---

COMPETITION TRIBUNAL OF SOUTH AFRICA

Case No: 63/LM/Jul06

In the matter between:

Lafarge Roofing (Pty) Ltd Acquiring Firm

And

Kulu Concrete Products

Kulu Roof Tiles Cape (Pty) Ltd

Kulu Roof Tiles (Pty) Ltd

(“Collectively known as

Kulu Group of Companies”) (Pty) Ltd Primary Target Firms

______________________________________________________________

Panel : N.Manoim (Presiding Member) Y Carrim (Tribunal

Member), and M Mokuena (Tribunal Member)

Heard on : 13 December 2006 and 27 February 2007

Order Issued : 12 March 2007

Reasons Issued : 18 June 2007

REASONS[NON-CONFIDENTIAL]

Approval

[1]. The Competition Tribunal issued a merger clearance certificate on 12 March 2007 approving this merger subject to two sets of conditions. The first set of conditions relates to the eventuality of supply constraints in Kwa Zulu Natal, the second, to the divestiture of a plant in the Western Cape. The reasons for our decision follow.

Parties to the merger

[2]. In this merger, Lafarge Roofing Pty Ltd (“Lafarge”) seeks to acquire the Kulu Group of companies (“Kulu Group”). Both firms are manufacturers of concrete roof tiles. Lafarge is at the time of this decision a wholly owned subsidiary of Lafarge Cement.¹ The Kulu Group of companies comprises the following firms: Kulu Concrete Products (Pty) Ltd Kwazulu Natal, Kulu Concrete Products Cape (Pty) Ltd and Kulu Concrete Products (Pty) Ltd Gauteng. ² Although not owned by a holding structure we are advised that the firms have a common controlling shareholder, Tony Redford. For operational purposes the three firms can be considered as being run as a group. The other owners of the Kulu Group are Mike Stirling and Rolf Moschinksy. Together with Redford they comprise the present management. ³

[3]. Lafarge has plants in Vereeniging, Brits, Bloemfontein, Durban, Port Elizabeth and Cape Town. Kulu has plants in Cape Town, Durban, Richards Bay and Germiston.

[4]. From this it is clear that the parties to the merger have plants located in close proximity in three provinces. We examine later whether these constitute firms located in the same geographic markets.

[5]. The product overlap between the firms is in the manufacture of concrete tiles. Although Kulu also makes standard size concrete building blocks, concrete common blocks, maxi concrete bricks, pavers and retaining blocks, these products are not of concern in this merger, as there is no overlap here with any equivalent product made by Lafarge.

Contending market definitions

[6]. The merging parties defined the market as the manufacturing or supply of pitched roof coverings. ⁴ The merging parties were non-committal on the geographic extent of this market, contending that the merger raised no issues however the geographic market was defined. ⁵

[7]. The Commission defined the market more narrowly as the market for concrete roof tiles. The Commission defined the geographic markets as being co-extensive with provincial boundaries. For this reason the merger created concerns for the Commission.

Relevant product market

[8]. The merging parties contend that concrete roof tiles are just one of the substitutes that compete in a market for pitched – as opposed to flat – roof coverings. Other products consumers can turn to according to them for roof coverings are steel, slate and thatch. On this basis they argue that the merged firm would only have a market share of 9% and hence the merger should not give rise to any further concerns.

[9]. The Commission has done a very thorough analysis of this aspect of the merger and comes to the conclusion from its market enquiries that other roofing products are insufficient constraints on the pricing of concrete tiles and hence cannot be considered in the same relevant market.

[10]. The key question here is the extent to which steel roofing products, compete with those of concrete tiles. Although other roofing materials are in use, such as slate and thatch, neither the Commission nor the merging parties view them as serious competitors.⁶

[11]. We therefore only focus on whether the steel products impose any competitive discipline on the pricing of the concrete product.⁷

[12]. This comparison is difficult to make for two reasons; the products come in a wide range from those aimed at the affordable segment to the luxury segment; secondly, as the merging parties have argued, customers have to consider not just the cost of a particular type of roof covering, but the cost of its installation. Labour and other associated structural costs vary depending on the type of covering selected. The Commission has resolved these difficulties, firstly, by comparing the costs of a standard roof product with a standard steel product, and then going on to consider the respective installation costs, in order to be able to compare the costs of the respective total roofing solutions.

