CT CR105Mar05 · case CR105Mar05
competitigntribunal edt af Pisa COMPETITION TRIBUNAL OF SOUTH AFRICA Case No: 17/CR/Mar05 In the matter between: Competition Commission Applicant Tracetec (Pty) Ltd Intervenor/ Complainant And Netstar (Pty) Ltd First Respondent Matrix Vehicle Tracking (Pty) Ltd Second Respondent Tracker Network (Pty) Ltd Third Respondent Vehicle Security Association of South Africa Fourth Respondent Panel Norman Manoim (Presiding Member), Yasmin Carrim (Tribunal Member), and Lawrence Reyburn (Tribunal Member) Heard on : 11-241 November 2008; 14-17 September 2009; and 29-30 September 2009 Order Issued : 19 April 2010 Reasons Issued: 19 April 2010 Reasons for Decision INTRODUCTION
[1]This case concerns whether standards set by an industry association created barriers to entry that prevented competitors of members of the association from competing in the market. The case has been brought by both the Competition Commission (“the Commission’) and the complainant and intervenor in the
[2]matter, a firm called Tracetec (Pty) Lid (“Tracetec”). The relief sought is purely declaratory. The four respondents in this matter all oppose the relief sought. The first three respondents Netstar (Pty) Ltd (“Netstar’), Matrix Vehicle Tracking (Pty) Ltd (“Matrix”) and Tracker Network (Pty) Ltd (“Tracker”) are firms engaged in the stolen vehicle recovery market (‘SVR’). We refer to them collectively as the “SVR respondents”. The fourth, the Vehicle Security Association of South Africa (‘“VESA’), is an industry association for firms engaged in the vehicle security industry that at the relevant time had a sub-committee that set standards for admission to membership of its SVR category.' PART A: BACKGROUND.
[3]
[4]
[5]In the late 1980’s, South African motorists began to experience unprecedented vehicle theft. In consequence, insurers began to look for means to minimise risk. Up till then security strategies had been aimed at inhibiting theft through devices such as gear locks and car alarms, but these were no longer considered sufficient. Products started being developed which their makers boasted could track vehicles after they were stolen and enable them to be recovered. SVR products were introduced into the market in the early 1990’s.* The products varied, but essentially they claimed to enable a vehicle which had the equipment installed, to be tracked and as an essential ancillary service to this, to be recovered either though the efforts of the tracking company concerned or some third party contracted for this purpose. Originally intended for luxury vehicles their usage was extended to less expensive vehicles. Insurance practice was not consistent, but insured motorists were either told to install an SVR device as a condition of insurance or were incentivised to do so by lower premiums. The insurance industry believed that SVR devices improved recovery rates drastically. One witness testified that prior to the introduction of SVRs recovery rates averaged 20% of ' The Commission initially also referred the complaint against two other firms, Bandit Limited and Global Telematics SA (Pty) Ltd, but withdrew against them on 15 November 2005. ? Netstar answering affidavit paragraph 7, page 186. The deponent is John Edmeston.
[6]
[7]
[8]
[9]stolen vehicles. Once tracking devices had been fitted, the recovery rate rose to 70%. * This created a market for SVR products as distinct from other vehicle security devices which could not track and recover vehicles. It is common cause in this case that the relevant market is a national market for SVR products. For this reason, beyond the observation made above, it is unnecessary for us to consider this issue any further. In the nascent stages of the SVR market, there were only two major players, Netstar and Tracker, and then a number of fringe players whose market shares at the time are not known. No standards existed in the industry and the market was a free for all. Two institutions did not like the /aissez faire manner in which the market was playing itself out. The first was VESA, the fourth respondent. VESA was formed in 1987. Its membership comprises firms involved in the vehicle security industry and includes manufacturers, suppliers, installers and maintenance firms.’ Firms are competitors of one another and are organised into sub-committees relevant to their market niche. The range of products represented varied from gear and wheel locks, to sophisticated electronic tacking equipment used in fleet management and SVR. Large insurers through their industry association, South African Insurance Association (‘SAIA’), wielded significant influence in the organisation at the relevant time of this case. VESA, as emerges later in this decision, is a complex institution. It is structured hierarchically, with a board of directors at the top and then with sub-committees beneath it which report to the board. !t also has its own secretariat comprising full-time employees, whose task it is to carry out its mandate. Yet the way the organisation ought to have functioned and the way it did are at variance, as the facts of this case demonstrate.’ Whilst VESA had ambitions to be the body 3 Testimony of Caroline Da Silva of SAIA, Transcript page 1813. * VESA Heads of argument paragraph 1.8. ° This is a problem that VESA very candidly acknowledge in their heads of argument where they state “The fourth respondent has identified the weakness in its organizational structure that a sub-committee may abuse its role in setting valid standards and to this end has strengthened and enforced the jurisdiction of the board of directors, who represent various 3
[10]
[11]
[12]
[13]representing all players in the vehicle security industry and to function as a quasi-regulator, supposedly in the interests of consumers, it laid itself open to capture by the short term interests of its members, as its history with the SVR industry illustrates. The second was the insurance industry itself. Organised through SAIA, representing all the large insurers and a large part of the rest of the industry, SAIA had no taste for the anarchy of the early days and wanted a ‘regulator’ to set standards for the industry, and then once set, to determine who met the standard.® The insurance industry justified this approach on the basis that it would reduce the costs of vetting companies by them individually and ensure higher standards. Unlike VESA, the insurance industry had clout because its decisions materially influenced demand for vehicle security products. Once SAIA wanted a standard setting body, the major SVR players decided to co- operate with VESA. VESA was eventually to set up an SVR sub-committee, comprising, inter alia, the first three respondents (the ‘SVR respondents’.) VESA developed certain standards over time which became the criteria for which VESA membership of the SVR committee and hence approval of their product was granted. How it did so and which of the organs of the organisation was responsible for their adoption is a contested issue in this case. What is not contested is that SAIA members adopted a policy in which they refused to endorse a company’s SVR product unless it was approved by VESA. During the time period that we consider in this case, which runs from 1999 until mid 2004, VESA developed a set of criteria for membership to its SVR committee. The history is complex and will be considered in more detail below, but it suffices that criteria for admission were established by VESA which were both technical and performance based. There has been little controversy over the technical specifications standard and we need not consider it further. It is disciplines, to ensure the legitimacy and high standards required of a vehicle security certification body” See paragraph 4.4. File E 1742.9 The use of the term ‘regulator’ is replete in VESA documents. It is not used in the traditional sense of a regulator that is a public sector agency acting in pursuance of legislative authority. 4
[14]
[15]
[16]
[17]the performance based criteria which are alleged to have been set by incumbent competitors to create a barrier to entry for new firms that are the focal point of this case. The controversial aspects of the criteria were that in order to qualify as a member a firm had to reach certain targets; it had to have been in operation for at least one year, have installed at least 3000 units, and have made 100 recoveries. The case of the Commission and the complainant is that the standards were set in such a way that they excluded competitors from entry into VESA, rendering them ineffectual competitors in the market. Whilst as a matter of law it was not necessary to be a VESA member to operate, it was, they allege, a de facto barrier to entry so high that no non-VESA member could succeed in competing effectively in a market where the customer base was primarily the insured motorist. Since the overwhelming number of insurers were SAIA members they adopted the SAIA position which was to give discounts on premiums only to motorists who installed a VESA approved device. Thus a new entrant had to meet the performance criteria during a period when the major source of demand in the industry — the insured motorist, was effectively foreclosed to it. One such new entrant was the complainant in this matter, Tracetec, which, since it first applied for VESA membership in 2001 and was rejected, has had a stormy relationship with VESA, eventually leading to Tracetec lodging a complaint with the Competition Commission in February 2004. ’ Concerned that its performance standards might be anti-competitive, VESA eventually in August 2003 adopted an alternative for aspirant members who could not meet the performance criteria — they could now lodge a R2 million financial guarantee in lieu of meeting the performance standards. Following the adoption of the financial guarantee alternative, several new firms secured VESA approval for their products and became members of the SVR Committee. According to the Commission the new standard was not exclusionary. Tracetec is of the view that it was and persists in this stance. ’ Bundle A 723. Tracetec statement of complaint. 5
[18]
[19]In May 2004, the three SVR respondents resigned from VESA. Following their resignation the industry has changed fundamentally and VESA approval is no longer a requirement for insurance approval for products. The case is therefore historical and so any form of behavioural relief would be academic. An administrative penalty is also not competent because the case is being brought under section 4(1) (a) and 8(c) of the Act for which a penalty is not competent for a first time contravention. Therefore only relief being sought by both the Commission and Tractec is a declaration that the conduct constituted a prohibited practice. When the Commission initially referred the matter the respondents were Tracker, Netstar, Matrix, Bandit Limited and Global Telematics SA (Pty) (Ltd). The Commission withdrew its case against the latter two on 15 November 2005. Tracetec was granted leave on 5 September 2007 to intervene in the Commission’s referral, and it joined VESA as the fourth respondent along with the three SVR respondents; it seeks relief against all four. The Commission, however, does not seek relief against VESA. B: THE PROCEEDINGS
