis. For example there would be subcontractors for each of the aspects of the operations such as lane equipment installation, management of information systems and, electro-mechanical work. This led to a plethora of operators or service providers who had to be supervised and managed individually. Over time, it became increasingly difficult to apportion responsibility for any revenue loss amongst several subcontractors and to implement or improve supervision and controls to remedy fraud or negligence. SANRAL embarked on a remodelling of these contracts by transferring the risk and compliancy requirements to a single operator.

26. The comprehensive toll road operations and maintenance contract (CTROM) model was implemented by SANRAL in order to better manage delivery, performance and risk in relation to toll road operation by awarding a comprehensive contract to one operator, who would be legally liable for compliance with all the terms and conditions of that comprehensive contract. Routine road maintenance services were included in these contracts in order to attract the medium and large contractors.¹⁶ The operator could subcontract with any number of other suppliers provided that it did so in terms of the specifications prescribed by SANRAL.¹⁷ These contracts are awarded through a comprehensive tender process of relatively long duration, usually five years.

27. In parallel with this development was the continuation of the toll road concession market in which SANRAL awarded a concession to a private sponsor, usually a special purpose vehicle, to construct, operate and maintain a toll road. As stated above these special purpose companies also conducted toll road operations as part of their concession agreements.

28. A toll road operator may render toll road management services to either the concessionaire who is entitled to recoup its investment through the collection of toll fees or SANRAL, who has issued a tender for the operation and maintenance of toll roads that it has constructed. The skills, technological expertise and experience required for the operations and maintenance of toll roads would be the same whether the construction of the road is done by privately sponsored concessionaires or whether it has been constructed by SANRAL.

29. Hence, according to the merging parties the market for toll road operations would include all those operators who render toll road management services to whosoever requires it, whether this is the concessionaire or SANRAL. The parties provided the following table of estimated shares for the toll road operations market:

Firm	% Market share
Tolcon (M&R)	24.1
Intertoll	41.0
Tracc / Basil Read	17.0
Concor	5.3
Dragados	12.6
Total	100

30. Post merger, the merged entity will enjoy 29.4% of this market.

SANRAL’s submissions

31. In its submissions to the Commission, SANRAL initially argues for a further segmentation of the market for toll road operations. It submits that the market should be divided into (i) the market for privately-sponsored toll road operations and (ii) the market for government-sponsored toll road operations roads. ¹⁸

32. According to SANRAL, privately-sponsored toll road operations involve the construction of the toll road itself. Government-sponsored toll road operations, do not involve the construction of the road itself and unlike in the privately-sponsored toll road operations market, these companies would manage toll roads and collect toll fees on behalf of government and charge a fee for the services rendered. This seems to be the first basis of market segmentation put forward by SANRAL.

33. As a second basis for market segmentation SANRAL submits that the market for government-sponsored toll road operations is “not an easy market and is in fact very competitive from a pricing point of view and has a different composition of efficiency gains than toll road operations of a privately-sponsored road which is operated within the confines of a concession contract.” ¹⁹

34. What SANRAL seems to be saying is that the basis for distinction was the size of the margin that the operator made. Firms made a lower margin in the CTROM (government-sponsored) market than those in the concession (privately-sponsored) market. Therefore, the market could be segmented on the basis of low and high margins and only three firms operated in the government-sponsored toll road operations market, viz. Tolcon (M&R), Concor and Intertoll (Group 5).²⁰ This merger would therefore result in the current toll road operators being reduced from three to two and this would increase the likelihood of co-ordination in the market. In addition, bidders may be less competitive because of the lack of threat of a third bidder. According to SANRAL, if the lowest tender were disqualified for some reason, SANRAL would be compelled to accept the higher tender. SANRAL believed that the reduced competitive environment would effectively defeat the benefits of the tender process.