[13]. It emerges from this that the standard steel product is generally more expensive than its concrete counterpart and that when comparing total roofing solutions, the steel roofing solution is about 20-30% more expensive than its concrete counterpart. ⁸ The Commission argues that this difference in price is significant enough to suggest the products are not regarded as substitutes.

[14]. Further complicating the picture is that even comparing the respective total costs of roofing solutions may not be a sufficient basis to consider their substitutability. According to the Commission’s feedback from the market, concrete tiles last longer than steel products, have less of a propensity to leak and have other attributes, which conceivably contribute to customers, perceptions of their superiority over steel. If this perception of value is translated into the total roof price then it would further increase the price differential.

[15]. The Commission has also studied the respective price movements of steel and concrete tiles to test whether they move in response to price changes in the other.⁹

[16]. The Commission concluded that the movements in the pricing of the steel product are largely dependant on the price of steel that a roofing manufacturer can obtain from sole supplier, Mittal SA. Similarly, the price of concrete tiles is largely dependant on the pricing of cement, with most manufacturers of tiles dependant on a single supplier. ¹⁰ The Commission concluded that the movements in their respective prices were explained by changes in the costs of their key inputs, and not as a response to a change in prices of the other product.

[17]. All this evidence points to the products not being substitutes. This view is strengthened by the views of customers and competitors, all of whom, when interviewed by the Commission, suggested that they considered concrete tiles to be in a separate market from steel roofing products.

[18]. No doubt there will be times when a rise in concrete prices not accompanied by a rise in steel prices, might bring prices closer together, but this is an insufficiently evidenced phenomenon to give us comfort that the one product is a constant discipline on the pricing of the other.

[19]. Although the merging parties never formally abandoned their adherence to their market definition, their subsequent posture in the case when they accepted both conditions proposed, which were predicated on the Commission’s market definition, suggests they did not press it with any great conviction.

[20]. We have therefore no reason for not accepting the Commission’s recommendation of the relevant product market, as being the market for concrete roof tiles.

Geographic market

[21]. The merging parties, as we indicated earlier, did not define a relevant geographic market, because their approach to the product market did not suggest any concentration would arise, post merger. The Commission’s approach has been to suggest that the markets are regional and co-extensive with provincial boundaries. On this basis the Commission contends for three regional markets where production facilities for the merging firms overlap:

1. Gauteng

2. Western Cape

3. Kwa - Zulu Natal

[22]. To be fair to the Commission, it has not suggested some serendipitous coincidence between the boundaries of the antitrust market - one driven by economic considerations and the province - one driven by political considerations; rather, its case is that these serve for the most part as a useful surrogate for an antitrust market, as firms have organized themselves on some regional basis when it comes to the choice of factory positioning and the supply of customers. For the most part this approach has worked, but we have differed with the Commission’s loose application of this approach in the Kwa Zulu-Natal region, where we have found a narrower definition of the geographic market was required, and having made that determination, the merger raised more problems in this market than the Commission considered there to be, because of its more expansive market definition. We will consider the competition implications for each region.

Gauteng

Table 1: Market Shares Based on Sales Volumes in the Concrete Tile Market in the Inland Region for 2005 ¹¹

Market Players	Market share (%)
Lafarge Roofing	24
Kulu	[11]
Marley Roofing	[32]
Concor	[6]
Ama Tiles	[4]
West Rand Tiles	[4]
Techon Tile	[3]

[23]. The Commission interchangeably refers to this as the Gauteng or inland market. It notes that post merger; Lafarge will have 37% of the market. The pre-merger HHI is 1622, and the post merger HHI is 2326, resulting in a change of 704. This suggests a highly concentrated market.¹²