[20]This case was first heard from 11 to 21 November 2008 and was then adjourned. The resumed hearing ran from 14 to 17 September 2009 and closing arguments were heard from 29 to 30 September 2009. The following witnesses testified: [20.1] For the Commission: Willem Pienaar, Manager: Underwriting of Santam Limited since 2001; Mr Johan De Wet, current director of business development of Cartrack (Pty) Lid since July 2001; Mr Phillip Maskrey, who joined VESA in August 1998 as a project manager and remained employed by VESA until June 2005 by which time he was divisional manager; and Conroy Oosthuizen, Former Managing Director of VESA from June 1998-September 2002. [20.2] For Tracetec: Stuart Pickering, the Managing Director of Tracetec; and Michael Holland, an economist from Price Metrics who testified as an expert. [20.3] [20.4] [20.5] [20.6] For Netstar James Hodge, an economist from Genesis Analytics who testified as an expert. For Matrix: Eugene De Meillon, a former director of Telesure investment Holdings (Pty) Ltd and former managing director of Hotline Administrative Services (Pty) Ltd, a wholly owned subsidiary of Telesure;®> and Stefan Joselowitz, Managing Director of Matrix from late 1995 to September 2007. Joselowitz is known in the industry as ‘Joss’ and this is how we refer to him in this decision. For Tracker: Caroline Da Silva who at the relevant time was a member of SAIA’s executive and was the head of Portfolio Management at Santam. Da Silva represented SAIA on the VESA tracking and recover sub-committee. Da Silva is now at Mutual and Federal, another short term insurance company; and Donald Beattie, a consultant in risk assessment who provided various services to VESA during the complaint period VESA did not call any witnesses. VESA did, however, give us revealing information when presenting final written and oral argument after the close of the formal evidence which was not challenged and which we assume to be correct. PART C: APPROACH [21]We first consider the factual history in this matter. We then go on to perform a legal analysis of the factual issues in this case. PART D: FACTUAL ISSUES. [22]When vehicle tracking was introduced into the market in the 1990s no industry standard setting body existed. According to one witness, Caroline Da Silva of SAIA, VESA had approached the participants in the industry, but at that time 8 As managing director of Hotline, Eugene De Meillon was involved in the management of Auto & General Insurance, Budget Insurance Brokers and First For Women Insurance Brokers. (Record p 315). 7 they had shown little enthusiasm.° It was, on her version, only at the insurance industry’s urging that they rejoined VESA.
[23]Although VESA was incorporated in December 1987, we do not have any documentary evidence in the record of its first interactions with the SVR companies. The earliest reference we have is found in the minutes of an annual general meeting of VESA held on 20 June 1996 where there is an agenda item headed “Tracking device.” Tracetec alleges that VESA established a tracking committee at this meeting.’ According to Tracetec there was a supporting document accompanying this minute entitled “VESA tracking sub-committee.”"’ From this document we observe that the committee had first met on 24 April 1997 and had held subsequent meetings in order to draw up a standards criteria document. Its mission statement states that its aim was to: «.establish minimum criteria for systems that have the ability to locate and safely recover stolen vehicles.”
[24]Anote to this mission statement observes that: “It is very difficult to establish criteria which will ensure certain recovery rates. The degree of success can only be determined once the system has been operational. The aim of the subcommittee is therefore to establish criteria, which if met, give a degree of confidence that successes can be realised.”
[25]Later in the same document under a heading ‘General’ the following observation is made: ° Transcript page 1776-7. “They had apparently, no progress with VESA in their current engagements.... So we encouraged them to go to VESA, because VESA is what we had always known as the insurance industry. '° File A 727-8. "l File A Exhibit TT2 pages 843-5.
[26]
[27]
[28]
[29]
[30]
[31]“if in its initial stages these criteria eliminate the opportunists, VESA Tracking will have achieved a major milestone.””” Netstar and Matrix are reflected as members of this sub-committee which is chaired by Netstar. Minutes of VESA board on 15 August 1996 show that tracking was discussed by the board since at least August 1996."° At this meeting it is recorded under the heading ‘tracking’ that the next meeting will submit draft specifications and then look at the day-to-day procedures of running VESA Track. These two themes — what the specifications are or should be and how the tracking committee is to be run, become recurrent themes filled with conflict and disagreement over the next few years. Netstar is present at all these meetings; Matrix’s attendance is more sporadic, whilst Tracker first appears in November. The discussion at the early meetings is inconclusive but it is clear that the idea is that a standard will be set by which firms will become members of VESA. The SVR respondents do not appear.a unified faction at this stage and indeed others e.g. a firm called Datatrak appear to dominate proceedings.** Tracker attends its first meeting in November, only to become mired in a controversy with Netstar over the latter's advertising claims.*° In meetings of what is described as the “VESA tracking meeting” in February and March 1997, Netstar’s Basil Papalexis comes to the fore, first presenting a vision document and then a specification for the industry which the meeting renames a guideline. ’ The minute of the March meeting records that the "2 File A 845. '3 File E page 35.1 '4 File E page 36 and 38.Matrix is minuted as attending an October meeting. 5 See minutes of meetings of October and November 1996. ' |ater a letter is written by Tracker to VESA taking issue with claims that Netstar was making in its advertising. File E page 39-41. 7 File E 44-45. Papalexis is described in later minutes as a VESA person, it would seem because he also sits on the VESA executive committee. However, he is also a Netstar person ~ in a letter of complaint from Tracker to VESA part of Trackers’ gripe was Papelexis’ dual role 9
[32]
[33]
[34]
[35]guideline is adopted by a vote and that it will be presented to the executive committee. In this guideline, section 7 is relevant to the issues of this case as it deals with “Key consideration for determining recovery performance.” Since the performance criteria are central to this case it is worth examining what they stated early in the history of VESA’s relationship with the SVR industry: “The size of each individual operators customer base was not considered relevant by the actuary consulted for this exercise. His opinion was that a base of 500 vehicles compared to a base of 20 000 vehicles and should over a period experience a similar incident rate and hence a recovery rate on this basis would be deemed fo be a fair comparison.”"® As we shall see later the minimum customer base suggested by the committee later becomes 3000. The actuary’s view on a meaningful customer base gets forgotten or disregarded. These guidelines do not come to any conclusions on appropriate numbers as performance criteria. However there is a suggestion that the recovery rate be at least within 20% of the industry aggregate recovery rate, calculated as a six month moving average. The guidelines go on to grapple with the issue of start up operations. The problem for start-up operations is that as the guidelines observe, they *..by definition do not have any performance history.” The guidelines seek to resolve this problem by suggesting the following “Financial criteria will therefore be subjective and could include: the content and scope of a strategic business plan; Management's (E 39). Papalexis writes letters which he signs as VESA Managing Director. (See A 868, undated letter to Tracker, but presumably written in June 1997). "® File E 60. 10
[36]
[37]
[38]business experience in the Tracking or related field) Management's individual credit standings....... lf we look at what is involved in establishing a tracking infrastructure then it definitely eliminates ‘ly by nights’. Where companies have not invested in infrastructures such as a radio network, they should demonstrate what type of investment they have made into their tracking business.’””” Thus the nascent guidelines recognised that start-ups would not be able to meet performance history criteria and so set up some ideas, albeit imprecise in their formulation, for resolving the problem. lt seems that some of the tracking companies became disenchanted with the approval process and were reconsidering their participation. In a letter written to Tracker, Papalexis records that the process of approval is no longer supported by some committee members, whilst others are reviewing their position on future involvement.”° In November 1998, the VESA board in retrospective mode, noted in discussing the item tracking, “ Although originally scheduled for September 1997, most of the role players withdrew from discussions” 7' This version of the history is also confirmed in testimony from Da Silva, who stated that it was as a result of SAIA’s intervention that the SVR companies came back to VESA. A letter written in October 1997 from SAIA to VESA confirms two meetings had been held under SAIA’s auspices and the upshot of them was that after discussing the tracking companies and their relationship to VESA it was agreed “that it would be preferable that the tracking companies become members of VESA”.”* It appears from a press clipping, believed to be dated in December 1997, that it was the three SVR companies that had met with SAIA. The press clipping reports that the three largest tracking companies, i.e. the SVR respondents, had met with short term insurers and © File E 63. 0 File E page 64 *1 File E 119.3. 22 File E 71. Letter dated 6" October 1997 from Taggart of SAIA to Papalexis of VESA. 23 File E 72. It is not clear from the article that it be housed in VESA, although VESA is quoted in the article commenting on the Datatrak closure as highlighting the need for some form of regulatory standard in the industry. 11 were setting up a governing body for the industry tasked with ensuring that companies trading “are above board”. At much the same time (December 1997) Datatrak, one of the earlier members of the tracking committee, announced it would cease operations forthwith. Managing Director Saul Tager is quoted as saying it would have taken the company another 25 months to break even and even then there was no guarantee. Datatrak had as its major shareholder listed electronics firm Jasco.
