---

COMPETITION TRIBUNAL OF SOUTH AFRICA

Case No: 84/CR/DEC09

In the matter between:

THE COMPETITION COMMISSION …....................................................................Applicant

And

AVENG (AFRICA) LIMITED t/a STEELDALE ….......................................First Respondent

REINFORCING MESH SOLUTIONS (PTY) LTD …..............................Second Respondent

VULCANIA REINFORCING (PTY) LTD …................................................Third Respondent

BRC MESH REINFORCING (PTY) LTD ….............................................Fourth Respondent

Panel : Norman Manoim (Presiding Member), Medi Mokuena (Tribunal Member) Lawrence Reyburn (Tribunal Member)

Heard on : 28 February 2011 – 2 March 2011

Final Argument on : 8 August 2011 and 10 August 2011

Order issued on : 7 May 2012

Reasons issued on : 7 May 2012

Reasons for Decision and Order

Introduction

1. On 2 December 2009 the Competition Commission (the “Commission”) referred a complaint against four firms (the “respondents”) alleged to have been involved in a cartel of wire mesh producers. One firm, BRC Mesh Reinforcing Limited (“BRC”), the fourth respondent, was granted conditional immunity by the Commission in terms of its corporate leniency policy (“CLP”); another, Aveng (Africa) Limited trading as Steeledale (“Steeledale”), the first respondent, has since the referral, settled with the Commission in terms of a consent agreement.¹ The two remaining firms, Reinforcing Mesh Solutions Pty (Ltd) (“RMS”), the second respondent, and Vulcania Reinforcing Pty (Ltd) (“Vulcania”), the third respondent, have defended themselves on various grounds. The hearing to which this decision relates concerns proceedings the Commission has brought against these two respondents.

Commission’s case

2. The Commission’s case is that the respondents contravened sections 4(1)(b)(i) and 4(1)(b)(ii) of the Competition Act No. 89 of 1998 (“the Act”), during a period that lasted from at least 2001 to 2008. The contraventions comprised agreements between the respondents and at times other firms on the following matters:

   1. A price list for reinforced mesh products;

The level of discounts offered to particular categories of customers;

The allocation of customers.²

3. The Commission alleges that the agreements came about as a result of meetings that initially took place under the auspices of an industry association, the South African Fabric Reinforcing Association (“SAFRA”) and later as a result of informal meetings involving some or all of the respondents. ³

Background

4. Wire mesh is an input into the construction industry. It is used to reinforce concrete. The wire mesh manufacturers purchase steel rod in coils from steel mills and put it though a process by which the material is drawn down to a specific diameter and then welded at fixed intervals corresponding to the desired mesh apertures.⁴ Customers include construction firms and resellers of building products. Customers thus vary greatly in size and nature as well as geographic location.

Firms competing in this industry produce a product considered by their customers as a commodity with limited differentiation. The business model of most of these firms is to add a margin to their input costs per unit. The inputs largely comprise steel rod, said to constitute roughly 60-80% of the price; labour used to convert the input into the final product; and transport.⁵ Most firms in this industry testified that they are faced with similar input costs. Steel prices from the mills vary only as to quantity purchased, while labour costs are similar.

It is common cause that a cartel existed in this industry for some years. No one is sure for how long as none of the current witnesses were present at its inception, but they were aware of its existence by the time they joined it. The industry leaders were Steeledale and BRC; historically they, or their predecessors, formed the core of the cartel. For the purpose of this decision their participation is not controversial.

What is in issue for us to decide in this case is in respect of Vulcania; whether it is liable and if so what would be an appropriate remedy to impose on it. As we shall consider later, Vulcania, whilst admitting attendance at meetings with its competitors on several occasions, denies that its actions amounted to an agreement to join the cartel and hence denies liability.

In respect of RMS, which admits liability, the issue is the extent of its involvement in the cartel and then two further issues – whether a penalty can be competently imposed upon it, for reasons we consider later, and if it can, what an appropriate penalty is.

The approach we have taken in this decision is to deal first with the liability of the respondents. In this regard we deal initially with factual issues that affect both respondents and then we consider them separately, to evaluate their respective participation.

Hearings

5. The case was heard from 28 February to 2^nd March. Final argument was heard on 8^th and 10^th August 2011. The following witnesses testified: for the Commission, Pierre Griffin (formerly sales and marketing manager of BRC Mesh), Martin Cawood (sales and marketing manager of BRC Mesh, who succeeded Griffin) and Jeremy Hartnady (marketing manager of Steeledale Mesh); for RMS, Carlo di Nicola and Trevor Hankey; and for Vulcania, Sean Greve.

PART A

LIABILITY:

Origins of the cartel

6. SAFRA, the industry association for the wire mesh industry, was formed in 1970. Its ostensible purpose was to promote the use of wire mesh by industry. Wire mesh competed with the reinforcing product known as rebar. SAFRA’s intention was to promote the use of wire mesh over rebar in the building industry.⁶

The problem faced by the wire mesh manufacturers was that they had to pay more for their rod inputs than did the firms supplying rebar. Also, mesh is a capital-intensive industry. Rebar whilst cheaper to manufacture is more expensive to install, as it comprises loose bars that have to be manually tied together. Whilst this latter consideration might have balanced the scales between the two functionally similar products, mesh firms were concerned about preserving the margins they earned on their product.⁷

At some date, none of the witnesses were quite sure when, meetings of the industry association, which were attended by competing wire mesh manufacturers, decided to discuss a recommended price for their product.⁸ At the very latest, on the documentary evidence before us, this had occurred by December 2001.⁹ The modus operandi of SAFRA was that a price was discussed at meetings and then agreed. Afterwards SAFRA would circulate a recommended price list to members. According to Pierre Griffin once the list had been circulated all four of the respondents in this case would adopt it. ¹⁰

The price increases followed a similar pattern over the years. The meeting would get information on likely steel price increases from the main manufacturer Iscor, later ArcelorMittalSA, (“AMSA”) and wage settlements from SEIFSA (the steel industry federation to which SAFRA was affiliated). They then applied these increases to a pre-determined formula resulting in the recommended price.¹¹ The formula was most closely associated with one individual from Steeledale, Costa Cassa, who had at times chaired SAFRA, and hence it was referred to as the ‘Costa formula’, although it was not necessarily his creation.

The important aspect of the formula was that it enabled the competing firms to uniformly pass on input cost increases by providing a common mechanism for deriving the final price. Absent such an agreement, although firms were faced with similar cost increases at similar times, it does not follow that in a competitive market the final price to their customers would have been identical. Increases in input costs might have been passed on in varying amounts or not at all; or passed on at different times, depending on how much stock of rod a firm had stockpiled at pre-increase prices.

Pierre Griffin of BRC, testifying for the Commission, explained that the timing of the increases was vitally important. If price increases were not implemented simultaneously, customers would have had the opportunity to search for better prices which might have led firms which implemented earlier to lose customers, possibly permanently, to firms which implemented later.¹²

Thus what the cartel’s agreements served to do was to resolve several important pricing issues for its members when faced with input price increases for steel rod and labour. They answered what their own responding increases would be, how they would pass them on to their customers, and when.

The kernel of this mechanism, which was confirmed by Jeremy Hartnady of Steeledale, who also testified for the Commission, was the Costa formula, which was set out in the documents discovered by his firm. Because as a matter of practice all customers were given a discount off the list price, the firms in the cartel had to reach agreement on the level of discount as well. On the example given by Hartnady, once all the calculations had been made, the increase on the list price was to be 10,4%.¹³ The effective increase, however, once the discount was applied, was 8,9%.

When did the collusion start? The document in the record setting out the Costa formula is dated May 2001 and reflects that the relevant increase was to be effective from 1 July 2001. Griffin testified that a base formula existed even before 2001.¹⁴ A minute of a SAFRA meeting in June 2001 records that “...the present method of escalation calculation remain.”¹⁵ Thus the probabilities are that the arrangement was in place from at least 2000. The exact date is not relevant in respect of Vulcania and RMS as it is common cause that their attendance at cartel meetings took place after this period. What is relevant for the purpose of remedies is that the cartel was in existence for some time before these companies are alleged to have joined.

Curiously, the evidence was that prior to 2005, the use of the SAFRA base price was transparent. Customers could get it from SAFRA and members of SAFRA gave it out to them. This conduct was not considered by SAFRA members to be anti-competitive.¹⁶ Sometime in 2005, however, the association was given legal advice by a Mr McDonald that the arrangement might infringe the Competition Act. This is Griffin’s recollection, but there is also documentary evidence to support this.

In the minute of the SAFRA executive meeting of 5 July 2005, it is noted that the opinion of McDonald had been circulated. The way members responded to this advice is noteworthy. Instead of abandoning the practice they resorted to subterfuge. In the minute of the same meeting, it is recorded, “It was agreed to maintain the status quo and not to change the current situation in this regard.” ¹⁷ Witnesses for the Commission confirmed that members continued to apply a SAFRA base price, but did so on their own letterheads. Griffin testified that he would check up on competitors from time to time to see whether they were following the agreed pricing.¹⁸

Hartnady also confirmed this modus operandi. He added that members might have “...tweaked the prices a little bit, just to make it look as if everything wasn’t copied verbatim and they would apply that to their own letterheads.”¹⁹

Trevor Hankey of RMS confirmed during cross-examination that this was the case.²⁰

Meetings between the firms also took place informally i.e. not as part of a SAFRA meeting. They were held in restaurants and hotels. It is not clear from the record or the testimony when the informal meetings started. From the notes discovered by Steeledale informal meetings date back to at least January 2001.²¹

Why did the cartel members need informal meetings if they were able to meet as members of SAFRA? One explanation is that SAFRA operated transparently, keeping minutes of meetings at which agreements reached were recorded, and had its own staff who were not industry players. The cartel members would not want the SAFRA staff to know of the cartel’s agreements on issues other than the pricing formula. Thus although informal meetings seem to have been in existence in parallel to those of SAFRA, they were used to discuss that which could not be discussed at SAFRA. This subject-matter included market division, customer allocations and, after McDonald’s 2005 warning, the pricing formula.