35. Furthermore, argued SANRAL, barriers to entry were high which meant that very few players would enter the market in the foreseeable future. It submitted that the toll road operations market is a highly specialised technical and financial field with high-risk exposure and low profit margins, and substantial capital facilities were required. Hence, it argued that the proposed transaction would substantially lessen competition and requested the Tribunal to prohibit or approve it on conditions that would address the competition concerns raised by SANRAL.

36. The Commission’s view was that its assessment of the market would remain the same whether or not there was any further segmentation because these were bidding markets. In addition, it argued that barriers to entry in the toll road operations market were surmountable and there were sufficient potential entrants to alleviate any competition concerns that may arise from this merger.

37. The parties, in response to SANRAL’s submission, agreed that a distinction could be made from a commercial perspective between privately-sponsored toll road operations and government-sponsored toll road operations, in that –

1. The party actually performing the toll road operations in privately-sponsored toll road operations generally formed part of the consortium that was granted the toll road concession. As privately-sponsored toll road operations were funded by consortium members and not government, this meant that the toll road operator would generally require significant financial resources in order to perform its funding obligations to (and participate in) the consortium.

2. With government-sponsored toll road operations, the tolls were paid directly to SANRAL and the toll road operator merely provided toll road operations for a fee. Toll road operators were not required to have the same financial resources as required for privately-sponsored toll road operations.

38. Both the merging parties and SANRAL agree that there is a difference between the two operations in the achievable margins, but while SANRAL argues that this should be a basis for market segmentation, the merging parties argue the contrary. The merging parties argue further that from a competition law perspective, the relevant market segment remained that of providing toll road operations and to distinguish between separate markets for privately-sponsored toll road operations and government-sponsored toll road operations would mean distinguishing a relevant market merely on the basis of the identity of the “customer” (SANRAL in government-sponsored operations and the private concession company in privately-sponsored operations). In any event, they submitted, the nature of toll road concessions was such that SANRAL (as the regulator tasked to act on behalf of road users) was ultimately the “customer” for both types of toll road operations.

39. Evidence led by the merging parties and SANRAL’s CEO, Mr Nazir Ali indicated that in CTROMs, (as opposed to privately-sponsored toll road operations), the margins are smaller due to the highly regulated nature of the contracts. Opportunities for improving margins were limited and were dependent to a large extent on how effectively operators could manage their costs. These opportunities were further limited by the fact that the technical specifications and contractual terms of these contracts were prescribed.²¹ The more stringent these specifications and conditions, the more costly the operations. The contracts often required the purchase of costly equipment and software. The size of the penalty imposed by SANRAL in the event of fraud or non-performance also impacted on margins in the CTROM contracts. ²²

40. The margins in privately-sponsored toll road operations, on the other hand, were higher because the operations were usually conducted within the concession contract, which included a range of other commercial activities. This enabled the operators to spread their risks and costs across the concession activities. Greater efficiencies and higher margins could be achieved by better planning, lower construction costs and discounts.²³

41. We accept that toll road operators, as part of concession agreements, may make a higher margin than those in CTROM contracts and could possibly require fewer financial resources. However, we would be hesitant to accept differences in margins as a sole basis for market segmentation.²⁴ By and large, margin achievements are dependent on far too many variables. ²⁵

42. In support of its contention that there were only three players in the toll road operations market, Mr Ali testified that only the three companies identified had actually bid for CTROM contracts and had been awarded these. ²⁶ These are listed as Tolcon (M&R), Concor and Intertoll (Group 5). It argues that the Basil Read/Bouygues (BRB) joint venture, which had been awarded the N17 CTROM contract, was not a participant in the market because it was a special purpose vehicle formed for the purposes of the N4 concession and that it had to obtain special permission from its shareholders to tender for the N17 CTROM.