[24]. The Commission, however, considered that barriers to entry in this market were insufficient to deter new entry or expansion by existing players. This assessment is reasonable. Apart from the merged firm there are three other large firms that compete in this market, namely, Marley Roofing, Concor and Brickor. The market also has three other smaller players. Thus, unlike the other regional markets we will consider, the Gauteng market whist highly concentrated has enough other players to exercise a competitive discipline on the merged firm, notwithstanding that it will leapfrog over Marley to become the biggest player in this market. The Commission’s research also confirms that entry barriers into this market are lower than on other regions. This is because inputs are more accessible, the customer base is larger and geographically more confined. This view seems to be fortified by the fact that there has been evidence of recent entry.¹³

[25]. None of those interviewed by the Commission appear to contradict this view in respect of this region.

In our view the Commission correctly found that despite the large post merger increase in concentration, the merger is unlikely to lessen competition in this market because barriers to entry are low.

Western Cape

Table 2: Market Shares for Concrete Roof Tiles in the Western Cape Based on actual production capacity for 2005. ¹⁴

	Actual Production Capacity for 2005 in million m²	Market Shares (%)
Lafarge Roofing	1.48	76
Kulu	0.46	24
Others	0	0
Total	1.94	100

[26]. As the Commission observes the merger will lead to a monopoly for Lafarge in this market. The Commission did not stop there, and very cautiously went on to examine entry barriers in the Western Cape. Finding them to be high on account of the relationships that exist between the tile manufacturers and their customers, where customers are bound to a supplier by contract, the Commission recommended divestiture of the Kulu Western Cape plant. Lafarge agreed to the proposal and the terms of the divestiture formed part of the Commission’s recommendations. It is therefore not necessary for us to consider this aspect in any further detail. In our view divestiture is the only appropriate remedy in this market, given the post merger monopoly the merger would have created.

Kwa Zulu- Natal

[27]. In its initial report the Commission defined the market as the whole Kwa Zulu- Natal province and identified the players in that market in the following table:

Table 3: Market Shares for concrete roof tiles in the KZN region based on current production capacity per million tiles per annum

Market Players	Current production capacity (mn p.a)	Market Shares (%)
Lafarge Roofing	9.6	24
Kulu (Durban)	7.2	18
Kulu (Richards Bay)	4.8	12
Saiba	2	5
Stanger Tiles	6	18
Kulu Crete	1.9	5
Marley Roofing	7.2	18
Kinaka Tiles	2	5
Total	40.7	100

[28]. From that table it emerges that the merged firm would have a high post merger market share (54%), but the Commission concluded that this did not raise concerns as it considered that barriers to entry in this market were not that high, as the market remained contestable, in contrast with that of the Western Cape.¹⁵

[29]. When the matter was first heard on the 13 December 2006 we raised questions about the adequacy of this market definition with the two witnesses who testified, Mr. Klaus Schubert from Marley Roofing and Mr Johan Van Jaarsveld from Lafarge. From this testimony we felt that the market was likely to be narrower than the provincial boundary and that this required further investigation by the Commission to see whether the market was perhaps a coastal market adjacent to the greater Durban area.

[30]. The Commission filed a further report on the 15^th of February 2007 in which it redefined the market as a Greater Durban market. The Commission in this supplementary report, states that customers are likely to source supply from plants up to 150 kilometers away from the manufacturer. Since both Kulu and Lafarge have a plant in the Durban area, the Commission defined the geographic market as being one within 150 kilometers of Durban.¹⁶

[31]. It recalculated market shares and these are inserted in the table below:

Table 4: Market Shares for concrete roof tiles in the Greater Durban Region based on maximum production capacity (thousand tiles per day) ¹⁷

Market Players	Maximum production capacity (mn p.day)	Market Shares (%)
Lafarge Roofing	45 0008	24
Kulu (Durban)	66 0009	35
Marley Roofing	36 000	19
Stanbrik Roof Tiles	20 000	11
Kulu Crete	20 000	11
Combined	111 000	59
Total	187 000	100