Griffin testified that informal meetings discussed pricing adjustments, allocation of customers and any deviations from agreements. He mentions that attending these meetings were Hartnady of Steeledale, Hankey of RMS, Greve of Vulcania, and himself for BRC.²²

Hartnady testified that the purpose of the informal meetings was, in his self-serving vision of the cartel, to: “...reinstate some sort of sanity into the market.” He testified that profit margins, as the cartel members perceived it, were getting down to break-even. ²³

The cartel’s pricing formula provided for steep discounts from the list price. Indeed this was an issue of concern to them as customers were suspicious of large discounts.²⁴ What the cartel members agreed upon as well, was the size of different discounts to different types of customers.²⁵ The largest discounts were given to fellow subsidiaries or associate companies within the same group, where a cartel member was part of a group of companies. Discounts also varied in size depending on whether customers were in the construction industry or were merchants and, in the latter case, whether they were large or small.²⁶

In the last years of the cartel, at a time when both Vulcania and RMS attended meetings, there was a trend to move away from agreeing on pricing to customer allocation. Hartnady explained in his testimony that there were different views on price increases and new firms were moving in and “...undermining settled scenarios...” For this reason firms started moving to agreements about customer allocation and market division to avoid a “free for all.” ²⁷

RMS – Liability in terms of section 4(1)(b)

7. RMS is a subsidiary of Capital Africa Steel (Pty) Ltd (‘CAS’). The position of this company in its group structure is a controversial matter in this case, as we discuss more fully later in the remedies section. In March 2002, a firm called Burbank Investments changed its name to Reinforcing Mesh Solutions with the intention of entering the steel mesh and reinforcing markets. Trevor Hankey joined RMS at this time to launch the mesh business. Hankey was previously a director of BRC, the fourth respondent. During the time he was with BRC, Hankey had attended various meetings of SAFRA on BRC’s behalf and is recorded as having attended some of the informal meetings of competitors in that capacity in November 2001.²⁸

Hankey acknowledged this prior involvement which, from correspondence to him in his BRC capacity, continued until at least 6 February 2002.²⁹

Hankey’s evidence, which corresponds with the position adopted by RMS, is that the firm was a late and reluctant member of the cartel. Hankey stated that when the firm entered the market it faced a number of obstacles which he attributed to the conduct of competitors. Producers of plant, he testified, would only supply equipment at excessive prices and RMS had difficulty in obtaining supplies of rod from steel mills.³⁰

Eventually RMS entered into a supply agreement with Iscor, as AMSA was then known, and commenced operations with its first run in November 2002.³¹

Whilst Hankey may have a motive to exaggerate these difficulties there is evidence that competitors, in particular Steeledale and Allens Meshco, did not welcome RMS’s entry into the market. A minute of one of the informal meetings, dated 3 October 2002, whilst cryptic, indicates concerns about RMS’s entry from those who were at this meeting. One entry states: “What will Steeledale (sic) approach be with RMS” The next entry, which may have been made at a meeting later in October, records the remark, “Allens Meshco recommends that we put a new price list that makes it difficult for new comers.” A suggested price follows. It is not clear whether this suggestion was followed.

8. Nevertheless RMS, despite its promising pro-competitive start, eventually joined SAFRA and the cartel. We only have Carlo Di Nicola’s version of how this happened. Di Nicola was the chairman of RMS and a former employee of Murray and Roberts, which owns BRC.

9. In its formative years, as we noted earlier, RMS was considered an interloper, disrupting a comfortably settled industry. BRC, in particular, threatened by RMS, responded by degrading the reputation of RMS’s products in the market. Upset by this, Di Nicola, who was still in contact with Murray and Roberts’ people, and being an alumnus of that firm, mentioned his concerns to a colleague there. Di Nicola states he had retained regular contact with his erstwhile employer in respect of other work that he had to hand over. In the course of such discussions those sympathetic to Di Nicola set up a meeting with the relevant BRC people to discuss the so-called ‘bad-mouthing’. At the meeting, held at a golf range, at which Di Nicola was present, a senior Murray and Roberts person told his colleagues to desist from that behaviour. It seems that everyone made up. That, according to Di Nicola, was as far as the conversation went at that time. In his witness statement Di Nicola dates this meeting as some time in the first quarter of 2004.³²

10. At a later stage, and Di Nicola was not clear about how long it took, he was phoned by Rick Allen of Allens Meshco to discuss “... how the market worked”.³³ At Allen’s suggestion, Di Nicola hosted a meeting at RMS’ offices which Allen and representatives of BRC attended. Thus although on Di Nicola’s version the initial contacts did not have a collusive nature, they subsequently, at the behest of Allen, did.

At this meeting the size of the market was discussed and, at Allen’s suggestion, everyone agreed to submit their last six months’ tonnages so that the total market size could be calculated at a follow-up meeting.

Present at the further meetings were RMS, BRC and Allens Meshco. Di Nicola claims that Allen represented Vulcania as well, as he knew what their tonnages were. (Under examination from Vulcania’s counsel he qualified this to say he did not know whether Allen represented Vulcania as he could have known its tonnages from being its supplier.³⁴)

Di Nicola said that at this meeting the suggestion was made that each firm reveal its customer list, and that each should ‘respect’ the others’ customers. He then instructed Hankey to do this. As he put it in his testimony:

“...Trevor there is nothing wrong with it, just show the people what you are dealing with and everyone else did the same and that famous list that is in the files is the outcome of that.”³⁵

11. They agreed to ‘respect’ one another’s customers although Di Nicola states there was not much overlap in any event.³⁶ Thereafter, Di Nicola testified, he was no longer involved in meetings with competitors, but Hankey was.³⁷ Hankey briefed him on these from time to time.

According the Hankey his first contact with competitors while representing RMS occurred when Fred Erasmus and Pierre Griffin of BRC contacted him early in 2003. They proposed buying RMS’s entire mesh production on an ongoing basis. Although Hankey was receptive to this proposal nothing came of it.³⁸

Hankey stated that no further contact with competitors took place till the following year. Pursuant to the meetings held with Di Nicola, he was invited, either by Hartnady of Steeledale or Erasmus of BRC, to attend a meeting with the other respondents, excluding Vulcania.³⁹ At this meeting client lists were discussed. Hankey suggested the meeting may have taken place on 24 March 2004. In his testimony Hankey stated that RMS was invited to join the cartel in 2003 and he joined SAFRA meetings in 2004, and in his words “...from there on we proceeded with informal meetings as a cartel”.

Hankey conceded that several informal meetings took place between 2004 and late 2007 or early 2008. Thus on his version RMS’s participation in the cartel lasted approximately four years. He said that the last meeting of the cartel was around January 2008. ⁴⁰

Records from SAFRA’s minutes of RMS’s relationship with SAFRA offer independent corroboration of Hankey’s version. Initially, and consistently with the oral testimony of Di Nicola and Hankey, RMS seemed reluctant to join SAFRA. In October 2002 and thus at a time before RMS had commenced trading, the SAFRA executive minutes reflect that “...Mr Hankey of RMS has indicated that his company wished to join the Association. Mr Mountford to follow up.⁴¹ However, at the next meeting it is recorded that a Mr Pastorino from RMS had told Mountford his company declined to take up membership.⁴²

But then on 19 February 2004, the minutes of the first SAFRA meeting for that year record that the organisation welcomed “... the new member represented by Mr Hankey”. Given the close relationship between members of the cartel and membership of SAFRA, it is likely that at least by the beginning of 2004, RMS was a member of the cartel.

The question is whether there is any evidence to show it was a member of the cartel earlier. The Commission has not challenged this version nor has it led any evidence to the contrary. Hankey also offers a plausible reason why other members of the cartel might not have made overtures to RMS before 2004. He surmises that the large firms hoped that RMS would fail in the market and only when it had not done so after a year, did they make the approach.

Hartnady suggested in his evidence that the cartel members had two options at the time for dealing with RMS, co-option or exclusion. The easier option was to get RMS to meetings and get it under the cartel’s control.⁴³ He did however confirm that prior to joining the cartel, RMS had approached BRC to supply it with wire and BRC had declined, because RMS was going to compete with it.

We find that the probabilities are that RMS only became a member of the cartel in 2004. It also is common cause that the cartel ended in early 2008, when Murray and Roberts, the owners of BRC, approached the Commission in terms of its leniency application in a variety of construction-related industries. The cartel in the wire mesh industry was revealed as a consequence of this approach.

There is no dispute about the content of the cartel agreements during the four-year period when RMS was a member. Members, as we have noted, agreed on the pricing formula and its implementation from time to time when faced with changes in input prices. They allocated customers and monitored adherence. The Commission has therefore established the allegations made out in paragraph 13.1.1 to 13.1.3 of the founding affidavit insofar as RMS is concerned.

12. We thus find that RMS has contravened sections 4(1)(b)(i) and 4(1)(b)(ii) of the Act.

We deal later with other evidence in relation to RMS’s degree of participation when we consider the remedy.

Vulcania – Liability in terms of section 4(1)(b)

13. Vulcania’s evidence was that it commenced attending meetings of the cartel only in 2006.

Sean Greve, the managing director of Vulcania, stated that in January 2006 he was approached by Adrian Mountford to join SAFRA. The firm joined SAFRA that same month. He attended his first meeting on 3^rd March 2006. At more or less the same time he attended informal meetings of the cartel members. He stated that he attended no more than five or six of these meetings. The other attendees were representatives of the respondents in this case. His description of the subject-matter of meetings is consistent with the testimony of other witnesses; in particular which manufacturers would serve which customers, that manufacturers should not supply other firms’ customers, and agreement on pricing and levels of discounts.

But Greve suggested that decisions taken at these meetings were not final, as they were subject to confirmation by and the approval of senior management of the respective companies. Vulcania, however, did not have a management level higher than Greve. Greve asserted that in these circumstances Vulcania did not consider these decisions as binding upon it and it did not adhere to them. In particular, according to Greve, Vulcania did not adhere to the pricing formula or respect the customer allocations made at these meetings.