43. In addition, in order to define the relevant market, it seems that SANRAL has placed some reliance on the fact that BRB has indicated that it does not intend to bid for another CTROM contract, after its contract for the N17 expires in 2008, due to the “imbalance between risk and reward”. The Commission seems to suggest that BRB’s involvement in the market is exceptional because participants that are in the toll road concession business generally do not participate in the CTROM business. ²⁷

44. While BRB may have voiced its hesitation to continue in the CTROM market, this is clearly a statement of intent, which may or may not occur sometime in the future. At this point in time, BRB is in both the CTROM and concession markets and therefore must be included as a participant.²⁸ The fact that contractual restrictions of this type are in place voluntarily, rather than imposed by regulation, implies that a competitor could negotiate out of these restrictions and bid for tenders in both the concessions and CTROM sub-markets, as has happened in the BRB case. Furthermore, the suggestion that BRB is somehow an exception seems misplaced given that the merging parties themselves participate in both the CTROM and concession space.²⁹

45. SANRAL’s own tender documentation reveals that several firms were invited to submit tenders for certain toll road projects.³⁰ These documents also reveal that the bidders competed on the basis of price rather than margin.³¹ Mr Ali himself asserts that the market for toll road operations is “very competitive from a pricing point of view”.³²

46. In any event, it seems that nothing much turns on these distinctions since in a subsequent submission SANRAL concedes that the relevant market is toll road operations. In addition, Mr Ali conceded, during his testimony, that there was no difference between the two types of operations in relation to the technical knowledge and level of skills required and in the functions of the two operations³³ and that the companies who conducted the toll road operations, as part of a concession agreement would be able to conduct toll road operations for a government sponsored toll road operation.³⁴ Furthermore as demonstrated in the N4 East extension tender, there could be a continuum between the two types of toll road operations.³⁵

47. For these reasons we agree with the merging parties that while a commercial distinction could be drawn in relation to the extent of financial resources required or margins achieved, these differences were not sufficient in themselves to further segment the toll road operations market into a concession market and a CTROM market for purposes of competition evaluation.

48. Nor can this Tribunal rely on the type of customer (concessionaire versus SANRAL) as a basis for further segmentation of the relevant market in this particular transaction. The type of customer serviced (e.g. business versus residential) has been used by competition authorities in other mergers or jurisdictions as one of many indicia to segment markets into sub-markets.³⁶ However, in this merger, no such distinction could be made because the ultimate customer is SANRAL, who determines the terms and conditions of both the concession and the CTROM tenders for national toll roads.

49. Accordingly, we conclude that the relevant market for purposes of this transaction is the provision of toll road operations.

IMPACT ON COMPETITION

50. In each of the markets identified in Annexure A, the post merger market shares seem notably high and prima facie raise competition concerns. In the market for toll road operations the market shares for the merged entity will be 29.4%, with an HHI of 2993 suggesting a fairly concentrated market.³⁷ An HHI above 1800 is generally considered to be an indication of a highly concentrated market. The change in HHI of 255 points is significant enough to cause concern and attract closer anti-trust scrutiny.

51. The Commission and merging parties argue that despite the seemingly high market share percentages and the change in HHI of 255 points for this market, these figures are not necessarily indicative or an accurate reflection of market power since market shares may fluctuate significantly as a result of the grant of one or more large project/s to a particular market participant.³⁸ All of the activities of the merging parties involve large projects and lumpy capital investments, which arose infrequently through a bidding process. The Commission and the parties argued that the transaction was unlikely to raise competition concerns, whether in the form of co-ordinated effects or unilateral conduct, as suggested by SANRAL since a “strong and peculiar feature of the relevant markets is that each of them are so-called bidding markets.” ³⁹ Furthermore, they argue, SANRAL has countervailing power to offset any exercise of market power on the part of a supplier.