In this market the post merger HHI is 4084 and pre-merger HHI is 2404 and thus a change of 1680. This is leaving out Kulu Richards Bay, which on the Commission’s version is not in this market, because located 185 kilometers from Durban, it falls outside of the Commission’s market boundary of 150 kilometers. On the merging parties’ version, Richards’ Bay would fall into the market because they have a more expanded notion of the extent of the market.¹⁸ If this is correct the merging parties would enjoy even more concentration as including Richards Bay only adds the Kulu plant and no other competitor. (We are unable to do this calculation as the Commission has changed from doing market shares based on annual production in its first report, to market shares based on daily production in the supplementary report. As the Commission’s figures were not disputed by the merging parties, and are more favourable to them than would be if we included Kulu Richards Bay for the purpose of calculating HHI’s, we accept that we should work with the Commission’s table and thus exclude the Richards Bay plant.)

[32]. When we resumed our hearing on the 27 February 2007, the parties filed a report prepared by Genesis, an economic consultancy firm. The Genesis report analyzed the transport costs of both firms and compared pricing of certain merchant customers and came to a conclusion on the boundaries of the geographic market in the KZN area. This market, which has Durban at its centre, is a coastal market extending to the North and South, but not extending very far to the inland regions of the province. It is thus a market narrower than the one originally contended for by the Commission.¹⁹

[33]. In its report Genesis has produced a useful diagram, which we set out below, where the geographic market is outlined in the red sphere as well as the locations of the respective firms that the merging parties and the Commission contend are the competitors in this market.²⁰

[34]. Although the Commission has delineated this Greater Durban geographic market in a different way to the merging parties, nothing turns on this as they are both in agreement as to which firms compete in this market. The merging parties concede this in their evidence.

“CHAIRPERSON: And then if I can take you to page 15 of the Commission’s report, if you have that in front of you, it’s the Commission’s current report.

MR MYBURGH: Their latest report?

CHAIRPERSON: Yes, yes, page 15 and the table of the who the players are.

MR MYBURGH: Okay.

CHAIRPERSON: Are those the players you see in that market, or do you see other people they haven’t included? Are you in agreement with them then about that? I know you’ve given a different boundary, but they say look these are the people in some derivative of this market.

MR MYBURGH: Yes that is our view.

CHAIRPERSON: Okay, so you don’t have a dispute with them about that?

MR MYBURGH: No.”²¹

[35]. Whilst we accept that it would be difficult to determine the geographic market with any greater precision, the likelihood is that the one contended for remains too broad. Because of the extent of this market, and the centrality of transport costs to a manufacturer’s ability to be competitive, firms tend to be less competitive as the distance of the customer from its factory gate increases. This means that even if we consider firms to be located in the same geographic market, a firm faces the most vigorous competition from a rival to which it is most closely located, and conversely, is less threatened in its most proximate markets, by a rival more distant from them. Stanbrik Roof Tiles (“Stanbrik”) is identified as a competitor in the greater Durban market. Yet, when asked by the Commission about sales in the Durban area it answered:

COMMISSION QUESTION: What percentage of your production is sold in the greater Durban area?

STANBRIK ANSWER: 0% production is sold in the greater Durban area. ²²

[36]. The same answer is given to the next question, which is what percentage of the market Stanbrik has in the greater Durban area. ²³

[37]. Whilst the Commission did not define the Greater Durban market in its question, and thus it is not clear whether Stanbrik understood it in the same way as the Commission did, it is at least clear that Stanbrik sees transport as a more significant barrier to entry than do the merging parties, and whatever its conception of the greater Durban area there is an area close to Durban, the heartland of the merging firms, in which it makes no sales. Indeed the version on transport costs differs amongst the merging parties. Whilst Lafarge appears to levy a uniform transport cost across the region that Genesis defines as the Greater Durban market, Kulu differentiates its costs across this region with the differential reaching as much as 50% more. In one zone, this same differential is reflected within a zone.²⁴ On this point, Mr. Haywood, the operations director of Kulu, was perfectly frank. Asked to explain the discrepancy he testified that the firm would want to maximize its revenues and if it did not face opposition in an area its charges would reflect this and when it was it would in his words “ bring ourselves in line” He explained that the significant differential between prices in Port Shepstone and Umzimkulu were attributable to the existence of competition in one and its absence in the other despite the fact that they were grouped in the same transport zone for the purposes of their distance from the Kulu plant.²⁵