In a nutshell, Greve contended that Vulcania’s attendance at cartel meetings was a sham, calculated to make its competitors on whose goodwill Vulcania depended for some of its supplies believe that Vulcania was a member of the cartel and thus obviate retaliation in the form of supply constraints. Once Vulcania had obtained an independent source of supply for rod in coil and installed the requisite machinery to process it, Vulcania ceased attending meetings of the cartel. This, he said, occurred some time in 2008.

In order to evaluate the credibility of this defence it is necessary to understand Vulcania’s history in greater detail. Vulcania was the last of the respondents to join SAFRA and to attend the so-called informal meetings. Vulcania evolved out of a firm known as Polymesh which had entered the reinforcing mesh market in the 1990s. Polymesh produced mesh from wire in coil as opposed to rod in coil, which is what the other respondents did. The consequence of this was a cost disadvantage of 10-15%.

This cost disadvantage forced Polymesh into becoming a niche supplier to small building contractors, supplying them with all their needs. Vulcania was formed in 2000 and took over the Polymesh business. Initially it too suffered from the same supply limitations. Matters only changed in 2005 when Vulcania purchased a 50% interest in a company called Steel Straighteners which was another mesh producer.

Its partner in the takeover of Steel Straighteners was an Allens Meshco subsidiary. After the takeover Allens Meshco commenced supplying rod in coil to Steel Straighteners.

Initially, despite this relationship with the Allens Meshco Group through Steel Straighteners, Vulcania still experienced problems in getting rod in coil supplied.

However the Steel Straighteners takeover had created an additional opportunity for Vulcania. Cape Gate, a steel mill that supplies rod in coil had lost a customer in Steel Straighteners as a result of the takeover. Vulcania was able to negotiate a favourable agreement with Cape Gate to supply rod in coil. At this time Vulcania did not have the requisite machinery to draw the rod wire. It took another year for Vulcania to commission, purchase and install this machinery. Thus according to Greve it took till early 2008, once the new machinery was running, for Vulcania to become a serious competitor.

Vulcania like the other respondents eventually became a SAFRA member. The SAFRA minutes for February 2003 recorded that Mountford was to contact Greve of Vulcania and Cardinal of Cape Gate to try to get them to join the association. This is in the context of SAFRA recruiting other members, as overtures to Cape Gate and a renewed request to RMS were discussed at this time as well.⁴⁴

No mention was made in the minutes of the outcome of this overture, if it took place at all. However the minutes of two subsequent meetings indicate that no new members had applied. One minute, from June 2003, records that the time was not ripe in the market for new members and that the association should wait till things ‘settle down’ because there were a number of ‘changes taking place in the industry’.⁴⁵

The next express mention of Vulcania in the minutes came in June 2004 when it was recorded that a Mr Erasmus had requested that a membership application form be sent to Greve.⁴⁶ Again the minutes recorded that other firms were to be approached to become members. including Steel Straighteners.

It seems that Vulcania was not interested in becoming a member as it was recorded at the next SAFRA executive meeting, in September 2004, that Mountford had again attempted to contact Greve “... but to no avail.”⁴⁷ The next entry on the subject came a year later, in the minutes of the SAFRA executive in August 2005, which recorded that Hankey (now of RMS) was to approach Vulcania to join.⁴⁸

Vulcania joined SAFRA in early 2006. The minutes of the SAFRA executive meeting dated 7 February 2006 reflect that Greve was welcomed as attending his first meeting, “...his company having recently joined the association.” ⁴⁹

At the meeting in April 2006 it was recorded under the heading “Escalation strategy” that “..Mr Cassar’s formula for escalations will be passed on to Mr Greve by Mr Hartnady.”⁵⁰ This is clearly a reference to the Costa formula discussed above.

Although Greve says in his witness statement that he attended his first SAFRA meeting in April 2006 he corrected this in his oral testimony and accepted the correctness of the minutes i.e. that Vulcania joined in February 2006.

The significance for Vulcania of the date when it joined SAFRA is that Greve testified that he only commenced attending informal meetings after he had attended his first meeting at SAFRA.⁵¹ This means that on his version, he only commenced attending the informal meetings from early 2006.

However other evidence suggests his attendance at informal meetings may have preceded formal membership of SAFRA. A schedule of customers faxed on 26 October 2005, discovered by Steeledale, headed ‘Allocations/pricing’ contains a list of customers’ names with the name of the supplier next to each. The suppliers’ names include BRC, RMS and Steeledale. But significantly included are customers seemingly allocated to Vulcania. If customers were allocated at an informal meeting to Vulcania, the probabilities are that Greve attended a meeting at about that time.

If Hankey, as the SAFRA minutes state, made an overture to Greve to join SAFRA in around August 2005, it may well be that an invitation to attend informal meetings and acceptance of the invitation were part of the same conversation, which would place Greve at an informal meeting as early as late 2005, thus consistently with the date of the schedule. Nor did Greve explain how he came to attend the informal meetings. He limited his explanation to the invitation to join SAFRA which he said came from Mountford. If Vulcania had commenced attending the informal meetings of the cartel at an earlier date this would be relevant to the issue of duration for the purpose of a remedy. However as the Commission has not challenged Vulcania’s version on this aspect, and as the addition of three extra months does not materially alter the duration of participation, we will accept Vulcania’s submission that its involvement did not exceed a period of two years.

Whatever the precise period of the informal meetings Greve recalls attending no more than five or six.⁵² Again this fact is not in dispute. The content of the discussion, according to Greve, related to pricing and levels of discounts to be given to the various classes of customers. Also discussed were which customers ‘belonged to’ which manufacturer and the cartel rule that manufacturers should not try to supply the customers of other manufacturers.

14. In this respect Greve’s version of the content of the discussions does not differ markedly from those of the witnesses called by the Commission. What he does dispute is the centrality of his own role in these discussions and the status of agreements reached during their course.

15. Vulcania contends that it cannot be held liable in terms of section 4(1)(b) in respect of its attendance at meetings with its competitors because: (i) it took an entirely passive role in these discussions and; (ii) the agreements reached at these meetings were inchoate in that they required subsequent ratification by more senior managers of the firms at meetings to which it was not party.

We deal first with the issue of Vulcania’s passive role in discussions. Since only Greve represented Vulcania at these meetings, evaluating that firm’s conduct means assessing his conduct.

Martin Cawood of RMS testified about an incident in which he, as a newcomer to the meetings on behalf of his firm, was being criticised for not having passed on price increases in accordance with the Costa formula. Rather than increasing the full selling price of the mesh by the same percentage as the increase in the relevant input cost, he passed it on proportionately; thus if the steel price increased by 10% and steel constituted 80% of the input costs of mesh, he would pass on a final price increase for the mesh of 8%, not 10%. Hence Cawood was charging a lower price to customers than were the other suppliers. He was hauled over the coals for this by the others at a meeting Greve attended.

Cawood testified that at some stage during the meeting Greve took out his calculator and performed a calculation to show that he (i.e Cawood), was not passing on the increase as agreed. This Cawood interpreted as indicating Greve’s willingness to go along with the cartel and Greves’ enthusiasm for its objectives. Greve denied having admonished Cawood at this meeting, saying that he was in no position to do so. What he appeared to mean by this is that Vulcania, given its small market share, was in no position to exert muscle over one of the bigger firms.⁵³ Under cross-examination he appeared to suggest that no one admonished Cawood; neither he nor anyone else present.⁵⁴

He does however admit that he took out his calculator, but says this was to assist Cawood to use the formula correctly.⁵⁵ Whilst this vignette might seem trivial it is relevant to the question emphasised by Vulcania that although Greve attended five or six meetings he was largely passive – as he put it in testimony “... taking instructions and listening.” On this dispute of fact we find Cawood’s version more credible. If Greve was the passive listener he would not have taken out his calculator to intervene in a dispute between the bigger players as to whether Cawood was adhering to a previous understanding over price increases. The fact that he did, suggests he was anxious to show the others that his firm was no less enthusiastic than they about increasing the full selling price and not merely adjusting the selling price to reflect input prices. It is fair to say that the other firms in turn would have inferred from this that Vulcania was a willing member of the cartel.

There is no reason for Cawood to have embellished his evidence on this aspect and given that he was the one being criticised he is likely to have had a clearer memory of it.

But this is not the only evidence of a lack of passivity by Greve. Two of the other witnesses, Hartnady and Griffin, testified that Greve had attended meetings and contributed to the proceedings at them. Greve tried to downplay this and stated that he merely contributed his customer list.⁵⁶ In its answering affidavit Vulcania admits that it:

“...created an appearance of co-operation with the other mesh manufacturers... because of its vulnerability in the market and only while it was perforce reliant on the benevolence of those other manufacturers” ⁵⁷

16. Thus even on its own version Vulcania represented sufficiently to the others that it was part of the understanding. It does not suggest that its alleged passivity was open to misinterpretation by the others. On the contrary, its sense of vulnerability required it to gain the confidence of the other firms that it was an adherent to their collusive agreements.

But his involvement went further than attendance at discussions. Greve does not deny contributing his customers’ names to the list that was prepared and of which we have a manuscript version in the record. He states however that the customers on the list were already his firm’s and that Vulcania gained nothing in the process. He also asserted that Vulcania ‘cheated,’ and did business with customers regarded as allocated to other firms.

Even if this is so, and we have to accept his evidence on this aspect, this does not exculpate Vulcania. Having participated in the price-setting and customer allocation processes, even if he never intended to implement the agreements reached, his conduct facilitated the cartels’ customer allocation endeavours. The firms in the room were given no reason to believe that Vulcania would compete for their allocated customers and hence they would have to lower their prices to them. Since these participants believed that Vulcania was going along with them, they did not withhold the quid pro quo: there is no evidence that any of them sought to do business with Vulcania’s customers. To that extent the cartel arrangements, whether Vulcania’s participation was real or a sham, protected Vulcania from potential competition, and to that extent Vulcania benefited from the cartel.