[39]It is not clear whether the Datatrak closure precipitated the insurers’ desire for a governing body or whether its troubles coincided with renewed efforts to get the reluctant trio back into the VESA fold.
[40]This early history evidences the following trends. The three SVR respondents were not initially enthusiastic about joining VESA as it offered few advantages for them. As emerged later when the three resigned from VESA, it potentially carried disadvantages for them because, if it set criteria that were easy to meet, VESA membership became not a barrier to entry but a major facilitator of entry.
[41]Pressure from the insurers, however, persuaded the three SVR companies to return to the fold, and VESA was selected, for want of anything better, as the governing committee.“ Once reluctantly back to the VESA fold, the three SVR companies were faced with the problem of how to frame the approval criteria. It was a task that Netstar, the largest firm, sat down to in May 1998, framing in a letter to SAIA’s Chris Bezuidenhout a proposal for performance criteria:”°
[42]Central to the proposal is that a system will not be approved unless either: “1.the system had been operational in the field for at least 12 months; and 2. at least 5000 units are installed in vehicles; 3. aminimum of 100 vehicle recoveries have been made 24 ~ VESA board minute dated 24 July 1998 also confirms this history as it describes a “renewed drive to bring tracking on board in VESA’ File E 81.2 * File E 73. 12
[43]
[44]
[45]
[46]4, a recovery rate of not less than 90% of the approved industry average has been achieved, or a performance guarantee of R2 million in favour of the insurance association to provide a level of comfort regarding financial stability and to compensate member companies and specifically their clients in the event of business failure. Once the criferia in terms of 4.1 are met, this guarantee may be cancelled.””° Two aspects of this proposal are noteworthy. {t sets out criteria for approval that mirror those subsequently adopted by VESA. It provides, however, an alternative for new start ups to meet, through the alternative of lodging a performance guarantee of R2 million, an idea that has a mixed history in VESA with those most hostile to it being the SVR respondents. Noteworthy as well are the recipients of this letter. Chris Bezuidenhout was clearly the most influential player in the insurance industry on this aspect, wearing not only a SAIA hat, but hats for VESA and Santam, the largest short term insurer, as well. The letter is copied to Tracker and Matrix, but not to any other SVR company. It is probable that the letter was the product of joint discussion between the three firms and was being sent to Bezuidenhout for his buy-in as SAIA’s support would be crucial to make the recommendation the approval standard for the industry. At that stage, as is clear from the letter, Netstar does not yet have a firm view of whether the governing body should be housed in VESA or SAIA. Interesting too is the comment that: “it is essential that the governing body be independent of the tracking industry”. On 16 July 1998, a meeting was held under the auspices of VESA, referred to in the documents as a tracking workshop meeting. We do not know the 26 File E 75-6. 13
[47]specifics of how the meeting came about, but no doubt it was a product of the desire by SAIA to have a standards body and the recognition by firms such as Netstar that they should participate in the process. The workshop was minuted. Whilst details are sketchy it is recorded that Da Silva addressed the meeting indicating SAIA had given VESA a mandate. The person chairing this meeting summarised the objectives as: [47.1] identifying the key elements that will demonstrate sustainable loss reduction: [47.2] Finding a simple, but effective method fo monitor claimed performance.
[48]
[49]
[50]
[51]it goes on to say the idea is to “group and rate rather than to accept and reject”.*” The workshop appears to have been well attended. Included in the list of invitees seem to have been a number of smaller players. 7° On 27 July 1998 the working group met again. The tracking methodology was discussed as based on the number of vehicles stolen as a percentage of the total client base, as well as the number of stolen vehicles successfully recovered. No figures were proposed yet. The start-up or new entrant problem was given attention and the idea of provisional approval was mooted as something that could be progressively upgraded until the “milestone are achieved” — presumably a reference to the performance history milestones.*° At the same time mention was made of a deposit to be paid. During August 1998, the group met three times — the minutes are cryptic and read like a work in progress rather than a systematic plan.*° Attendance from some of the industry players was variable and, not surprisingly, in the minute of the meeting on 26 August, mention is made of the need ‘for the sake of transparency’ to inform other tracking companies of 77 File E page 80. 78 File E page 66. 29 File E 83. Minutes of meeting of the vehicle and tracking recovery committee dated 27 July 1998. 6 August, ( File E 85), 11 August, (File E 90) and 26 August (File E 96). 14
[52]
[53]
[54]progress made so far. The requirements for the category of provisional membership were lower than those for full members. Whilst full members had to meet the multiple requirements of having a minimum 3000 client base, one year of operation, and 100 recoveries with a recovery rate of 90% of the industry weighted average for six months, a provisional member need only demonstrate that it had a basic infrastructure, had made recoveries in South Africa, and had received security clearances for its control room personnel.*" Meanwhile, at VESA board level, (at this time with no representation from the tracking industry) it was agreed that despite the fact that there was still dissent over the involvement of VESA in the verification process, all companies currently interested in joining VESA would be allowed to do so for a probationary period of six months, in which time VESA would be able to monitor the industry and finalise the specifications.** This proposal was never put into practice. After noting the difficulties in getting this committee started, the board records that all is on track for applications to be sought at the end of November and for the results of evaluations to be completed by the end of February.*° It is not precisely clear when the standards that are in issue in this case were approved. It would appear that they were approved by the tracking and recovery sub-committee at the end of September 1998. We know standards for approval must have been in place by the next meeting in November 1998, as the committee minutes record that all successful applicants would be notified by the end of February 1999.*° There are also records which show that Tracker, Netstar and Bandit applied for membership during the course of December 4998. We also have the letter in which VESA informed Bandit that its product had been approved due to it successfully meeting “...the minimum | requirements as detailed in the relevant VESA specification.” Since this evaluation was carried out by Les Dauth, an independent evaluator from a firm called Risk Analysis and Control, who was later employed by VESA, it must 31 File E 100. This appears to have been the position at the end of August. 32 The board meets on 7 September 1998. File E page 100,1. % File E 119.3. 34 File E 102, Minutes of meeting dated 9 September 1998. At this meeting it is minuted that a number of minor revisions were made to the mandatory requirements. 35 Eile E117. Minutes of meeting of Tracking and Recovery ( T&R)committee dated 18 November 1998. 15
[55]
[56]
[57]have been done in terms of the standards set by the tracking and recovery committee as they were by the end of September 1998. There is no other mention of the standards in the minutes after the September meeting, nor do they appear to have been placed before the VESA board at this stage. * The next meeting of the tracking and recovery committee we have is dated 24 February 1999 and is entitled, “Tracking press release meeting’.*” This meeting discussed a press release and a launch function. It was only attended by VESA staff and representatives of the three SVR companies. Bandit and Global Telematics are recorded as having sent apologies. No mention is made in the minutes of who had been approved, in what category and why. Despite this being a press release meeting, provisional approval was discussed. It is recorded that companies given provisional approval status could only market themselves as being provisionally approved and were not entitled to make use of the VESA logo.” The press release, unlike the minute, does tell us who has been given approval. They were the three SVR respondents, Capital Air and Bandit. Notably in the press release Conroy Oosthuizen, then the chief executive officer of VESA, stated that there were about 120 companies in SA operating in tracking and recovery and fleet management, that there were about 30 companies capable of doing a good job, but only seven companies to date had met the required standard. (Note here he is including fleet management as well, but even counting them there would be only six companies, Global Telematics being the fifth). There is no document, apart from the letter to Bandit referred to earlier, evidencing this approval and who gave the approval — the VESA staff, the executive committee, or the embryonic tracking and vehicle sub-committee. In all likelihood the firms were, like Bandit, approved by the secretariat after an evaluation by Dauth. In an internal memorandum to its branches and franchises, Netstar claims to have had “full approval” since 16 March 1999.