[38]. The evidence of Mr. Van Jaarsveld of Lafarge was that transport costs to a contractor (third parties are mostly used we were told) may not be passed on at exactly that rate to the customer.²⁶

He went on to say:

Mr. Van Jaarsveld: What I am trying to say is that in Amanzimtoti we may pay the contractor R550, 00 and we charge the customer R 720, 00 in this case. In the Pietermaritzburg case because it’s uphill, we may be paying R 720, 00 that’s in the zoned areas.

Ms Carrim: You may keep the margin then..

Mr. Van Jaarsveld: Correct.

[39]. The relative distances of the competitors from Durban according to the Commission are:

Marley – 45 kms

Stanbrik – 120 kms

Kulu Crete – 150 kms²⁷

[40]. Thus whilst firms may be regarded as competitors in a particular geographic market, it would be artificial, after having come to that conclusion, to have no regard to whether they are equally effective competitors of the merged firm over the whole of that region. One of the requirements that we consider in merger analysis is whether the merger will lead to the removal of an effective competitor. Because in this market, transport costs are fundamental in determining a firm’s competitiveness in a particular area, the closer the proximity of firms the greater their rivalry. In this respect because the Durban plant of Kulu is closest to the Lafarge plant it will be its most effective competitor, judged from location. According to the Genesis report, 68% of Lafarge’s sales take place in a segment of the market closest to Durban, where Kulu and Marley are the best placed geographically to compete for those sales. ²⁸ Thus although these firms have sales throughout this Greater Durban region, the bulk of sales are made much closer to the location of their factories.

[41]. We see a similar pattern examining customers buying patterns. One merchant BBS, which has a branch in Richards Bay, and thus is situated at the outer extremity of the Greater Durban market, informs the Commission that it sources nearly half its tiles from Kulu Richards Bay and a similar amount from Marley in Tongaat, whilst it sources only 2% of its tiles from Lafarge in Durban.²⁹

Analysis

[42]. Whilst we accept the definition of the geographic market in this region, as it is now contended for by the merging parties in the Genesis report, we remain of the view that not all the competitors in this market, and as we have seen there are now less of them than in the original KZN market contended for initially by the Commission, are as effective competitors to the merged firm over all areas of this market, albeit that they are capable of supplying the entire area. ³⁰

[43]. We now go on to consider whether the merger will lead to a substantial lessening or prevention of competition in this market, by considering some of the factors that the Act directs us to take into account in terms of section 12A(2).

Levels of concentration

[44]. The conventional measurement of concentration used in merger analysis is the Herfindahl-Hirschman Index (HHI). The United States antitrust agencies’ Horizontal Merger Guidelines, suggest that a market with an HHI of 1800 is highly concentrated and that a change in the HHI from the pre to post merger status of more than 100 is an issue of concern. ³¹ With that as the comparator, the concentration levels in this market post merger are stratospheric. The pre-merger HHI is 2404, thus we commence with an already highly concentrated market, whilst the post merger HHI is 4084 - a change in concentration of 1680. Expressed differently, we are going from a five firm market to a four firm market where the number one firm is buying number 2. Post merger, the two top firms will have almost 80% of the market. Another measure of analysis sometimes used is by summing the market shares of the four largest firms in the market or the CR4. Typically a CR4 of more than 75% is considered a highly concentrated market. Here, post merger, we exceed this figure with summing the market shares of just the two largest firms.

Removal of most effective competitor

[45]. We have noted that the merger leads to a market in which there are only four competitors. Apart from the merged firm, only Marley of the remaining three is a large and enduring concern. Stanger Brick and Tiles, now trading as Stanbrik, is a very recent entrant into this market, having only commenced operating in the concrete tile market in [CONFIDENTIAL], as an adjunct to its existing business in bricks.