Whatever its private reservations, Vulcania held out that it was a member of the cartel and by doing so aided its purpose. Repeated attendance at informal meetings – at least five or six on its own version – in conjunction with attendance at the alter ego, in the form of SAFRA executive meetings, only strengthened the assumptions the remaining firms could make that Vulcania was one of them. The incident with Cawood further reinforced perceptions that Vulcania was a willing partner.

European case law has long considered these defences offered by Vulcania and rejected them. Perhaps this is best set out in a decision of the European Court of Justice (ECJ) in the Aarlborg Portland AS case.⁵⁸ We quote what is a lengthy extract because the court’s approach is strikingly apposite to the facts of this matter.

“81.   According to settled case-law, it is sufficient for the Commission to show that the undertaking concerned participated in meetings at which anti-competitive agreements were concluded, without manifestly opposing them, to prove to the requisite standard that the undertaking participated in the cartel. Where participation in such meetings has been established, it is for that undertaking to put forward evidence to establish that its participation in those meetings was without any anti-competitive intention by demonstrating that it had indicated to its competitors that it was participating in those meetings in a spirit that was different from theirs (see Case C-199/92 P Hüls v Commission  [1999] ECR I-4287, paragraph 155, and Case C-49/92 P Commission v Anic  [1999] ECR I-4125, paragraph 96).

82.The reason underlying that principle of law is that, having participated in the meeting without publicly distancing itself from what was discussed, the undertaking has given the other participants to believe that it subscribed to what was decided there and would comply with it.

83. The principles established in the case-law cited at paragraph 81 of this judgment also apply to participation in the implementation of a single agreement. In order to establish that an undertaking has participated in such an agreement, the Commission must show that the undertaking intended to contribute by its own conduct to the common objectives pursued by all the participants and that it was aware of the actual conduct planned or put into effect by other undertakings in pursuit of the same objectives or that it could reasonably have foreseen it and that it was prepared to take the risk (Commission v Anic, paragraph 87).

84. In that regard, a party which tacitly approves of an unlawful initiative, without publicly distancing itself from its content or reporting it to the administrative authorities, effectively encourages the continuation of the infringement and compromises its discovery. That complicity constitutes a passive mode of participation in the infringement which is therefore capable of rendering the undertaking liable in the context of a single agreement.

85. Nor is the fact that an undertaking does not act on the outcome of a meeting having an anti-competitive purpose such as to relieve it of responsibility for the fact of its participation in a cartel, unless it has publicly distanced itself from what was agreed in the meeting (see Case C-291/98 P Sarrió v Commission  [2000] ECR I-9991, paragraph 50).

86. Neither is the fact that an undertaking has not taken part in all aspects of an anti-competitive scheme or that it played only a minor role in the aspects in which it did participate material to the establishment of the existence of an infringement on its part. Those factors must be taken into consideration only when the gravity of the infringement is assessed and if and when it comes to determining the fine (see, to that effect, Commission v Anic, paragraph 90).” ⁵⁹

17. This approach has been more recently followed in another ECJ decision, T-Mobile Nederlands, ⁶⁰ where it was held that a national court is required to apply the presumption that, where parties remain active in a particular market where the cartel conduct took place, such undertakings took account of the information exchanged with their competitors, unless the undertakings concerned adduced proof to the contrary.

It was further held in T-Mobile Nederlands that what matters is not so much the number of meetings held between the participating undertakings as whether the meeting or meetings which took place afforded them the opportunity to take account of the information exchanged with their competitors in order to determine their conduct in the market in question and knowingly substitute practical cooperation for the risks of competition. Where it can be established that such undertakings successfully concerted with one another and remained active on the market, they may justifiably be called upon to adduce evidence that that concerted action did not have any effect on their conduct in the market in question.⁶¹

18. The court stated also that an exchange of information between competitors is tainted with an anti competitive object if the exchange is capable of removing uncertainties concerning the intended conduct of the participating undertakings.

19. More recently the same approach has been followed by the European Union’s General Court in the Denki Kagaku case.⁶² Here the General Court held that;

“A party which tacitly approves of an unlawful initiative, without publicly distancing itself from its content or reporting it to the administrative authorities, effectively encourages the continuation of that agreement and compromises its discovery.”

20. In our law a contravention of section 4(1)(b) is sufficiently established if the firms concerned are proven to have entered into one of the agreements prohibited, even if the effects of that agreement are not established in respect of that firm. This is also the case in European law. Again in Denki Kagaku the Court stated:

“ According to settled case law, for the purpose of applying Article 81(1) EC, it is sufficient, in order for an agreement to fall within its scope, that the object of an agreement should be to restrict, prevent or distort competition, irrespective of the actual effects of that agreement. Consequently, in the case of agreements reached at meetings of competing undertakings, first that provision is infringed where those meetings have such an object and are thus intended to organise artificially the operation of the market and, second, the liability of a particular undertaking for participation in the infringement is properly established where it participated in those meetings with knowledge of their objective, even if it did not proceed to implement any of these measures agreed at those meetings.” ⁶³

21. This is also consistent with the approach we have taken in previous cartel cases. In Pioneer ⁶⁴ we observed that:

“The European Competition Commission has also stated that a party is guilty of participating in a cartel regardless of whether that participant actively participated in the day-to-day implementation of the agreement, a concept which is known as the concept of “a single overall agreement”. In the PVC decision⁶⁵ the European Commission found that all fifteen firms in the cartel were party to the agreement even though some had not attended every meeting or been involved in every decision made. Nor did the fact that some members intended to deviate from the cartel exclude them from the agreement. Their participation in the overall agreement was sufficient to establish their guilt. This position was upheld by the Court of First Instance.⁶⁶ The CFI in Trefileurope v Commission also held that the fact that an undertaking does not abide by the outcome of meetings which have a manifestly anti-competitive purpose does not relieve it of full responsibility for its participation in the cartel, if it has not publicly distanced itself from what was agreed in the meetings.⁶⁷ In Thuysen Steel v Commission the CFI found that attendance of meetings that involved anti-competitive activities was enough to establish participation in the cartel, in the absence of proof to the contrary.⁶⁸

22. Vulcania’s defence that it did not contravene section 4(1)(b) because it was passive during meetings and did not implement the agreements reached at the meetings is rejected.

23. The next aspect of Vulcania’s defence was that the discussions held at the meetings did not amount to agreements or understandings because they required ratification by another meeting that of the “senior people” to have effect. This defence, like the previous one, is bad in both fact and law.

It is bad on fact as there is no evidence from any witness other than Greve that agreements reached at the informal meetings required some committee of more senior people in the firms to acquire the status of an agreement.

Vulcania alleged that other witnesses called by the Commission supported this version in their witness statements.⁶⁹ They did not. Pierre Griffin, who dealt most extensively with the relationship between meetings of senior people and the other meetings, stated that senior management would be informed of the meetings and would ‘sanction’ them.⁷⁰ What Griffin was saying was that senior management permitted these meetings – that is quite different to saying that the agreements were subject to their subsequent approval.

Indeed the evidence of Di Nicola was that once the principals had set the process in motion he left it to Hankey to reach agreements. This version is also the more plausible. The ‘big’ men agreed on the need to co-operate and the operational people discussed the specifics. Given that Greve was both the ‘big’ man and the operational man in Vulcania, which is a much smaller firm than the others, there would have been little point in the cartel referring its decisions to others at a level where Vulcania was not represented. That was the very point of having Vulcania in the meeting-room. Nor is there evidence of report-backs from some other group or other individuals. No decision of the operational group was, on the evidence, ever vetoed or amended by a higher authority. Greve’s one example of a ‘report-back’ was nothing more than a report of the view of Rick Allen on a discount structure.⁷¹ The fact that Allen’s view was reported reflected the fact that he had not been at the relevant meeting (he was based in Cape Town and the cartel meetings took place in Gauteng). We do not see this as proof that the informal meetings could not take binding decisions.

24. Greve’s evidence on this aspect cannot be accepted.

25. Nor as a matter of law does it have substance. The Act makes it abundantly clear that agreements amongst firms do not require the formality of agreements in contract. The Act has extended the ordinary definition of the term agreement when used in relation to a prohibited practice to include “...a contract, arrangement or understanding, whether or not legally enforceable”. It is clear that the meetings arrived at ‘understandings’ or ‘arrangements’ between competitors on customer allocation to the effect that a member would not compete for another firm’s customers, as well as establishing the formula for adjusting selling prices to cater for increases in input costs. It is also clear from all the witnesses that the firms declared their intention to implement the decisions reached at these meetings, even though some ‘cheating’ occurred in practice.

As Greve himself conceded, Vulcania benefitted from the cartel arrangements in that other firms ‘respected’ its customers.

Asked what he meant when he referred to a customer as allocated to him (Vulcania) he explained:

“Apex who was my customer was allocated to me and what I mean by that is the other guys wouldn’t ...well, they said they didn’t ever go there and try and take the business away from me, from Apex.”⁷²

26. Greve here testifies to the existence of an understanding. This suffices for liability, even if ratification of some other individual had been required by some firms before they could implement the understandings. As a matter of fact we have found that there was no such requirement, and that agreements reached at the meetings were implemented without further ado, although they may be been reported to more senior people in the participants’ groups.

Vulcania is therefore liable in that it participated in meetings with competitors at which prices and discounts were set and customer allocation was agreed upon.

It remains for us to determine for what period Vulcania was a member of the cartel. As we noted earlier, Greve testified that he participated in the informal meetings after Vulcania joined SAFRA. The minutes show that this happened in February 2006.⁷³ It may well be that Vulcania joined the informal meetings earlier than this, as one of the documents purporting to show customer allocations was dated October 2005 and reflected customers of Vulcania. This is also consistent with the SAFRA minutes where it was recorded that Hankey was asked in August 2005 to approach Vulcania to join SAFRA. It is highly probable that Vulcania was attending meetings of SAFRA in late 2005 and that the customer allocation reflected discussions in which Greve had taken part and which had occurred by October 2005, at least.