52. In our evaluation of the impact on competition below, we proceed on the basis that the relevant market is the broader market for toll road operations, rather than the narrower markets suggested by SANRAL. Hence, in our view, this is not a 3-2 merger and the market would include at least the five players identified in the table in paragraph 30, which have been awarded a toll road operations contract, whether this be a CTROM or concession contract.⁴⁰ In accordance with this Tribunal’s traditional approach to mergers⁴¹ we conclude that this transaction is unlikely to lead to a substantial lessening of competition and that the remaining players, post merger, would be able to constrain any exercise of market power by the merged entity. But in the event that we are wrong on our definition of the relevant markets, we are persuaded that this market has the features of a traditional bidding market as discussed below, and that this would further ameliorate any competition concerns identified by SANRAL.

Bidding markets

53. The Commission’s reliance on the fact that any competition concerns that may attend this merger will be ameliorated by the fact that these are bidding markets stems from the understanding that the manner in which project specifications are developed and in which tenders are adjudicated in a particular market limit the ability of a supplier to exercise market power.⁴²

54. In traditional bidding markets,⁴³ the bidding process would usually involve a pre-qualifying process, with a fair amount of interaction between the suppliers and the client. The tender is usually at a very high level of detail. The pre-qualifying process requires all prospective bidders to qualify with pre-determined criteria. Prospective bidders, who qualify at this stage, would compete only on price at the last stage of the bidding process and the winner would take it all. Once a contract is awarded, the consequence is a relative increase in market share for the winner. This high market share endures for as long as the contract endures. Customers in these markets under consideration are large sophisticated clients who set the specifications and terms of the tender.⁴⁴ The investment required is usually of a large capital nature, with tenders occurring infrequently and contracts enduring for a relatively long period of time. ⁴⁵

55. Klemperer⁴⁶ identifies the features of an ideal (classical) bidding market, relying on the European Commission’s definitions. The term bidding market is associated with contests where:

1. Competition is ‘winner take all’, so each supplier either wins all or none of the order;

2. Competition is ‘lumpy’;

3. Competition begins afresh for each contract, and for each customer;’

4. Entry of new suppliers into the market is easy; and

5. A ‘bidding system’ or ‘bidding process’ is involved.

56. However Klemperer cautions regulators against suspending antitrust scrutiny of a transaction simply because it is a “bidding market”. He argues that many markets that are claimed as classical bidding markets (and which therefore would provide protection against the exercise of market power) are in fact not such markets. Therefore, one must carefully consider the extent to which the special features of auctions and bidding processes in a particular transaction mean that competition evaluation should indeed be different than for ordinary economic markets.

57. According to Klemperer if a market displayed all of 1-3 (“Bertrand market”) or 1-4 (“contestable market”) features identified above as a bidding market then it would make sense for competition authorities to tolerate (assuming 5) the creation of maintenance of highly concentrated markets.⁴⁷ He goes on to argue why the same range of competitive problems may beset bidding markets and ‘ordinary’ economic markets. He argues that with the rise of e-commerce, government privatisations and both public and private outsourcing the role of auctions has greatly increased in the economy. As bidding markets move more and more away from the Bertrand or contestable type of markets, regulators need to be more circumspect about markets described as “bidding markets”. Hence regulators need to examine the features of each so-called bidding market, the product and the bidding process itself in its evaluation, as it does in other ‘ordinary’ markets. This was precisely the approach adopted by the Tribunal in the Cementation case.⁴⁸

58. This Tribunal has previously cautioned that all markets claimed as bidding markets are not necessarily capable of limiting the ability of participants to exercise market power. In Murray & Roberts Limited and The Cementation Company (Africa) Limited the Tribunal stated that “there appears to be no particular reason why the existence of bidding markets should prevent the exercise of market power in the market for providing, for example, protective clothing or explosives or some or other input that is required […] on a daily basis.” ⁴⁹

59. This is because there may be a number of reasons why confidential bidding was preferable in markets in which working equipment or other inputs are supplied on a relatively small scale and a regular basis. Examples of such bidding processes can usually be found where a public entity or a private corporation is the purchaser of motor vehicles or cleaning and maintenance services or IT and other services. The reasons why bidding processes are preferred may be numerous varying from the need to promote fairness and transparency to promoting competition between bidders to obtaining the best possible price. Such tenders and bidding processes occur regularly and are relatively small and do not provide a greater degree of protection form market power.