[46]. Stanbrik is wholly owned by its founder.³² By comparison, the merged firm not only operates nationally, but its controlling shareholder is a multinational. ³³

[47]. Kulu Crete likewise is wholly owned by a single person, so it too suffers from the same potential constraints as would Stanbrik. From a location point of view this firm is least favourably placed relative to the merged firm, situated as it is at southern boundary of the relevant market.³⁴

[48]. Marley, whilst a well established firm and more favourably located to the merged firms Durban heartland has to date not been a vigorous competitor of the merged firms. According to the data submitted by Genesis, Marley has priced at the same prices as Lafarge both of which are more expensive than Kulu. ³⁵

[49]. Kulu suffers from none of the limitations of the other competitors of Lafarge. Although Kulu does not have a major corporate shareholder, it has three shareholders, is well established, and has operated nationally.

[50]. Its Queensburgh plant is closer to that of the Lafarge Durban plant than any other competitor, and given what we stated about the relationship between distance and competitive effectiveness in our analysis of the geographic market, this makes Kulu the best situated competitor to Lafarge.

[51]. According to the Genesis data, Kulu prices lower than either Lafarge or Marley. Internal documentation from the Lafarge board, suggests that Kulu has lower production costs and this may explain its ability to price lower in the market.³⁶

[52]. But there are other reasons besides pricing and location for why Kulu can be considered to be Lafarge’s most effective competitor in the Greater Durban area.

[53]. Kulu has the largest share of the market, calculated in terms of daily production capacity, with 35 %, while Lafarge is number two, with 24 %. Both firms produce the full tile product range, whereas of the remaining firms only Marley produces a similar range. In contrast to Marley, Kulu’s tiles are interchangeable with those of Lafarge, a fact raised in the due diligence report. ³⁷ Thus consumers in the course of a job could substitute with Kulu tiles if dissatisfied with pricing from Lafarge. Whilst this is not determinative for every customer or for every job, it does exercise some constraint on Lafarge’s’ pricing power pre-merger.³⁸

[54]. Thus the merger leads to the removal of Lafarge’s most effective competitor in this region pre-merger.

Barriers to entry

[55]. In its supplementary report the Commission echoes the views of the merging parties that barriers to entry in the market are low. They cite low capital costs for this. Whilst this may be correct, the Commission has not adequately considered problems associated with the most significant barrier to entry in this geographic market, which is obtaining supply of concrete, the key input for a tile producer. It is common cause that all the firms in this market are supplied concrete from one supplier, Natal Portland Cement (NPC). Concrete is an essential input and comprises roughly 56 % of the costs of production of a cement tile.³⁹ NPC has experienced shortages recently, and as a result, had to impose supply quotas on its tile customers. This emerged during the first hearing in the matter, during the testimony of Mr. Van Jaarsveld of Lafarge, who stated:

“MR WILSON: In the KZN area do you know where you source your cement requirements from?

MR VAN JAARSVELD: Yes.

MR WILSON: Who is that?

MR VAN JAARSVELD: All our cement comes from NPC and we’ve got a quota and if we run over that quota we stop the factory we don’t get anything more than that.

MR WILSON: And have had shortages along the way?

MR VAN JAARSVELD: Yes we’ve had days where we stood.

MR WILSON: When was that?

MR VAN JAARSVELD: In the last four months quite a number of days.”⁴⁰

[56]. Subsequent to the hearing on that day, the Lafarge version on the supply problem underwent revision. In reply to questions from the Commission on this point the merging parties stated:

“Both Lafarge and Kulu indicate that they did not experience cement shortages during the year. However, for those customers that did experience shortages, this has been addressed by NPC’s investment in further capacity.”⁴¹

[57]. When questioned on this seeming inconsistency, Mr. Van Jaarsveld, who testified again on the 27^th February 2007, explained that although they had been short supplied they knew what they were getting and hence did not consider that they had been restricted. ⁴²