This means that Vulcania was a member of the cartel for at least two years prior to its dissolution in 2008. As we cannot be certain of that date we will assume that this period was not much longer than two years.⁷⁴ For our purposes it suffices to conclude that it was at least two years.

We thus find that Vulcania has contravened sections 4(1)(b)(i) and 4(1)(b)(ii) of the Act from early 2006 to early 2008.

We discuss later the extent to which Vulcania’s late arrival in the cartel and limited participation should influence the remedy imposed upon it.

PART B

REMEDIES

27. In this case the Commission seeks as a remedy the imposition of a penalty of 10% of RMS’s and Vulcania’s respective annual turnovers. This is the maximum permissible under the Act.⁷⁵ Before discussing remedies we need to deal with two initial arguments which were raised as to why an administrative penalty, which is the remedy the Commission seeks in respect of both respondents, is inappropriate. We consider these two arguments respectively.

Vulcania’s initial legal argument:

An administrative penalty is not appropriate on the facts.

28. Vulcania argues that if we find it liable we ought not to impose a penalty as a remedy, given the limited circumstances of its involvement. Whilst it is correct that the imposition of an administrative penalty is discretionary, this is not an appropriate case for not imposing a penalty. Cartel contraventions, as we have noted before, are the most serious contravention of the Act and require appropriate deterrence.⁷⁶ Firms which attend meetings with competitors where collusion takes place must be sent a message that such activities are unlawful and are viewed very seriously. This does not mean that we will not consider a penalty that is proportionate to Vulcania’s limited degree of involvement in the cartel, and we go on to consider this later after considering the legal argument advanced by RMS.

RMS’ initial legal argument:

Firm has no turnover in relevant financial year

29. RMS contends that although it is liable in terms of the Act, a penalty cannot be levelled on it as it had no turnover in the relevant financial year which must be taken into account for assessing the cap on the penalty. This argument entails an interpretation of what year is contemplated in the relevant section that provides for the cap, namely section 59(2).

30. That section states:

“An administrative penalty imposed in terms of subsection(1) may not exceed 10% of the firm’s annual turnover in the Republic and its exports from the Republic during the firm’s preceding financial year.”

31. The crisp issue is what year is contemplated in this section as the firm’s ‘preceding financial year’. RMS contends that the relevant year is the year prior to the date of imposition of the fine by the Tribunal. Thus is if the Tribunal were to impose its fine on the 1^st August 2011 and the firm’s financial year ends 31 July, the “preceding financial year” would be the year ending 31 July 2011. Conversely, if the fine had been imposed one day earlier on 30 July 2011, the “preceding financial year” would be the year ending 31 July 2010.

The Commission initially argued that the relevant year for the purpose of determining what is the “preceding financial year” should be the last financial year during which there is evidence of the infringement having taken place.

But other candidates for the preceding financial year also exist. One might be the year preceding the date of the initiation of the complaint, and another the year preceding the date of the referral of the complaint.

Nevertheless, ordinary language lends some support to the approach that the relevant year is the year preceding the imposition of the penalty. The difficulty, of course, is to find what time or event the word ‘preceding’ refers to, and to that extent there is uncertainty or ambiguity or perhaps a lacuna in the language of section 59(2).

The only verb in the section is ‘impose’ and hence it may seem that the phrase ‘preceding financial year’, might qualify the term ‘impose’. Supporting this approach is an obiter dictum from the CAC in Southern Pipeline Contractors and Conrite Walls (Pty) Ltd v Competition Commission⁷⁷ (‘SPC’ case) where Davis JP stated:

“Although not strictly necessary for the determination of this case, I tend to accept the approach ... that a plain reading of section 59(2) supports a conclusion that the base year for the determination of the cap is the financial year preceding that in which the penalties are imposed.”⁷⁸

32. If this were so, it would follow that the contention of RMS is correct and RMS could not be subjected to an administrative penalty because in the relevant financial year it did not have a turnover - there is evidence before us that the company has been dormant for several years.

This would be a most unsatisfactory outcome as it would mean that no effective remedy can be imposed on a firm that does not trade in the financial year that pre-dates the year of imposition of the penalty. This case would be a perfect example. In our system, where penalties are imposed by the Tribunal and not the Commission, the prospect that a firm may engage in some restructuring to avoid a large penalty before the date of imposition is not remote. Several years may pass between the date on which a complaint is initiated (when the respondent first becomes aware of proceedings against it) and the date on which a penalty is imposed by the Tribunal.

In this case the Commission alleges RMS has done just that. When the Commission originally brought the case the second respondent was cited as Capital Africa Steel (Pty) Ltd (CAS), trading as RMS. In its answering affidavit CAS took the point that the wrong entity had been cited. According to Malcolm McCulloch, CAS’s executive chairman, the entity that ought to have been cited was RMS, its wholly owned subsidiary. He stated that the business of RMS was transferred to CAS on 30 April 2008. CAS, he stated, now conducts the business that was RMS, but was not the entity that participated in the conduct alleged.⁷⁹

In response to this the Commission amended its referral and substituted RMS for CAS as the second respondent. That happened prior to the commencement of the hearing. During the course of the hearing, while Di Nicola was being cross-examined about CAS by the Commission, counsel for RMS objected and stated that the cross-examination was irrelevant given that CAS was no longer a respondent.⁸⁰ In response after some debate the Commission sought an adjournment and during the adjournment period filed an amendment application to join CAS. CAS opposed this application, but on the eve of the hearing of the amendment application the Commission abandoned the joinder application, citing the recent CAC decision in Loungefoam as being against it as CAS was not the firm against which the complaint had been originally initiated. ⁸¹

As RMS was the only respondent against whom the Commission could proceed, the question then was, did RMS have turnover in any relevant year for the purpose of section 59(2), given that the case was being heard in late 2011, whilst RMS had ceased trading in April 2008?

The Commission alleges that the sale of assets to CAS was nothing but a contrived transaction by both firms to avoid liability for an administrative penalty under the Act at a time when CAS knew that the cartel was about to be exposed.

RMS rejects this. According to both RMS and CAS (the latter in the answering affidavit of McCulloch), the reason why the assets were transferred was that there was a change of shareholders in CAS. The company had previously been wholly owned by Wilson Bayley Holmes Ovcon Limited, which now disposed of half of the equity to Brait (40%) and Carlmac (10%). In the course of the restructuring certain subsidiary businesses were transferred to the parent company, RMS being one of them. CAS and RMS contend that at the time the transfer of ownership to CAS was made they had no knowledge of any legal action against RMS and that the restructuring was undertaken for purely commercial reasons relevant to group considerations at the time.

It is not necessary for us, when determining the question of the correct financial year for the imposition of the penalty, to decide whether CAS and RMS conspired to evade the provisions of the Act or whether the restructuring took place with no ulterior motive in order to meet genuine commercial strategic needs. But what these facts show is that an interpretation of the Act which means that in all circumstances the penalty can only be levied on turnover in the financial year before the date of imposition of the penalty is clearly susceptible to being flouted, hence making remedies under the Act nugatory. This would be an absurd result that would make a mockery of the Act. RMS suggested that the only remedy that could be imposed was a declaratory order to the effect that a contravention was found to have occurred, but that would be a worthless remedy and in fact an invitation to other firms who suspect that they are liable for large penalties under the Act to resort to the hasty disposal of assets. This would defeat the purposes of the Act. Even an interdict imposed on a firm that no longer trades would be pointless and ineffective, and would have the opposite of a deterrent effect

A better approach is the one favoured by the European authorities the Britannia Alloys case.⁸² Here the Commission imposed a fine in 2001 using the firm’s 1996, as opposed to 2001, turnover as the upper limit. The relevant European regulation required the cap to be based on the ‘preceding business year’ however the firm had not achieved any turnover in the year preceding the Commission decision. The Court held that the Commission was entitled to refer to another business year :

“... in order to be able to make a correct assessment of the financial resources of that undertaking and to ensure that the fine has a sufficient deterrent effect” ⁸³

The Court upheld the Court of First Instance’s (C.F.I.)( the predecessor of the General Court) which explained that the calculation of the upper limit presupposes that the Commission not only has at its disposal the turnover for the preceding financial year but also those figures that represent a full year of what it termed “...normal economic activity over a period of 12 months.”

Endorsing this approach of the CFI the ECJ held that “ ...in certain situations, the turnover of the business year preceding the adoption of the Commission decision does not provide any useful indication as to the actual economic situation of the undertaking concerned and the appropriate level of fine to impose on that undertaking.” ⁸⁴

33. The respondents argued that the European Courts have greater latitude than the Tribunal does, to adopt a purposive approach, and that when ordinary meaning is clear we cannot depart from the language of the statute. The first part of that proposition may or may not be true, but the proposition as a whole is undercut by the fact that the language of section 59(2) cannot with candour be described as clear. Some interpretive input is needed for it to have rationality. But in any case our courts and others do from time to time, when the ordinary canons of interpretation of statutes and other documents fail them, resort to a purposive approach to get to the mischief that is sought to be addressed by a statute or other legal document in which the language is not clear.

Indeed our courts have been as robust as the European courts were in Britannia in applying a purposive approach when appropriate to avoid an absurdity. This approach was recently followed in a tax decision in Commissioner, South African Revenue Services (SARS) v Multichoice Africa (Pty) Ltd and another,⁸⁵ where the court referred to the well known tests adopted in earlier cases.

“In Venter v R⁸⁶ Innes CJ held that a court may depart from the ordinary meaning of the plain words of a statute where to give effect thereto 'would lead to absurdity so glaring that it could never have been contemplated by the legislature'.⁸⁷ In a separate, concurring judgment Solomon J held that departure from the ordinary meaning of plain words in a statute is warranted if the result of a literal interpretation would be 'something which is repugnant to the intention of the legislature.”⁸⁸

34. In Jaga v Donges NO and another ⁸⁹ the court held that there must not be excessive peering at the language to be interpreted without sufficient attention to the contextual scene.