60. Put simply, there is no magic in the words “bidding markets”. Just because merging parties invoke the word “bidding markets” does not mean that the Tribunal must suspend its antitrust scrutiny of the transaction before it. Not all bidding markets provide protection against the exercise of market power simply by virtue of being bidding markets. However, a merger in bidding markets that display the characteristics of a Bertrand type market is less likely to result in a lessening of competition in the form of either unilateral or co-ordinated effects.

61. The markets identified by the Commission in this merger are not markets for some or other input that is required by the customers for the merging parties on a regular basis. These markets are markets where the product or service that is the subject of the bidding is a large, lumpy capital investment project, taking place over some duration of time and which occurs infrequently.

62. We turn to consider the characteristics of the toll road operations market.

63. In general, the bidding process for toll road operations takes place in the following manner. SANRAL in its discretion decides to issue a tender for either a toll road concession or a CTROM. It does so after some investigation and approval of its Contracts Committee. Once such a decision is made, seemingly with a fair amount of interaction with key players in the industry, SANRAL prescribes the specifications of the tender and embarks, with the assistance of consultants, on a pre-qualifying process whereby it invites potential bidders to submit expressions of interest. Many companies, both national and international, who are known by these consultants as potential players in the market are invited to participate in the process.⁵⁰ A proportion of those approached are likely to submit expressions of interest.⁵¹ A list of these is drawn and on the basis of some pre-determined criteria by SANRAL,⁵² usually related to experience and credibility, a shorter list of firms is drawn up. These short-listed bidders are then asked to submit formal bids. It is at this stage of the process that companies may decide or decline to submit a formal bid.

64. The tender is usually designed along SANRAL’s specifications. However, a fair exchange of information and interaction regarding the tender specifications and possible business models could take place during the pre-qualifying stage.⁵³ After bids have been submitted a bidder is chosen on the basis of price and the winner takes it all. The duration of the contracts are usually five years (CTROM) but could be extended.⁵⁴ Tenders are put out afresh at the end of the contract period. The investment required is usually in the region of R100million or more.

65. This process demonstrates that SANRAL goes into the market, for a large capital investment with an approximate idea of the price, quality, and duration of contract and technical specifications of its purchase. It approaches prospective suppliers after a process of technical and financial modelling and international benchmarking conducted by its consultants⁵⁵ and sets a very high level of detail in the bid. While SANRAL may not have specific knowledge about the costs of equipment,⁵⁶ it has sufficient knowledge about the technical specifications and the potential costs of the operators, which enables it to estimate the margins and the areas in which bidders may compete.⁵⁷ This high level of detailed knowledge about its projects makes it very difficult for bidders to construct a collusive bid.

66. In addition, the tender process, which is very tightly controlled by SANRAL, isolates suppliers and makes it difficult for them to collude. During the pre-qualifying stage, SANRAL approaches a group of pre-selected bidders, who are then short-listed by SANRAL. Some of this group are rejected because they are not suitable or credible applicants, according to criteria pre-determined by SANRAL. Those who remain and whom SANRAL invites to submit a formal bid, may do so or decline to do so. Until that point in time none of the interested prospective bidders know with any certainty whether they will be on the shortlist or not, nor are they able to influence the short listing which is done according to pre-determined criteria set by SANRAL.⁵⁸ Obviously, the more experienced players will be able to predict with a measure of accuracy due to their longer involvement in and knowledge of the sector, whether they are likely to be short-listed, but unlike in a beauty contest type application⁵⁹ in which public hearings on bids are often mandatory, each application is confidential and bidders do not have insight into the other bidders applications.⁶⁰

67. All the bidders at this last stage are credible bidders and are required to comply with the technical specifications prescribed by SANRAL, the only basis of competition being price. ⁶¹ SANRAL is even able to predict with some measure of accuracy, the extent of the price competition because the Minister prescribes toll fees and bidders tend to compete largely in the area of cost management.⁶² The selection of the successful bidder is made by SANRAL in confidence and the winner is awarded the full contract, rather than only some aspects of it.⁶³ Given this kind of process, it seems extremely unlikely that any of the bidders, assuming some collusion on their part, would be able to assess whether any of them were cheating or were deviating from the terms of their collusive agreement.