°° This memorandum is intended for a wider audience (all franchisees and fitment 38 File E119. That Dauth was then an externally contracted evaluator is recorded in the T&R Committee minutes of 30 October 1998.File E115. 7” File E 166. File E 167. * File E 170-171. 16 centres are circulated and told to inform their staff of its contents.) Indeed the memo is sent out by the firm’s public relations department. It takes care to explain the following: [57.1] That a new sub-committee has been formed within VESA to deal with tracking and recovery with the object of determining and maintaining specifications for the approval of operators. [57.2] That a clear distinction is made between approval and provisional approval and that the insurance industry will only support those that have full VESA approval. [57.3] That only three companies have full approval, Matrix, Tracker and Netstar and that Bandit has provisional approval. This thus goes further than the VESA release discussed earlier which does not make this distinction in respect of Bandit, as becomes more controversial later. The memo makes no mention of Capital Air. [57.4] The criteria for approval are set out.
[58]At around the same time an article appears in the /nsurance Times, the industry news letter.*° Largely a reworked version of the VESA press release, it lists as approved companies the three SVR respondents, Bandit and Capital Air. The concept of provisional approval is not alluded to. Barry Scott of SAIA is quoted as saying that VESA had approached companies active in the field and the result was this “independent regulatory body.”
[59]What seems clear is that by February 1999, the sub-committee had set the approval criteria, had agreed with the secretariat who complied, and had communicated both the standard and who had complied with it, to the public at large. Later correspondence from Netstar seems to confirm this history.*" For this reason we regard February 1999 as the date when the first performance * File E 172. “1 In a letter to VESA board dated 20" September 1999, containing an appeal (which we deal with later,) Netstar’s Edmeston writes: “A draft set of management rules and a draft system specification of minimum requirements was drawn up prior to February 1999 for the purposes of VESA evaluation of the first batch of applicants applying for VESA approval’. (E 228) Edemston attaches this draft to his letter, and this is what we rely on for proof that this was the draft standard at the time. See E 229 for this assertion and E 231 for the terms of the draft standard. 17 approval standard came into force, albeit in a more diluted form than it was to assume later.*?
[60]Let us summarise what the standard was at this stage: [60.1] For provisional membership a firm had to: [60.1.1] Demonstrate a number of recoveries within SA’s borders, but with no minimum specified; [60.1.2] Have basic infrastructure; and [60.1.3] Have a security clearance of all personnel directed involved in tracking and recovery. [60.2] For full approval a firm had to: [60.2.1] Have a security clearance of all personnel directly involved in tracking and recovery; [60.2.2] Have a recovery rate that was 90% of the six month floating industry average ( weighted); [60.3] And additionally comply with two out of four of the following: [60.3.1] Have a minimum client base of 3000 installed units; [60.3.2] Have a period of operation of one year; [60.3.3] Have made 100 successful recoveries; [60.3.4] Have basic infrastructure.** 42 That it was somehow diluted is the view of Edmeston in the letter referred to in footnote 41 supra, where he remarks, “Furthermore the requirements for full membership status were also lowered to make the transition from provisional to full approval status easier to attain.” E 229. 18
[61]
[62][62.1] [62.2] We will refer to this later as the ‘February 99” standard. Of vital importance was that it was understood that approval was being granted to a product of a company, not a company itself. If a company developed a new product, that product was required to meet the performance criteria standard irrespective of the fact that the company had had other products already approved by VESA. As we shall see this became an issue of controversy later. In terms of their respective roles in setting the standard and implementing it we can come to the following conclusions: The tracking and recovery committee set the standards. The three SVR respondents, whilst not the only members of the committee, agreed to that standard and the probabilities are that without their consent the standard could not have been set. We say that the probabilities favour such an interpretation because this brief history illustrates that until all three companies were on board, it was not possible to agree a standard, and the later history shows that when they could not agree to changes in the standards, the committee disintegrated. It also illustrates that the three firms were not willing to agree on any more diluted form of standard that might have been less exclusionary in its effect. By initially withdrawing from VESA and only coming back at the instance of SAIA, the three respondents demonstrated how without their consent no standard could be introduced into the SVR industry. Notably the crucial letter in May 1998 from Edemston to Bezuidenhout is circulated only to the three SVR respondents. Tracetec and the Commission suggested that this letter was the original source of the performance standard proposal and the probabilities suggest that this is correct. Netstar chose not to have Edmeston or any other of its employees testify and counter these allegations. VESA is responsible for the manner in which the standard is set in two respects. The vehicle tracking and recovery committee was one of its sub- committees, and it did not prevent the committee using the organisation to set 43 In one document this requirement seems to be absent for full membership but mandatory for provisional. This seems anomalous and may be an error. (See E 231) De Clerk's interpretation, which seems more sensible, is that the fourth requirement for full membership is not the possession of a basic infrastructure, but achieving the 90% recovery rate. (See E 225). This interpretation is the one we will follow. 19
[63]
[64]the standard. Second, once the standard had been agreed to by the committee, the VESA secretariat saw to its implementation by facilitating the approval process. Provisional membership was a contested notion amongst VESA members. What was clear was the reason for a firm being only provisionally approved; the firm could meet some part of the performance standards, but not all, and until it could, it was to be classified as a provisional member. Less clear was what this entitled provisional members to, other than a hope of one day becoming a full member. This is a hope even a firm without provisional membership might aspire to. Bandit as the first firm so categorised took advantage of the opaqueness of the classification to market itself as VESA approved. This aroused the chagrin of Netstar who demanded that the rule be adhered to so that provisional members could not: “& . readily deceive the public into believing their system is fully approved, such provisionally approved operators/systems are clearly precluded from advertising their affiliations to VESA.” At its meeting in June the tracking and recovery committee spent much time discussing provisional approval. The minute records that: “Considerable opposition was expressed towards the whole concept of provisional approval. Members felt that this created confusion and- could cause major problems in the future.” *°
[65]Although the minute records that considerable opposition was voiced to provisional approval it is likely that only the full members felt this way. Capital Air was not at the meeting and Bandit is hardly likely to be against the only category to which it could gain admittance. At the same meeting a hardening of “4 File E 177 Letter from Netstar to VESA dated 13 May 1997. It seems other letters were written by Netstar on this topic at around the same time as a VESA letter addressed to Netstar refers to an April letter from it on the same subject. File £173. At the meeting of the T&R committee on 11 May 1999, Edmeston is minuted as complaining about misrepresenting provisional approval status. Oosthuizen blames the misrepresentation on SAIA which, he says, was unwilling to identify companies which had received provisional approval. “5 File E 183. 20 attitude developed towards performance criteria. It is noted that presently firms must meet two out of the four remaining criteria and it is suggested they should meet three out of four.