In our view the clear purpose of section 59(2) of the Competition Act is to prevent a penalty being levied which is disproportionate in relation to the size of the firm concerned and hence the cap imposed on turnover. Since however the preceding financial year may not always reflect the firm’s ‘real economic situation’, as the ECJ referred to it, it is permissible to rely on another year which reflects the firm’s turnover in a year of ‘normal economic activity.’⁹⁰

35. We thus find that in normal circumstances, when a contravention is long-standing (e.g. of some years’ duration) and the complaint procedures leading up to a hearing in the Tribunal have not suffered undue delay, the meaning to be given to section 59(2) is that ‘the preceding financial year’ mentioned in that section, refers to the last completed financial year before the imposition of the fine. Where there is for any reason no turnover in that year or where because of some financial engineering there is minimal turnover not reflective of the ordinary business activity of the respondent firm during the course of the contravention, the Tribunal may have regard to the earliest preceding year of normal turnover, to avoid an absurdity or an outcome that would be repugnant to the intention of the legislature.

In the case of RMS this would be the financial year ending June 2007 when RMS had a turnover of R 363,7 million. We find that this amount may be taken into account for the purpose of determining the cap referred to in section 59(2). There is thus turnover to which a cap on a penalty can apply and as we consider a penalty appropriate (there was no dispute on this point) we now go on to consider how the amount of the penalty should be determined.

APPROACH TO PENALTIES

36. We now consider the factors we must take into account in imposing administrative penalties on RMS and Vulcania in terms of section 59(3) of the Act.

We have previously recommended in SPC⁹¹ that the Commission, like its European counterpart, adopt guidelines to its approach to the imposition of penalties. At the time of deciding this case no such guidelines had been published. The CAC in SPC has recommended having regard to the EU guidelines published in 2006.⁹²

We have done so in this decision, but adapting some features of the approach to meet the requirements of our Act. The CAC had also noted in SPC that the factors taken into account in the EU guidelines do not correspond identically to the factors set out in section 59(3). For this reason, whilst influenced by the EU approach, we have adapted it to suit our legal framework. Where we have done so, we indicate the difference in approach and the reason for doing so. There are some new aspects to our approach in this decision but we have also followed past decisions.⁹³ Our approach requires six steps. We discuss each in turn. In summary, they are:

Step one: determination of the affected turnover in the relevant year of assessment.

Step two: calculation of the ‘base amount,’ being that proportion of the relevant turnover relied upon.

Step three: where the contravention exceeds one year, multiplying the amount obtained in step 2 by the duration of the contravention.

Step four: rounding off the figure obtained in step 3, if it exceeds the cap provided for by section 59(2).

Step five: considering factors that might mitigate or aggravate the amount reached in step 4, by way of a discount or premium expressed as a percentage of that amount that is either subtracted from or added to it.

Step six: rounding off this amount if it exceeds the cap provided for in section 59(2). If it does, it must be adjusted downwards so that it does not exceed the cap, as explained by the CAC in SPC.

Step one

37. Like the EU we first determine what the affected turnover is. The affected turnover of the respondent firm is based on sales of the products or services that can be said to have been affected by the contravention. The European authorities point out that an affected turnover approach is not to be conflated with a relevant market approach – the former does not demand the rigour or precision required of the latter.⁹⁴

The year selected for this purpose is the last financial year of the period for which we have evidence that the cartel existed.

Where contraventions occurred over several years, as they frequently do, there is the temptation to take the affected turnover for each year of contravention into account.

Whilst this approach might, superficially, seem the more accurate one, it loses sight of the purpose of the overall determination which is to find a proxy for the harm done, not an exact calculation of profits or damages. As the authors of “Economics for Competition Law” have noted, economic theory identifies two possible reference points to determine the optimal level of the penalty: harm to society caused by the contravention, and the illicit gains made by the perpetrator. In practice the difference is not be a matter on which adjudicators need dwell. Penalties set with reference to value of sales and duration serve as a adequate proxy for the economic importance of the infringement. As they put it:

“The term economic importance can be interpreted as importance to either the economy as a whole or to the perpetrators. ... But the Guidelines stop short of actually trying to measure either the harm to the economy or the illicit gain. Instead they apply a number of rules to approximate these effects...” ⁹⁵

38. Since equivalence between these two amounts is in most cases unlikely – an inefficient monopolist or cartelist may raise prices above the competitive level, but not see the same amounts translate into net profits -- the affected transaction figure is merely a rough figure that serves as a guide, with some, but by no means perfect correlation to harm caused or benefits gained.⁹⁶ Note that in approaching a base year we are using it as base to build a penalty, not assess damages. Considering each year in a multiyear cartel would entail lengthy and futile sub-enquiries into each year’s accounting intricacies. Further, there comes a time when duration may in the case of long cartels be rounded off to the benefit of the respondent. An approach that required each year to be taken into account would not allow this. Note that the Act does not set out a formula for how duration is to be taken into account – this is a matter of discretion.

Furthermore, the effects of the prohibited practice may not have ended by the time the case is initiated or indeed even by its conclusion. Hence again a precise assessment of affected turnover can never be a figure that perfectly replicates the harm caused or illicit gains made. Finding a base year therefore serves as a proxy for calculating a penalty. By adopting a consistent policy regarding the year in which the turnover should be selected – the final complete year of the cartel for which there is evidence – a measure of certainty can be created.

Step Two

39. Once we have ascertained what makes up the affected turnover, the next step is to calculate the ‘basic amount’ which then becomes the building block for the penalty calculation. Determining the basic amount is again an exercise to be performed with discretion. In the EU the basic amount is expressed as a proportion up to a maximum of 30% of the affected turnover.⁹⁷ The more egregious the conduct, the more likely the basic amount will tend towards the maximum. The factors taken into consideration in determining this basic amount are the nature of the contravention, the combined market share of the undertakings concerned, the geographic scope, and whether or not the contravention has been implemented.⁹⁸

In cartel cases our approach to the assessment of the basic amount (i.e. step 2), is to examine the effect of the cartel as a whole. When we get to step 5, we examine the circumstances of the individual firms. Thus two firms who are respondents in the same cartel case would be treated in the same way in step two for the purpose of calculating the percentage to be applied to their respective turnovers to derive the basic amount; but might not be treated identically in step 5, as the individual circumstances of the firms may be different.

Section 59(3) of the Act, which sets out the factors to be taken into account in determining penalties, provides as follows:

“When determining an appropriate penalty, the Competition Tribunal must consider the following factors:

the nature, duration, gravity and extent of the contravention;

any loss or damage suffered as a result of the contravention;

the behaviour of the respondent;

the market circumstances in which the contravention took place;

the level of profit derived from the contravention;

the degree to which the respondent has co-operated with the Competition Commission and the Competition Tribunal; and

whether the respondent has previously been found in contravention of this Act.”

40. Without making the list exhaustive, the factors referred to in section 59(3) to be considered in step 2 would be the nature, gravity and extent of the contravention (corresponding to section 59(3)(a)), any loss or damage suffered as a result of the contravention (section 59(3)(b)) and market circumstances in which the contravention took place (section 59(3)(d)). Later, in step 5, we look at sub-paragraphs (c),(f) and (g) of section 59(3), all of which refer to the respondent specifically, as we noted above.

Although the level of profit derived from the contravention, which is a factor to be taken into account as set out in section 59(3)(e), does not refer to the respondent expressly, unlike sub-paragraphs (c),(f) and (g), such a reference is certainly there by implication, which means that this factor could be examined at either stage 2 or 5, depending on whether the facts were common to all firms in the cartel or differed from firm to firm. Thus we must avoid approaching these steps in a rigid manner. Where firms in a cartel merit different treatment in terms of the factors taken into account in step 2, that difference may be taken into account in step 5. On the facts of this case such an approach has not been necessary.

Whilst some factors in section 59(3) are self-evident, others such as the ‘nature’, ‘gravity’ and ‘extent’ of the contravention need more elaboration and would depend on the type of contravention being considered.

Because the factors set out in section 59(3) apply to all contraventions listed in section 59(1) and not simply contraventions of section 4(1)(b)(i) it follows that the application of the factors to different forms of contravention will vary. Some contraventions concern abuse of dominance and resale price maintenance, others concern contraventions of merger regulation provisions and still others are a species of civil contempt in that they apply to breaches of the Tribunal’s orders. ⁹⁹ This means that some factors will loom larger than other factors in assessing a penalty, depending on the nature of contravention.

When we consider ‘nature’, ‘gravity’ and ‘extent’ in the context of section 4(1)(b) contraventions, we have regard to the fact that this was cartel conduct and the nature of the cartel. Cartel conduct is considered a more egregious form of contravention than resale price maintenance.¹⁰⁰ When considering extent and gravity, if it was a cartel, the extent of the cartel – was it national or regional – how much market share did it account for, was it consistent or sporadic in nature, are some factors to be taken into account. The answers to these enquiries then inform an approach to the proportion of the affected turnover which constitutes the basic amount, which would lie somewhere on the continuum between zero and 30%.

Step three

41. Duration is provided for in the next step by multiplying the basic amount by the number of years the contravention lasted. In this way proportionality is recognised: long-lasting contraventions, which are more harmful, are more heavily fined than contraventions of a shorter duration. This is the approach taken in the EU and was also the approach followed by the CAC in the SPC case.

Step four

42. Here we introduce a further adjustment not provided for in the EU Guidelines nor did it need to be considered by the CAC in SPC, but it does arise on the facts of this case.

If the cap on 10% of total annual turnover provided for in section 59(2) is already exceeded at this stage, it means that adjusting the penalty for firm-specific factors in mitigation or aggravation may prove to be pointless since the adjustments may leave the final amount above the cap. This outcome would offend against the principle of proportionality because considerations that must be taken into account in determining the penalty are rendered moot. Thus if a high proportion of a firm’s total turnover constitutes affected turnover, if the contravention is a serious one but it is of only medium duration – say two years or somewhat more - there is a considerable likelihood that the amount calculated at this stage will already exceed 10% of total turnover.