68. Mr Ali suggests that if there were only two players⁶⁴ in the market this may lead to collusion on allocation of markets, trading conditions or increased prices and possibly predatory pricing.⁶⁵ However the fact that these tenders are infrequent, concern very large projects and last over a fairly long duration, does not in our view provide an incentive for such collusion. By its own admission, the market for toll road concessions has grown at a much slower pace than that anticipated by SANRAL. Tenders are few and far between than was initially anticipated. There is also a measure of unpredictability in the awarding and duration of these tenders. At times a tender has started out as a CTROM and been converted to a concession⁶⁶ and at times the duration of the contract has been extended by an additional short period. ⁶⁷

69. A bidder is unlikely to stay out of one bid in favour of its partner-in-collusion when there is some uncertainty as to when its turn will come up next, or if at all, or about the extent of foregone profits in the case of a possible conversion to a more lucrative contract (CTROM to concession). Likewise, a bidder is unlikely to engage in predatory pricing if it is uncertain about when, or if at all, it will be able to recoup any losses incurred due to such strategies.

70. Nor is there any evidence that a predator in the CTROM business is likely to recoup its losses through the concession business. All the bidders in the toll road concession business tend to be large consortiums, consisting of several shareholders. Unless the toll road operator in the CTROM business is a shareholder of a concessionaire, he will have to bid against others for the business of the concessionaire, the basis of competition being similar to that of the CTROM contracts. He would also have to recoup his investments (and any losses incurred as a result of predatory pricing in the CTROM business) over a much longer period of time.

71. While the merged entity will have a fairly large market share (estimated as 29.4%), the remaining players in the market are highly credible and experienced competitors. It would seem unlikely that they would allow the merged entity to gain market share through a series of anti-competitive conduct.⁶⁸

72. These combinations of factors persuade us that this market is competitive and is likely to continue being competitive, despite SANRAL’s concerns about collusion.

Countervailing Power

73. While the Tribunal has not always found favour with the argument of countervailing power, in this particular merger, SANRAL, is the ultimate customer in toll road operations and in road construction, a large customer in engineering (general) services and is clearly a sophisticated customer, who is knowledgeable about international benchmarking, pricing and margins. In our view SANRAL is a price maker in the relevant market⁶⁹ since as a regulator and customer it is able to determine not just the technical specifications, quality of service, duration and commercial terms of a tender, but is also able to determine the size of the market. The markets for toll road concessions and operations are completely dependent on whether SANRAL decides to issue a tender or not.⁷⁰ In our view this countervailing power on the part of SANRAL acts as a disincentive to suppliers to exercise market power.

Barriers to Entry

74. SANRAL listed several key barriers to entry which had arisen and prevented further entrants from entry and participation in this market notably that it was a highly specialised technical and financial field with high risk exposure, low profit margins and substantial capital facilities required.⁷¹ The Commission however found that in the market for toll road operations entry barriers were in fact surmountable.