[66]Complaints about the misuse of the VESA approval status appear in the next minutes and an unnamed company is recorded as having apologised for engaging in misuse.” In July a letter was received from the company, SIT Ltd, which made a grovelling apology to VESA. The reason for the apology becomes clear; SIT had applied for approval and did not want to jeopardise its chances for approval.*”
[67]At this stage it would appear that VESA officials were performing the function of approving members, as the minutes record that the committee would be kept up to date with regards to companies and products being approved and evaluated.
[68]The concept of provisional membership bedevilled the early history of VESA. It became unpopular, but it is not clear exactly with whom. Earlier minutes we have examined show that some of the SVR firms did not like it, and wanted it to exist, if at all, as an internal classification not to be relied on for marketing. Bandit had as we have seen exploited it but received heavy censure for doing so and was to apologise for abusing the status. At an executive meeting of VESA Oosthuizen is recorded as stating that it had not been well received by the insurance industry.*®
[69]At the hearing, this was a subject of contention between the Commission and the SVR respondents, with the Commission contending that it was the SVR respondents who did not want this category to exist as it was the chief means of lowering barriers to entry for start ups. The SVR respondents pointed out passages in the record which suggest that the insurance companies did not like it — a fact confirmed by Da Silva in her testimony and supported by a statement 46 -: File E 187 ‘7 File E page 191. The letter states “..and that [we] shall follow the correct route until we have been formally and officially approved by VESA’s evaluation committee.” 48 File E 196. 21
[70]
[71]
[72]
[73]made by her in the minutes of a meeting she attended of the Tracking and Recovery committee in November 1999.*° The two notions are not mutually exclusive. The Bandit advertising episode makes it clear that the provisional status was susceptible to being exploited by start-ups, something the three full members did not like, as it elided the two concepts and suggested to consumers that there was no distinction. Insurers, wanting VESA to give them the quiet life, also had good reason not to like it for fear that a provisionally approved product was the proverbial lemon. Less concerned with competitiveness in the market and with new entry as we discuss later, they too wanted certainty that this ill-defined category did not permit. However, VESA was not a homogenous entity. Whilst the tracking sub- committee was driven by the short term interests of its competitor members, the board comprising other industry players was more concerned about the standards as entry barriers. At the board meeting the following remark is minuted in respect of a discussion on tracking and recovery: “Care must be taken that any revision to the Tracking and Recovery Specification e.g. deletion of provisional membership, should not be perceived as creating barriers for new members. 90 This extract demonstrates that at this early stage in the history of its tracking sub-committee, there is already a level of awareness by the VESA board that standard setting has the potential to create barriers to entry. The new Competition Act had only been in operation for a week at the time of this meeting. As the remaining history will show, despite this awareness, the VESA board did little about the problem, until threats of litigation under the *° Transcript pages1790, 1793 and 1795. At the meeting of 8 November 1999 Da Silva is minuted as saying the SAIA did not agree with the system of semi-approval and a system should either be approved or rejected. File E237 5° File E 196 Minute of the board dated 7 September 1999. 22
[74]
[75]
[76]
[77]
[78]Competition Act caused it to become more assertive in respect of its sub- committee. Meanwhile at tracking and recovery committee level, the February 99 standard was being reconsidered. The minutes of May 1999 record that the exclusions listed on the VESA specification are to be reviewed and submitted for comments. That this is a reference to the approval standards is evident from the next rather cryptic remark under this topic of exclusion: “Members cautioned that revisions to the specification should not make membership exclusive.TM' Revision of the performance standard was discussed again at the committee meetings in June and July.*” The tracking and recovery committee met again on 13 September to discuss the new specification. Indeed the minute states that it is a “specia/ resolution meeting.”* At this meeting two significant votes took place. First, a revision of the specification for provisional membership was carried unanimously. Six firms voted on the resolutions — the three SVR respondents, Bandit, Datatrak and Capital Air. This resolution changed the requirements for provisional approval from what they had been in the February 99 standard by making them more onerous to the applicant. It is recorded that the changes will not have a retrospective effect (presumably so that Bandit at least was not faced with the higher hurdle). Provisional members now had to be in operation for a minimum period of six months and have a minimum client base of 1000. 5! File E 175 Minutes of T&R meeting dated 11 May 1999. At meeting of 10 June see File E 184 items 5.3 and 5.4. Here provisional membership is questioned and making full membership comply with three out of four, not two out of four is mooted. For meeting of 19 July see File E 187 items 5.1 and 5.2. Here specification was discussed but not voted on and it is minuted that a special meeting will be announced to “adopt” this specification. Later Edmeston claims in his correspondence that the specification was finalized at the July meeting. File E201 23
[79]
[80]
[81]
[82]
[83]
[84]However, the consensus on raising the bar for provisional membership broke down over doing the same for full membership. Here the proposed resolution would make firms have to comply with three out of four of the following: 3000 units installed, one year of operation, 100 recoveries and a 90% recovery rate of the industry weighted average. The full membership proposal was rejected on a split vote, with four firms against it (Matrix, Bandit, Capital Air and Datatrak) whilst Tracker and Netstar are on their own supporting it. What accounts for the split in the vote? We get some guidance on this point from letters later written to the VESA board by Netstar and Tracker, who appeal against the outcome and ask the board to not ratify the decision containing the minimum proposals. Netstar suggests that the board: “consider and recommend to the committee what minimum specification the VESA board proposes to be reasonable. “4 According to Netstar and Tracker it is self-interest on behalf of the four, who, knowing that they could not qualify for the raised standard, voted it down.” De Clerk explains why the move from two to three made such a difference to the resolution’s opponents.*® Matrix’s Echo system, although in operation for one year and achieving a 90% recovery rate, did so from a low base as it had only recovered six vehicles from eight incidents. Bandit had not yet been in operation for one year and its recovery rate was based on 12 incidents. Datatrak was “waiting in the wings” for the one year period to elapse, but its recovery rate of 106 vehicles out of a base of 917 vehicles was questionable. None of these firms had reached the 3000 unit mark although some had been in operation for one year. Hence the 100 recovery 4 File E 230. 55 De Clerk states: “The proposals for more stringent requirements that were tabled at the meeting have been outvoted solely because of self — interest and not in the interests of industry.” See letter to VESA op cif, File E 225. * | etter to VESA op cit File E 225.. 24
[85]
[86]
[87]requirement being made mandatory was now the barrier to full membership if the requirements went from two to three. Tracker makes it views on the necessity for stringent standards abundantly clear: “Typically with the emergence of a new industry there were always many suspicious opportunists trying to capitalise on the ignorance of confused and not so well informed clients. We cannot allow this to happen in our industry. *” Netstar asserts that providing for provisional membership was a major concession by itself, and Tracker so agreement could be reached on standards and in the “...interest of not exercising restrictive or anti-competitive trade practices ...” Like Tracker, Netstar goes on to accuse the four of voting against the resolution because none of them could meet the proposed specification for full approval status. Edmeston goes on to suggest that they were even willing to support the proposal if it were to apply to future applicants and not themselves. Netstar goes on to suggest why the first draft specifications were totally inadequate because they allowed operators who had not demonstrated a reasonable recovery service on an ongoing basis to be approved. This: “ debases the standards Netstar, its co-members and VESA should seek to uphoid..... This contention is made with hindsight and the experience of seeing the slow growth in volume and slow progress in demonstrating their ability to recover vehicles of the 3 smaller VESA members as well as that of the Eco [the Matrix] product” * 57 E 224 58 Note that Echo is a Matrix product. Matrix as appears from the letter qualifies for approval in terms of its MX2 system 25
[88]Netstar asks the board to not ratify the decision and to recommend to the committee what minimum specification it considers reasonable.