[58]. It is fairly clear that Mr. Van Jaarsveld, when first asked about the supply problem in December 2006, was completely candid with the Tribunal. Subsequent attempts after this testimony, to place a more favourable gloss on the supply problem are unconvincing.⁴³ Moreover, NPC itself acknowledges the supply shortages in its submission to the Commission.⁴⁴ Mr. Haywood from Kulu, testified that by February of this year [2007] supply constraints will start to be overcome, because NPC is expanding production at its new plant in Samuma in Port Shepstone. ⁴⁵ NPC informed the Commission that more clinker capacity would be available from August 2007. ⁴⁶

[59]. We do not know at the time of deciding this merger, whether these expansion plans will solve the supply problems. If they do not, and quotas are applied again, they will prevent a rival which wishes to increase its markets share in response to a price rise by the merged firm, from being able to do so meaningfully. This is because the quotas are based on the previous years supply. This constrains the ability of a smaller firm, say Stanbrik, to increase supply in response to a price hike as it would not get the supply necessary to increase production over and above what it had ordered historically. When we consider barriers to entry, we consider not only whether firms can enter the market, but whether firms can expand in the market. ⁴⁷ The merged firm would have no reason to feel constrained by its rivals, if it knew that in response to an increase in price by it, its rivals could not expand their supply to customers who wanted to switch. Furthermore NPC past practice when it had to limit volumes was not to supply new customers. ⁴⁸

[60]. This is not to say that the merging parties’ sanguine expectations of improved supply out of NPC may not prove to be well founded. Rather, the risk is so great that unless the supply problem could be remedied it would create an insuperable barrier to entry as NPC is presently the only source of supply to concrete tile firms in this market.

Conclusion on the Greater Durban Market

[61]. The merger leads to further concentration in an already concentrated market, the removal of Lafarge’s most effective competitor. While this on its own may not be decisive in condemning a merger if entry was easy, we have noted that because of the history of supply constraints, entry may be constrained in the future. For this reason the probabilities are the merger will lead to a substantial prevention or lessening of competition in this market. We discuss below how this can be remedied.

Remedy

[62]. In this merger we have imposed two remedies. The first is a structural remedy in relation to the Western Cape market, which as we noted earlier, was the subject of an agreement between the merging parties and the Commission. We have not tampered with this arrangement, apart from some fine tuning on procedural issues, to resolve a dispute on these points between the merging parties and the Commission.

[63]. A divestiture of the Kulu Western Cape plant will at least restore the competitive state pre-merger provided the purchaser is a viable competitor.

[64]. The second remedy is to deal with the substantial lessening of competition that we concluded the merger will bring about in the Greater Durban market. The purpose of the remedy is to lower entry barriers into this market and to allow for the expansion of existing and future competitors by ensuring that the source of cement supply is not constrained in the future as it once was in the past. Given the demands on cement suppliers at present due to ambitious infrastructure programs, the World Cup in 2010, and the fact that the new NPC plant may take time to meet its production expectations, the supply constraint problem is most critical in the near future. For this reason we have imposed the condition for a limited period of three years. The merging parties agreed to the condition. Had they been unwilling to accept such a condition we would have had to seriously contemplate whether the merger could be approved without the divestiture of one of the merged firm’s two Durban plants.⁴⁹

[65]. Although there is always a concern that a behavioural condition such as this may be difficult to enforce, we take note that in this case, past practice indicates that NPC has been able to administer a quota to deal with shortages, and secondly, that competitors, who in terms of the order have been informed of the condition by the Commission, will be the parties best placed to police its adherence and hence this will not require any resources from the Commission.

Public Interest

[66]. This merger raised no public interest concerns and hence they have had no bearing on our consideration of the merger and the remedies discussed above.

Conclusion

[67]. The merger is approved subject to the conditions contained in the Annexure. A confidential version of the conditions was sent to the merging parties on 12 March 2007, and a non-confidential version was also circulated on the same day.

___________________ 18 June 2007

N. Manoim Date

Tribunal Member

Y Carrim and M Mokuena concurring.

Tribunal Researcher : J Ngobeni

For the merging parties : Adv Jerome Wilson instructed by Webber Wentzel Bowens

For the Commission : Seema Nunkoo and Dumisani Motsamai

---