75. The Commission submits that SANRAL itself contributes to the raising of entry barriers by its stringent regulatory or contractual specifications. SANRAL’s requirements were so stringent that only experienced and established players could meet them. An example of this is the shortages prescribed by SANRAL in CTROM contracts. The internationally accepted standard for shortages i.e. the extent of allowable shortages in toll fees collected before penalties are applicable is circa 5%.⁷² SANRAL on the other hand prescribes a shortage of 0%. This means that if an operator incurs any shortages whatsoever in the toll fees to be collected, that operator is liable to pay a penalty. According to market participants, this is overkill on the part of SANRAL since the additional costs associated with the collection of fees do not justify the extra income generated by this and serves as a discouraging factor for potential entrants. ⁷³

76. As a second basis in support of its contention, the Commission asserts that supply side substitution from large construction companies was possible. International suppliers of toll road equipment were also potential entrants in the toll road operations market. In support of this, the Commission points to the BRB presence in the toll road operations,⁷⁴ as well as to Toll Link which has recently formed a consortium with Intertoll for a CTROM tender.⁷⁵

77. In our view the issue of a shortage percentage may be relevant to the degree of regulatory barriers, but the fact that the market is a regulated one and requires a high level of technical and specialised knowledge suggests that the barriers to entry in the relevant market are high for new players. The barriers to entry do not seem as easily surmountable as that argued by the Commission. Nor does it appear that supply side substitution is as easy as that suggested in the Commission's report.

78. In other words, the barriers to entry in the CTROM business could be relatively higher than those in the concessions business due to lower margins, switching costs⁷⁶ and more stringent terms and conditions, but because the business of operating toll roads, whether these be in a CTROM or concession contract, requires specialised technical skills and knowledge, suggests to us that the barriers to entry are high.

79. This is supported by the evidence of Mr Dreyer, who confirms that while Intertoll’s roots can be found in the parking garage management services,⁷⁷ supply side substitution cannot take place easily without a significant upgrading of skills and knowledge,⁷⁸ This may explain why Toll Link, an equipment supplier is part of a consortium with Intertoll rather than bidding on its own.⁷⁹ It appears that International players who have shown an interest in the market recently seem keener to partner with local players rather than enter on their own.⁸⁰ Hence, we are of the view that barriers to entry in the toll road operations market are high.

80. SANRAL, as customer and regulator, could obviously facilitate entry through a number of mechanisms without compromising on health, safety and environmental standards. It could amend its tenders, combine tenders to improve economies of scale for operators or benchmark its terms and conditions more favourably thereby attracting a wider pool of bidders and still retaining its objectives of efficient delivery and risk transfer. It could also decide to increase the number of toll road concessions versus CTROM tenders in order to attract new players into the market or it could increase the size of the market by issuing more tenders.

81. However even if there was no such intervention by SANRAL, and new players could not enter this market easily due to the regulatory requirements and the high level of technical and specialist knowledge required, we are persuaded that the remaining players and prospective bidders in the relevant market for toll road operations (both CTROM and concession) pose a credible constraint on the merged entity to ameliorate any competition concern arising from this merger.

82. The market share figures submitted by the merging parties and the Commission reflect only those players who have been awarded tenders, not all of those who had submitted bids or who are prospective bidders in the relevant market. SANRAL’s own documentation, in particular its list of pre-qualifying prospective bidders, suggests that there are alternative credible competitors to the merged entity in the relevant market, who even if they are not successful in a tender and hence would not necessarily be reflected in the market share figures, would, by virtue of being on the short list provide a constraint on the merged entity.⁸¹ In a bidding market such as this one and where there will be at least more than two prospective bidders post merger,⁸² the mere fact that another entity submits, or could submit, a competing bid is likely to place a sufficient constraint on the merged entity.⁸³

NOTE ON UNDERGROUND MINING

83. As stated earlier, our analysis of the market for toll road operations applies equally to all the other relevant markets identified by the Commission. The market participants and their relative market shares are reflected in annexure A.