[89]_Edmeston makes a veiled threat in his letter that: “Netstar, for one cannot afford to associate itself with an organization whose credibility is vulnerable and where the self-interest of certain members places the credibility of the industry at risk.” °°
[90]The two letters are written in remarkably similar terms and clearly the appeals were done in consultation, or one of the firms followed the approach of the other. ©
[91]This episode has been the subject of conflicting interpretations. The SVR respondents rely on it to suggest that there is no collusive arrangement | between them — here we have Tracker and Netstar aligned against Matrix and the other firms. What is interesting is who is aligned on the respective sides of this issue. Netstar and Tracker support raising the requirements because, as the overwhelmingly largest firms, they can afford to do so. Matrix finds common cause with the newer entrants as it realises that it will either not meet the new standard or not meet it for new product launches. At this time Matrix is launching new products and has met opposition that its products fail to meet the standard.
[92]The fact that Matrix does not align its interests here with Netstar and Tracker in raising the hurdle for approval does not make its interest in making a hurdle which it can achieve, but other newer rivals cannot, any less compelling. Firms in collusive relationships are not necessarily the best of friends. As the authors of a leading US textbook trenchantly remark: 59 -: File E 230. ®° We know from a fax cover sheet that Netstar sent its proposal, which is sent later to VESA, to Tracker as well. File E 227 26
[93]
[94]
[95]“Smoke filled rooms and hard drinking may moderate disagreements but they cannot eliminate differences in price and output preferences.”*' Whilst Tracker and Netstar had a common incentive to want to raise the requirements for full membership from that agreed in the February 99 standard, all firms were agreed on not lowering that standard for full membership and were agreed on raising it for provisional members. !f this lower standard was exclusionary, it does not matter that some of the firms to the agreement wanted to set the standard even higher. Expressed differently, the desire of some for a more ambitious standard does not negate the fact that they all agreed on a less ambitious, but possibly still exclusionary, standard. Both Netstar and Tracker also advanced themselves as the champions of the consumer, unwilling to drop standards for narrow self-interest like the other four. Yet as we will see later, they too would drop this high-minded posture when it suited them. The real reason for their appeal was that they considered the present standard to be too inclusive and they feared that if it was not raised, this would soon allow more entry and hence pose a threat to their strong positions in the market, particularly from firms with newer technologies. It is also noteworthy that the veiled threat both made of not being able to afford to associate with an organisation that loses its credibility by relaxing its standards, was no idle one and was later carried out, at a time when, ironically, willing assistance was forthcoming from Matrix.°* But, as we show below, the significance of this threat was not lost on the board, which despite its concern over the legality of the approval standards was equally concerned that if the requirements were lowered to a level unacceptable to the two major firms, those firms would leave VESA. Without them, VESA could not claim to be the body regulating the SVR industry. On 8 November the Tracking & Recovery committee met again for the first time since the appeals were lodged. Nothing yet had come of the appeals. Oosthuizen reported that the board wanted to meet the committee to discuss 81 Antitrust Law and Economics, Ernest Gellhorn, William Kovacic and Stephen Calkins. Thompson West, Fifth Edition, page 194 ®2 Netstar’s view on this has been quoted above. Tracker stated; “We can certainly not afford to be associated with an organization that run (sic) the risk of losing its credibility by giving the wrong guidance to those that rely on them for making purchasing decisions.” File E 226. 27
[96]
[97]
[98]
[99]various issues raised by the insurers. ®° The minute is garbled, so it is unclear if the two issues are separate or seen as one — the appeal and the concerns of the insurers. Meanwhile the insurers were back in attendance at this meeting and Da Silva gave SAIA’s views on provisional approval. She is recorded as saying that insurers did not agree with the principle of “semi approval’. Datatrak, which at this stage was neither a full or provisional member, but for some reason attended meetings, indicated that entrants would be adversely affected if provisional approval was done away with. Da Silva’s response was that a financial guarantee system should be considered. The guarantee should protect the client base in the event of the company withdrawing from the market. The meeting reached no resolution on this issue and Oosthuizen announced that all existing requests for approval would be delayed pending the outcome of the appeal. Thus interestingly the appeal had succeeded beyond its proponents’ wishes. It had not raised the bar to full membership, but it had frozen admission to membership for some time. The battle against provisional membership intensified. SAIA had notified at the next meeting of the committee on 22 November, its request for provisional membership to be discontinued citing as its reason that new operators might meet the provisional status but then drop out. At this same meeting Oosthuizen proposed a distinction be made between fleet management products and tracking and recovery products. The proposal was well received. Later, as we shall see, the real reason for this proposal emerged. An interesting discussion took place at this meeting between Oosthuizen and Chris Bezuidenhout of SAIA. This discussion was not minuted but we know about it from an email from Harry Louw of Netstar who had attended the meeting and reported back on it to his colleague at Netstar, Katherine ® File E 237 Da Silva who had not attended T & R meetings since August 1998 was accompanied by her colleague Linda Du Plessis. At the hearing, counse! for Tracetec pointed out that Da Silva attended only 7 out of the 45 meetings of the SVR Committee or its predecessors. (Transcript 1896) 5 File E240 28
[100]
[101]
[102]Krafezyk.= According to Louw, Bezuidenhout had suggested that VESA membership be stopped for 12 to 18 months to allow the existing members to become profitable. Louw goes on to describe this discussion: “It is alleged that fly-by-night operators are ‘confusing’ the brokers and paying them incentives to sell their products. Clearly some of the brokers are difficult to control, and Chris Bezuidenhout (or is it the insurance industry) are looking to VESA to assist them to contro this situation...... Both Tracker and Netstar cautioned against appearing anti-competitive, and suggested that the VESA requirements should in fact determine who can and can’t obtain recognition.” This is an important passage that demonstrates two issues: first, the agenda of the insurance industry to control SVR companies through VESA and the industry's willingness to tolerate minimal entry; and second, the sophistication of the response by Netstar and Tracker, (assuming that this plan was indeed suggested by them), which avoided misuse of the committee in the crude exclusionary manner suggested by Bezuidenhout. In February 2000, the tracking committee met again for the first time since November 1999. Oosthuizen announced that the committee had been re- branded to be called the SVR committee, so as to distinguish it from the newly established fleet management committee. All are recorded as being happy with this, but Matrix is not present at this meeting. Provisional membership was again on the agenda. This time VESA had an attorney present who warned that scrapping provisional membership (which the minute describes as a SAIA request) would give rise to three dangers: [102.1] Creation of barriers to entry for companies who do not comply [102.2] VESA being supportive of a “monopoly” by the existing players © File E 243. 29 [102.3] No recognition by the insurance industry of firms who are not VESA
[103]
[104]
[105]