The reason why this outcome is of less concern in the EU than in South Africa could be the result of two factors that distinguish the EU’s approach to penalties from that of South Africa. Firstly, in the EU the cap is based on world-wide, not EU- wide, turnover, whilst the affected turnover from which the basic figure is derived is an EU turnover figure.¹⁰¹ Furthermore, the EU will in appropriate cases, fine parent companies for contraventions by controlled subsidiaries. Parent companies will generally have a greater total turnover than the affected turnover of a subsidiary. Thus on both approaches the world-wide turnover taken into account for the cap is generally likely to exceed the basic turnover and hence “hitting the cap” before mitigating and aggravating factors are considered, does not normally arise in practice. ¹⁰²

For this reason we must approach our statute differently to avoid this problem. Hence we have considered it appropriate to apply an adjustment to the basic amount already at this stage of the enquiry.

Step Five

43. Finally, once this basic amount has been calculated by following the steps followed above, it is adjusted downwards or upwards by a percentage that depends on the balance of mitigating or aggravating factors present. In making this assessment we consider the remaining factors set out in section 59(3), namely sub-paragraphs (c), (e), (f) and (g). Whilst not all these factors are relevant in each case and some may be neutral, we nevertheless have regard to them in the assessment. Of course some factors may indicate aggravation and others mitigation at the same time. The final assessment requires both to be netted off, with the result that there will be either an increase or a decrease in the basic amount.

Step Six

44. Once we have applied step 5, we then ensure (again) that the amount arrived at does not exceed the statutory cap set out in section 59(2), namely 10% of the firm’s total turnover in the preceding financial year. We dealt above with the question of what that year is, but note that this does not necessarily mean that it is the same financial year as the one used to calculate the affected turnover in step 1.

APPLICATION OF SECTION 59(3) FACTORS TO VULCANIA AND RMS

We first consider those factors that are common to both firms and then consider the firms individually.

Nature, gravity and extent (Vulcania and RMS)

45. Cartel activity is viewed by competition authorities as the most egregious of all contraventions. Under the Act contraventions of section 4(1)(b) allow no defence of justification. For this reason the penalty for a contravention of section 4(1)(b) has to have a serious deterrent effect.

The cartel itself appears to have operated nationally although we accept that Vulcania’s and RMS’ presence would have been limited to a more local effect; probably within the greater Gauteng region.

The evidence of Hartnady is that not all the firms involved in the industry were members of the cartel. This is corroborated by SAFRA’s minutes and informal meeting notes which indicate the presence of other firms which were in the industry but which were not part of the cartel. We can thus assume that the cartel was more effective in the Gauteng region but less effective nationally.

The gravity of Vulcania’s and RMS’s conduct is less severe than that of Steeledale and BRC. They were not founding members of the cartel and therefore we can infer that the cartel could and did operate without them for some years. In this respect the evidence from the SAFRA minutes we considered earlier indicated that the other cartel members exhibited an ambivalent attitude toward having them as members. Whilst at one stage wanting to court new members, SAFRA minutes show that later its members were reluctant because of ‘market circumstances’ to bring in new members. Still later they encouraged them.

Loss or damage suffered as a result of the contravention and level of profit derived from the contravention (Vulcania and RMS)

46. Loss or damage is difficult to prove in cartel cases where the counter-factual – what the market might have looked like under competitive conditions -- is not available. In this case, given the long duration of the cartel, the difficulty of constructing the counter-factual was compounded, and no effort was made to present it to us.

It might also have been useful to have evidence of what happened to prices after the existence of the cartel became known.

Although the Commission applied for a postponement to lead such evidence we did not grant the postponement for reasons we discuss later in this decision. This means that we must accept Vulcania’s and RMS’s version that the cartel arrangements were not lucrative and that the firms involved did not make undue profits as a result of its existence.

There is also evidence that rebar was considered a substitute for customers seeking an alternative form of reinforcement to wire mesh. To some extent the installed price of rebar would have acted as a price ceiling on the installed price of wire mesh.

Wire mesh is a commodity product where input prices comprise a very high proportion of the final price (some witnesses stated up to 80%). Whilst the cartel endeavoured to keep margins above the level that would have prevailed in a competitive market, it is unlikely that the excess would have been more than 10%.

Absent evidence from the Commission to the contrary, we will accept that the damage suffered by customers and the excess profits gained from the cartel by its members were not inordinate. That is not to say that cartels constitute harmless behaviour. Firms do not collude for the sake of collusion; they do so because the benefits of collusion are calculated to be better than the benefits of competition between them.

Market circumstances of Vulcania and RMS

47. Both respondents were intermediaries between the selling power of the steel mills and the buying power of some of their customers. That circumstance does not excuse collusion, particularly when other firms managed to expand in the market without joining the cartel, as Barnes apparently did. However it does provide some context as to the financial pressures which prevailed at that time in the mesh reinforcement sector. We discuss later when we deal with the firms individually the particular pressures experienced by them as opposed to other members of the cartel.

Vulcania

Duration (Vulcania)

48. Vulcania’s evidence is that its involvement in attending cartel meetings lasted from February 2006 to February 2008. Although it is possible this period may have been longer, as we discussed earlier in the section on liability, the Commission does not contest this. We can therefore accept and there is no dispute about this that it was involved for at least two years.

Behaviour of the respondent (Vulcania)

49. It is common cause that Vulcania entered the market with certain disadvantages and that until it had overcome these, it was limited to serving markets that were not of great interest to the major players. Vulcania’s main line of defence is that it joined the informal meetings at a time when it was vulnerable in the industry. It did not have access to supplies of rod in coil from a steel mill and lacked the equipment necessary to draw rod in order to become a low-cost producer. Vulnerable to exclusion from the market, it succumbed to pressure to join the cartel and attend its informal meetings. Once it had a secured supply of rod in coil and had acquired and installed equipment to produce mesh from rod in coil, it alleges it ceased attending the informal meetings.

Whilst this factor of perceived coercion will be taken into account in mitigation, its extent as a mitigating factor is limited. All new firms face difficulties entering markets and resistance from incumbents is one of them. This does not justify joining a cartel.

Vulcania’s second plank of its defence is that it was passive at the five or six meetings that it attended. This version is contradicted by the Commission’s witnesses and in particular the incident with Cawood. Whatever Greve’s personal reservations were, he did not express his reluctance to the others nor did they have any other reason to perceive it. On the contrary, the other firms, having brought Vulcania into the conspiracy, presumably because they considered it enough of a threat, had every reason to believe that it would not be a disruptive force in the market. On the contrary they acted on the assumption that it would abide by the customer allocations that had been agreed upon, given its regular attendance at meetings during this period and its acquiescence at those meetings regarding matters of customer allocation.

We accept however that Vulcania played a minor and less active role in these meetings than did other firms. There is no evidence that it initiated any of the cartel’s activities.¹⁰³

Another important aspect of the Vulcania’s defence is that it quietly defied cartel decisions rather than implementing them. There is again no evidence to counter Greve’s testimony that it did not implement the decisions.

Importantly for Vulcania, its assertion that it did not respect the cartel arrangements was supported by witnesses called by the Commission. Hartnady in his evidence in chief said that certain players did not regard the market allocation list “...as a serious document” and he expressly included Vulcania in this category.¹⁰⁴

There is also some contemporaneous supporting documentary evidence. In a Steeledale marketing plan document dated March 2007, Steeledale remarked that certain competitors’ market share had grown because of alliances with their former franchisees. The competitors mentioned were RMS and Vulcania. It was also mentioned that the entry of those two firms into the Specimesh market had had an adverse impact on Steeledale’s growth and that price levels were lower than anticipated.¹⁰⁵

However even if Vulcania sold to customers allocated by the cartel to other members this does not seem to have been on a scale significant enough to attract the attention or at least the wrath of the other members. There seems to have been no manifest disruption of the customer allocations and no upsetting of the cartel’s prices. There is no evidence that the other firms punished Vulcania for non-compliance. Nor did Greve testify that other cartel members ever competed for Vulcania’s customers. As he put it:

“We were under the radar and we grew our business and we managed to not upset them.”¹⁰⁶

50. Whilst we have some regard to the mitigating fact that Vulcania did not implement, or at least fully implement, the cartel’s agreements, Vulcania’s conduct did not seriously impede the cartel arrangements. This factor accordingly constitutes only minor mitigation.

Another plank to Vulcania’s mitigation argument is that it withdrew from the cartel as soon as it was self-sufficient in its supplies of rod in coil.

However Greve has given inconsistent evidence on this point. In his witness statement he stated that as soon as Vulcania became self-sufficient it ceased attending meetings.¹⁰⁷ In his evidence in chief, he, unsolicited, suggested that the last meeting he attended was the last of the cartel’s informal meetings. The implication of this answer is that Vulcania did not withdraw from the cartel prior to its demise – the cartel ended of its own accord when a leniency application was made. ¹⁰⁸

Later, when responding to a question from Vulcania’s counsel whether he went to any further meetings, Greve’s answer, contradicting the prior one, was that there would have been no point: it would have been a waste of time as “...we could stand on our own two feet at any stage.”¹⁰⁹ This answer suggests that his last meeting was not the cartel’s last meeting, but that Vulcania did withdraw prior to the dissolution of the cartel, albeit at a very late hour.¹¹⁰

Fortunately for Vulcania the Commission has not challenged Greve’s evidence on this point. We will therefore accept that despite this inconsistency in his oral testimony, the version given by Greve in his witness statement is the correct one and that Vulcania withdrew from cartel meetings prior to the end of the informal meetings, because its supply was now secure and it was no longer vulnerable to exclusion by its rivals . We recognise this as an important mitigating factor.

To offset this mitigating behaviour there is one aggravating feature. This is that Greve, the most senior person in the company, partook in cartel activities. The Office of Fair Trading in the United Kingdom takes into account, in determining penalties, whether senior management participated in the cartel meetings and it lists such participation as an aggravating factor in determining the appropriate penalty amount for cartel activity.¹¹¹ There is no reason not to follow such an approach in our law. In Vulcania’s case, Greve was the chief executive officer at the relevant time and this makes its participation in the meetings more serious.