84. The Tribunal had a residual concern in relation to the market for underground mining. The Commission, without making a final determination, had identified several possible narrower functional product markets in underground mining viz. design and planning, shaft sinking, mine development and extraction of ore. The merging parties argued that while both of them were involved in the broader underground mining market, they focussed on different aspects of the market, suggesting that there was no overlap in their activities. Mr Trevor Robinson (Concor) and Mr Henry Laas (M&R Cementation) submitted that no competition concerns arose in this market because M&R tended to work more in the capital expenditure aspects of the market and Concor tended to focus on the working costs aspects of underground mining.⁸⁴

85. Mr Mario Gericke of Northam Platinum Gericke submitted that the merging parties were both involved in the broader market for underground mining, but that Northam Platinum did not necessarily see them as competitors because it had different contracts with each of the merging parties.⁸⁵

86. Although the merging parties argued that no competition concerns arose in this market due to a difference in focus, no further evidence was led to justify a narrower definition of the market. Both parties were involved in one or more of the services identified by the Commission in underground mining and while they were currently focussed on different aspects of that market, no evidence was led that, post merger, they would elect to focus on only one of the services contained within the broader market of underground mining. All the witnesses in fact confirmed that the relevant market for purposes of this transaction was the market for underground mining.⁸⁶

87. We are satisfied that the relevant market is the broad market for underground mining and the relative market shares and participants as listed in Annexure A. While the post merger market shares prima facie raise concerns, the accretion in market share is only 5%. Our evaluation of the competition impact of the transaction (above) leads us to conclude that despite the relatively high market share in this market, the transaction does not lead to substantial lessening or prevention of competition.

CONCLUSION

88. For the reasons stated above, we find that the transaction is unlikely to substantially prevent or lessen competition in any of the relevant markets. There are no public interest issues and we accordingly approve the transaction without conditions.

_________________________ 11 October 2006

Y Carrim Date

Concurring: N Manoim and D Lewis

Tribunal Researcher: M Murugan-Modise

For the merging parties: Advocate D Unterhalter instructed by Werksmans.

For SANRAL: Mr S Langbridge (Bell Dewar Hall).

For the Commission: M Mohlala (Mergers and Acquisitions) and D Motsamai (Legal Services).

ANNEXURE A⁸⁷

Estimated national market shares for the provision of building construction services, focusing on the large project sub market for 2005

Competitor	Market share (%)
WBHO	23
M&R	13
Concor	6
Grinaker-LTA	19
Group 5	19
Stocks Building Africa	19
Others	1
TOTAL	100
HHI (Pre-merger)	1874
HHI (Post-merger)	1974
Change in HHI	156

Estimated national market shares for the provision of civil engineering services, focusing on the large project sub market for 2005

Competitor	Market share (%)
Grinaker-LTA	20
M&R	16
Concor	7
Group 5	16
Steffanutti & Bressan	8
WBHO	7
Others	26
TOTAL	100
HHI (Pre-merger)	1208
HHI (Post-merger)	1400
Change in HHI	192

Estimated national market shares for the provision of engineering services for 2005

Competitor	Market share (%)
M&R	19
Concor	7
Group 5	30
DSE	21
Cosira	18
SMEI	3
S&P	2
TOTAL	100
HHI (Pre-merger)	2089
HHI (Post-merger)	2355
Change in HHI	266

Estimated national market shares for the provision of roads construction services for 2005

Competitor	Market share (%)
WBHO	34
Grinaker-LTA	24
Group 5	12
M&R	8
Concor	7
Raubex	8
Haw & Ingles	7
TOTAL	100

Estimated national market shares for the provision of underground mining services for 2005

Competitor	Market share (%)
M&R	43
Concor	5
Grinaker-LTA	7
Shaft Sinkers	12
Dellmann-Haniel SA	2
JIC Mining	6
Others	25
TOTAL	100
HHI (Pre-merger)	2232
HHI (Post-merger)	2662
Change in HHI	430

Estimated market shares for the provision of facilities management services for 2005

Competitor	Market share (%)
Tolcon (M&R)	24.1
Intertoll (Group 5)	41
Tracc	17
Concor	5.3
Dragados	12.6
TOTAL	100
HHI (Pre-merger)	2893
HHI (Post-merger)	2993
Change in HHI	255

---