[106]members.°” No resolution seems to have been reached on this issue and the minute | records that the CEO would set up meetings with each member individually to continue deliberations. A VESA board meeting was held two days fater but there was no mention of provisional membership in the report given on this committee by Oosthuizen.” The only item discussed was the separation of fleet management from SVR. Much seemed to be happening outside of the meetings. Oosthuizen wrote to Edemeston of Netstar in April 2000, confirming a discussion held earlier and confirming discussion points. One of the discussion points was that an approval committee be set up consisting of current members of the Fleet Management and SVR committee members to “ensure the credibility of new applicants and products entering the market”.®° By Apri! 2000, the separation of Fleet Management and SVR, thus far presented in the minutes as a subject on which all were agreed, emerged as contentious. In a letter from Matrix to VESA, Keith Rampton, Matrix’s Executive of Operations, accused VESA of unilaterally restructuring the committees with no agreement being reached as to which members fall into which categories.”° Matrix, he writes, finds itself in the position of being dropped from the SVR committee despite having products that fit into both. He states that the Eco system was a pure recovery system. He alludes to a “... hidden agenda here,” but he does not elaborate on what he means by this. The probabilities are that he is referring to Netstar and Tracker attempting to remove their erstwhile ally from the committee due to the fact that Matrix had not supported them on tightening up on the full membership criteria. Had Matrix voted with them the resolution would not have been defeated and led at worst to a tied vote. Rampton may well have interpreted his company’s removal from SVR to fleet 87 E 273. 88 File E 282.1. 8 File E 288. File E 287, letter dated 19 April 2000. 30
[107]
[108]
[109]
[110]management as punishment for not voting with the big two and hence the reference to a “hidden agenda”. He demanded immediate reinstatement to the SVR committee and noted that Matrix had been “compromised.” By April 2000 VESA records show that the SVR committee had only three members (Bandit, Netstar and Tracker) while Fieet Management, where the unhappy Matrix now resided, had eight, despite the fact that it had only recently been formed. ” On 3 May, the fleet management committee met. The minutes record that both Matrix and Capital Air complained about the way the re-definition had been carried out and the fact that they had been subsequently ‘de-listed’ from SVR which “... has been detrimental to their business’.” Both are recorded as having lodged an appeal against the re-definition. As in many other VESA meetings, the issue was not resolved and was left for later consideration. On the same day two hours earlier, it seems) the SVR committee met with only Tracker, Netstar and Bandit in attendance {apart from the officials). The following issues were confirmed: [110.1] The separation of the SVR and fleet management committees was approved; [110.2] provisional approval had been discontinued with effect from September 1999; [110.3] Bandit was granted full approval status (Bandit’s situation during this period is curious. If provisional approval was abolished with effect from September 1999, but Bandit was only approved as full member at this meeting in May 2000, it is difficult to understand how its representatives attended meetings during this period when its status seems to be that of a non-member). ”! See File E 286.2 and File E286.3. ?2 File E 303. 31 [110.4] Members were asked to attend approval committee meetings in order to
[111]
[112]
[113]
[114]“include industry knowledge and opinion regarding approvals under consideration”. This meant the SVR committee not only set the standards but had also appointed itself as the arbiter of the standards by becoming the approval committee. The Matrix controversy finally resulted in a meeting between VESA officials and Matrix’s management on 16 May.” The upshot was that it was agreed that Matrix would rejoin the SVR committee and that the ECO and MX2 would be transferred to the SVR category. The fate of a third product, Matrix’s MX3, was left for discussion in the SVR committee, once Mairix had applied for approval. Oosthuizen is recorded as commenting that he was not certain the Eco numbers were ‘sufficient,’ to which Rampton responded that Eco had been approved by VESA prior to the minimum criteria being set. Matrix’s perception that it was being harmed in the market by the re- classification was not without foundation. !ts rivals were quick to ensure that the industry was made aware of the reclassification. In a letter from Tracker to a broking firm, Tracker pointed this out claiming “.... none of the Matrix products are approved in this sub-section.” (i.e. the stolen vehicle recovery sub- committee). At the same time Tracker points out that it has VESA approval.” This letter seems to have fallen into the hands of Matrix as Matrix complained about it to VESA.” The dispute between Matrix and VESA continued into June with further correspondence and meetings. Whilst the MX2 was reinstated to the SVR category, according to VESA officials the Eco did not qualify, not having met 73 The minutes of this meeting were compiled by Matrix File E 309. 74 (etter from Chris Kustner of Tracker to R Thompson of Meridian Brokers dated 23 May 2000. The basis for the letter seems to be that the broker was offering greater discounts on two of the Netstar systems and Kustner used recent VESA stats to suggest that Tracker’s recovery rate for the most recent period was better than that of Netstar. File E 311-2. File E318. Letter from Rampton to VESA dated 5 June 2000. Rampton states the letter to Meridian brokers clearly shows how VESA has caused confusion by ’misrepresenting’ the Matrix products. 32
[115]
[116]
[117]1118] the 3000 target. Matrix claimed that it had been approved in the previous year (June 1999). Matrix upped the ante and a letter from its attorneys was sent to VESA at the end of June 2000 demanding the reinstatement of the Eco system to the SVR category.” The next meeting of the SVR committee following all this controversy took place on 4 July 2000. Matrix was represented at this meeting as were Tracker, Netstar and Bandit. The chairperson announced that the Matrix MX2 and Eco had been ‘redefined’ back under SVR. The meeting also recorded that: “an extended period be granted to allow those members whose systems do not conform to the new specifications i.e. up to December 2000 after which they will be re-evaluated. In the interim, their systems should be re-admitted to the listing to enable them to obtain the insurance support necessary to sell their systems. The proposal was accepted.””® This decision is extraordinary in the light of recent history. Having done away with the system of provisional approval, the committee had reinstated what amounted to provisional approval for incumbents only. Secondly, by admitting them to the listing, they were treating these products as if they were the subject of full approval. Why did the committee do so? The probabilities are that the members would rather have the incumbents in the room on side rather than against them threatening litigation and competition scrutiny, which might have been the outcome of the episode with Matrix. At the same meeting it was proposed that the committee henceforth act as the Approval Committee for any new system or member. 6 Letter from Matrix to VESA dated June 2000 File E 322. File E336 has what appears to be the 1999 application form, which shows a subscriber base for Echo of 450 ” Eile E 337 letter dated 22 June 2000 from Werskmans to VESA. 78 File E 355. Minutes of SVR committee dated 4 July 2000. 33
[119]The SVR committee’s expedient approach to dealing with incumbents’ products
[120]was demonstrated again in relation to the Sleuth product of Netstar. introduced in late 1999 as a new product, it came up for approval at the October 2000 SVR committee meeting. Whilst it had recorded more than 3000 installations it had not been in operation for more than one year and was way short of 100 recoveries. Indeed it had recovered only 23 vehicles out of 29 reported thefts. Nevertheless Don Beattie the consultant hired by VESA to perform evaluations recommended approval based on Netstar’s track record on “other product lines” and there were no: “.. substantive reasons to apply disqualification. The numerical non- compliance deviations at present will obviously achieve target in the near future.” At the meeting, which was held on 24 October 2000, Netstar motivated why Sleuth should be approved even though it fell short on the 100 successful recoveries and one year of operation. Netstar solved the dilemma by calling for a revision of the rules. Matrix and Bandit (and the VESA personnel) supported Netstar on this, but Tracker voted against the proposal. The reason for Tracker doing so are clear from the letter to the broker we considered earlier. Tracker, unlike Netstar, had at that time just one product, whilst its rivals were introducing new products. By making rivals’ new products conform each time with the approval criteria, it created barriers to entry to these new products. Not surprisingly it is recorded after this discussion that: “Members felt it was important to maintain transparency and not to be seen to be self serving.”*° *" ”* File E390. 8 File E 403. 8! There is another reason why Tracker may have wanted to delay the entry of the Sleuth product. In October 2001, once Sleuth had become VESA approved, a minute from the Tracker board of directors indicates that “According to VESA statistics for July and August Netstar’s fitments were greater than Tracker. Netstars Sleuth product and their relationship with intermediaries (broker commissions) were the main components for the increased fitments.” File E 631.1 34
[121]The criticism is not attributed but could apply equally to the positions of any committee member. A meeting was arranged for 8 November 2000 to discuss the proposal for a revision of the rules.