Level of profitability of Vulcania

51. Little is known to the Tribunal on this topic. Vulcania asserted that it did not derive any benefit from the cartel arrangement. The Commission argued that Vulcania benefited from not having the other members of the cartel compete for its customers, meaning that Vulcania did not have to lower its prices when selling to its customers. Vulcania did not give evidence of why its level of profitability might have been lower than that of the other cartel members nor did the Commission show that profits were higher than they might have been in a competitive market.

Degree to which Vulcania has co-operated with the Commission and Tribunal

Vulcania has shown no special co-operation in this case. It has not offered information beyond that which the law required it to give. On the other hand, the fact that it has defended itself in these proceedings in the limited fashion that has been outlined above cannot be considered as an aggravating factor, and the Commission’s suggestion to that effect must be rejected. This factor is therefore a neutral one, neither aggravating nor mitigating Vulcania’s conduct.

Whether the respondent had previously been found in contravention of the Act. (Vulcania)

52. Vulcania has not previously been found to have contravened the Act.

RMS

Duration

53. We have found that RMS took part in the cartel for at least four years. This is not seriously contested by RMS.

Conduct of RMS

54. In assessing RMS’s conduct we have to take into account both aggravating and mitigating factors.

The first aggravating aspect to RMS’s conduct is that despite the fact that SAFRA members were warned in 2005 by Mr McDonald, an attorney, that these activities were illegal, RMS, along with the other cartel members, continued with them, and moreover moved the price discussions into informal meetings to keep them secret. RMS was aware of McDonald’s warning as it was a member of SAFRA at the time. (We cannot make the some assumption about Vulcania, which only joined SAFRA in February 2006 and may not have known about McDonald’s advice.)

Another aggravating feature of RMS’s conduct was the involvement of senior management. RMS commenced involvement when its chairman, Di Nicola, was approached by others to attend meetings where inter alia market division was to be discussed, and consented to this. Hankey, the chief executive officer, later joined discussions at the informal meetings. Hankey was both a shareholder and director of the RMS business.¹¹²

Thus RMS was party to discussions on market division, customer allocation and price setting. Its involvement during this period was as a full participant in the full range of the cartel’s activities.

However there is also some evidence that mitigates RMS’s conduct. First, as with Vulcania, there was some evidence of coercion. Whilst the coercion was not as strong as that applied to Vulcania there was pressure from the other cartel members to join the cartel at a time when RMS’s supply arrangements were not yet secure and RMS had difficulty in obtaining the required equipment from Clifford Engineering.

More persuasively in RMS’s favour is that it attempted to disrupt the cartel both before it joined and later as a member. RMS’s own evidence to that effect was supported by contemporaneous sources from other parties.

An internal Steeledale marketing document dated February 2005 records the following:

“The anticipated impact of Reinforcing and Mesh Solutions entering the market has been grossly under-estimated with tonnages being lost to them particularly via the steel merchants and large contractors. The company is at present on a major national expansion drive, buying market share in most regions.”¹¹³

55. There is also evidence that RMS was punished for undercutting the cartel and was forced to make amends. Hankey testified that he had sold mesh to a customer of Aveng and he was forced to buy the same volume from Steeledale at the list price without a discount.¹¹⁴

Level of profitability of RMS

56. RMS stated that at its highest its profitability was 10%. Therefore it did not excessively profit and hence the damage caused by its participation in the cartel was correspondingly limited.

Degree to which RMS has co-operated with the Commission and Tribunal

57. Like Vulcania, RMS has shown no special co-operation in this case. It has not offered information beyond that which the law required it to give. Information that it gave the Commission to the effect that that RMS and not CAS was the offending party, has proved self-serving. Again, the fact that it has defended itself in these proceedings in the limited fashion that it has, cannot be considered as an aggravating factor, contrary to the Commission’s suggestion. This factor is therefore a neutral one and neither aggravates nor mitigates the contravention for the purposes of determining the penalty to be imposed.

Whether the respondent had previously been found in contravention of the Act.

58. RMS has not been found to have previously contravened the Act.

PENALTY CALCULATIONS

Penalty calculation in respect of Vulcania

59. Vulcania asserted that the last full financial year in which it participated in the cartel was the financial year ending 31 December 2007. We know that Vulcania continued to participate in the cartel for some period during 2008, but this ended in January of that year. Therefore we accept the calendar year 2007 as the correct year for the purpose of assessing the penalty.¹¹⁵ Vulcania has provided figures to the Tribunal indicating that its turnover for reinforced wire mesh for the financial year ending February 2007 was R 31,6 million rand.¹¹⁶

If the maximum proportion for this calculation by the EU is used as a guideline (recall that the EU applies a maximum penalty of 30% in the most serious cases) we would regard this contravention as one for which 15% would be an appropriate figure. This gives a basic figure of R4,74 million (31,6x15%). This figure of 4,74 is then multiplied by 2, to provide for the two-year length of Vulcania’s participation. This leads to a figure of R9,48 million.

However we have accepted the reasoning that Vulcania was not an instigator, that it was in some respects coerced to join, that it left of its own accord before the cartel was detected, that it profited little and caused little damage by its participation, that it refrained from implementing cartel decisions, and that at times it may have disrupted the cartel’s effectiveness. Cumulatively these factors mean that this penalty can be substantially reduced. The one aggravating feature in its conduct is the presence of its most senior manager at the cartel meetings. For these reasons the mitigating factors set out in section 59(3) lead to a reduction of 40% in the basic amount. This gives a final figure, as rounded off, of R5,6 million.

This amount does not exceed the 10% annual turnover of Vulcania for either 2010 or 2011 and therefore does not need to be rounded down in order to fall below the cap.¹¹⁷

Vulcania ran at a loss in 2010 although it was profitable for the 2011 financial year. Given this financial background we shall allow 50% of the fine be paid within 60 business days of the date this decision and the balance not more than a year later.

Penalty calculation in respect of RMS

60. The relevant year for RMS will be its financial year ending June 2007 as this represents the last complete year of cartel behaviour, given that the cartel is alleged to have ended in early 2008. The figure for RMS’s turnover in wire mesh was given by the Commission as R62 million. The Commission distinguished between turnover for mesh and hard drawn wire. As we are not certain from the submission if hard drawn wire was part of the affected turnover we shall give RMS the benefit of the doubt and restrict the affected turnover figure to R62 million.

We apply the base percentage of 15% to this figure. This is the same as was done for Vulcania as its circumstances are the same in this respect. This gives a figure of R9,3 million. Multiplying this amount by four to take account of the duration of RMS’s participation in the cartel, takes us to a figure of R37,2 million.

This amount is in excess of 10% of the turnover for the 2007 financial year, which was R363 million. A proportion of 10% of this amount is R36 million. We therefore round down to R36 million.

Taking into account both mitigating and aggravating features we consider that on balance, the mitigating evidence, particularly the evidence that RMS disrupted the cartel, both prior to joining the cartel and during its period of involvement, outweighs the aggravating evidence. We find that this would justify reducing the basic amount by 40%, leading to a penalty of R21,6 million. This is the same discount as applied to Vulcania. In RMS’s case although there is more aggravating evidence present, the mitigating evidence of disruption is much stronger; hence on balance it receives the same discount.

The application for postponement

On the 6^th of August 2011 the Commission applied for a postponement¹¹⁸ of the matter in order for it to be allowed to investigate certain matters that relate to the level of the penalty to be imposed in terms of section 59.

The application had followed the release of the CAC decision in SPC. The Commission argued that the decision meant it had to lead evidence it was otherwise from past practice not expected to lead, concerning certain factors set out in section 59(3), in particular evidence concerning loss or damage as a result of the contravention and the level of profit derived from the contravention. The Commission argued that in the past such evidence had been presumed flowing from a finding of liability and the Commission was not required to lead any further evidence on these issues.

61. The application was opposed by both respondents. We decided to refuse the postponement.

62. The application was made late in the day in what have been already protracted proceedings, at a time when all parties were ready to present final argument. The respondents would have been severely prejudiced from further delay which could not have been remedied by a costs order given that we do not make costs orders in prohibited practice cases brought by the Commission.

Secondly, it was not clear whether the Commission would obtain evidence that it intended to lead. Its application was to enable it to further investigate to see if it could obtain such evidence. Thus we had no confidence that the postponement would yield any new evidence on these issues. Nor do we accept that the approach of the CAC was novel or should have come as a surprise. If the Commission was able to obtain such evidence during its investigation of the cartel it should have done so.

Thirdly, the respondents are, as we have noted, the smallest players in this cartel. They stand the most to lose by a further postponement, whilst given their lack of centrality to the cartel, the administration of justice has little to gain by postponing proceedings against them further.

63. For these reasons we consider that the application for postponement was not well founded and it was refused.

ORDER

1. Vulcania is found to have contravened section 4(1)(b)(i) and 4(1)(b)(ii) of the Act for a period of two years from January 2006 to January 2008;

2. Vulcania is ordered to pay an administrative penalty of R5,6 million. Of this amount, R2,8 million is to be paid within 60 business days of the date of this order, and the balance within one year of the date of the first payment, but no later than one year and 60 days after the date of this order .

3. RMS is found to have contravened section 4(1)(b)(i) and 4(1)(b)(ii) of the Act for a period of four years from January 2004 to January 2008;

4. RMS is ordered to pay an administrative penalty of R21,6 million. Of this amount, R10,8 million is to be paid within 60 business days of the date of this order, the balance to be paid within one year of the date of the first payment but no later than one year and 60 days after the date of this order.

5. There is no order as to costs.

____________________ 7 May 2012

Norman Manoim Date

Lawrence Reyburn and Medi Mokuena concurring.

Tribunal Researcher: Songezo Ralarala

For the Applicant / Commission: NH Maenetje SC and DG Ngcangisa instructed by the State Attorney.

For the 2^nd Respondent: MJ Engelbrecht instructed by N Altini of Cliffe Dekker Hofmeyr Attorneys.

For the 3^rd Respondent: JPV McNally SC instructed by C Artemides of Christelis Artemides Attorneys.

---