[122]Matrix group chief executive officer Stefan Joss could not attend this meeting so he wrote to Oosthuizen outlining his views on the revision. Clearly the stricture against being self-serving had not inhibited Joss, who wrote in his concluding paragraph: “In conclusion, | feel that this restriction has evolved into a totally impractical mechanism, especially as it applies to established operators. if it stays at all, it should apply to new entrants only and as for established operators (with overall subscriber base of in excess 3000 to 5000) any new product offerings should be left to the integrity of the individual organisation, providing of course that the product meet the minimum technical requirements as laid down by VESA.” ©
[123]Yet the letter is damning for VESA and the SVR respondents in another respect. Joss basically articulates the central concern of Tracetec and the Commission in this case that the standards are a barrier to entry: “The dilemma we are ail facing is that VESA is ultimately successful (and by this i don’t mean that VESA hasn't achieved a certain admirable success to date) and it would be technically impossible for either the established players or new entrants to sign up 3,000 subscribers on a non-VESA approved stolen vehicle recovery product. if the VESA brand name develops to the level that we would aif like to see it, neither the insurance industry at large nor the consumer would purchase a non-approved product. If this was the case, then all we have achieved is inhibited the introduction of valuable technology and the fight against vehicle crime will ultimately be lost as the criminals become more and more proficient with overcoming the existing 82 E 407. 35
[124]{125}
[126]
[127]offerings. Clearly this is not the industry’s, VESA’s or the consumer's interest. Having written thus, Joss was in his evidence forced to concede that the standards were exclusionary, but he maintained that this was only if VESA had brand equity and, in his opinion, it did not. On 8" November a new resolution was adopted by the SVR committee as follows: “A new stolen vehicle recovery specification for existing members was established: (Our emphasis) e Minimum client base was 3000 now 2000 e Period of operation was one year now 8 months e Number of recoveries per product was 100 now 30." Interestingly the standard mirrors in all but one respect the circumstances at that time of Netstar’s Sleuth system.” As it happened it did not take long after the meeting for Sleuth to meet the newer lowered standards for existing members and in a letter dated 22 November 2000, Edmeston informs Oosthuizen that they have met this standard and they have achieved 41 recoveries out of 51 theft incidents. °” A smali fracas over advertising indicates again the importance that the SVR respondents attached to the VESA name, contrary to the position now adopted by Joss. Aft a general meeting of VESA in 2000, various awards were given to members. Tracker received one for “outstanding contributions to vehicle security”. Tracker allegedly released this to the press claiming to have been given the award for the best SVR company by VESA. This led to angry letters from Matrix and Netstar who induced VESA to insist the paper publish a correction. In his angry letter Edmeston states; 83 E406 para 6. 84 Transcript pages 1450 and 1576. * File E410 8° Sleuth would not have met 30 recoveries but was not far short of this reduced target at 23. Perhaps even Netstar would not have influenced a standard of as low as 20. 87 It is unclear when precisely this product was approved, but VESA’s February 2001 recovery rates figures list it as a product for the first time with effect from December 2000. 36
[128]
[129]
[130]
[131]
[132]“The use of VESA’s name and reputation in this manner is surely not 188 permitted by VESA, without explicit approval by VESA In contrast to its attitude to incumbents, the SVR committee retained a hard line on new entrants. A firm called Global Telematics applied to become a member and its application was considered at the SVR committee’s meeting on 3 May 2001. The application was rejected and Global Telematics was asked to remove the phrase “VESA approval pending from its advertising”. The following justification of this decision appears in the minutes: “Members believe that the 3000 min client base before being able to apply to VESA is not exclusionary. Approximately 70-75% of the vehicle park is uninsured and consequently VESA approval would not normally be a major concern. Thus a relatively large poo! of vehicles is available from which to reach the required number of fitments.” °° This anticipates a defence that that the respondents ail rely upon in this case. It is not clear why the committee needed to minute a justification of its performance criteria at this point in time, but perhaps it was for the benefit of the consumption of the VESA board and SAIA. In August 2001, at the next meeting, Beattie is recorded as remarking that Global Telematics is ‘progressing’ and that it appears it will meet the criteria soon.?° At the October 2001 SVR meeting the Matrix MX1 system was approved but 91 not its MX3 system, although the minutes do not tell us why.” Matrix’s 88 File E 434. The problem was resolved at the next SVR meeting with an undertaking by Tracker that the award would not be ‘misquoted’ in any additional advertising material. File E 451. The award was discontinued by the board on the recommendation of the SVR committee. File E 487. °° File E 487. %° File E 499. °1 The minute records that no consensus was reached on the MX3, however a subsequent letter from VESA to Matrix indicates that it was not approved, but that VESA would advise them later about this. 37
[133]1134]
[135]
[136]Rampton was understandably unhappy with the decision and minuted his objection.” At the November meeting the MX3 was approved. Again the minute does not explain why - it simply records that the objections to the approval had been withdrawn. °° The MX1 became a subject for disagreement the following year when at an April meeting of the SVR committee, at which Tracker and Matrix were the only companies present, De Clerk asserted that the MX 1 product approved was not the one currently being distributed. Rampton of Matrix argued that the system was the same as the one approved, but had been the subject of technological advances.” In February 2002, the SVR committee discussed new approvals and the issue of a financial guarantee. Although the minutes do not make this clear, the financial guarantee had earlier been proposed by Da Silva in November 1999, as a solution to the new entrant problem.°® The committee notes that after discussions with VESA’s lawyers it was concluded that “... it was nof viable to proceed with the financial guarantee in lieu of achieving the current specification.” The minute further recorded that the current specification would remain in force, and if challenged, “... an independent evaluator will be asked to investigate the current criteria to ensure that they are reasonable.” This passage clearly shows the approach of the committee which at this time still comprised the same four SVR companies and the VESA officials, with SAIA although a member, rarely present. The committee’s approach in not adopting the financial guarantee alternative at this time, and instead taking a wait and see approach on challenges to the existing criteria, are all indications of its indifference to the exclusionary possibilities of the existing standards. °2 Like Sleuth the MX1 would not have been approved unless the lower standard adopted for existing members was applied. On its application form the MX1 is reflected as having 2170 installations thus falling short of the 3000 for non-members but above the newer 2000 standard for incumbents. (File E 520). The application is dated 17 August 2001. °° File E547. * File E 607 % File E 583. *6 At the meeting of the T&R committee on 8 November 1999. 38
[137]
[138]
[139]
[140]|
[141]
[142]Yet despite adhering to their standards until challenged and rejecting the financial guarantee alternative, members were not wholly clear on the interpretation of their own standards, as the minute of their next meeting suggests. The minute records that the following definitions need to be clarified by members: “what constitutes an incident — what constitutes a recovery — what constitutes exclusions” ” The solution, in conformity with past practice on these problems, was to procrastinate and defer to a later specification review meeting. Meanwhile in April 2002, the subject of the approval of Global Telematics arose again. In a letter to members, VESA advised that Global Telematics had passed ail the evaluations, met all the specification except the 100 recoveries. It was noted that the required recoveries would be achieved in the next few days. Members were asked if they would have any objection. We do not know if any did, but Global appears in the recovery rate statistics published monthly by VESA for the first time in June 2002, suggesting that it had now been approved.® It needs to be recalled that Global Telematics’ membership had first come up at a meeting in May the previous year. In May 2002 there was a combined meeting between members of the SVR and the VESA board. The debate over the MX1 continued. Tracker was not present at this meeting but Edmeston of Netstar indicated that it too, wanted to upgrade products and wanted to know why an upgrade should be treated as a new product. Again the issue was deferred.** In June 2002 Edmeston asked for approval of an updated unit. Edmeston was frustrated that the launch of the unit 7 File E 604. % File E 656.2. ® File E647. 39
[143]
[144]
[145]
[146]was being delayed by a hold up in VESA procedures. Like Rampton from Matrix, he was arguing that this was not a new product, but an upgrade. ‘°° The Matrix M1 was still a cause for discord at the SVR meeting on 18 July 2002. At this meeting Global Telematics was present for the first time. The minutes record that the MX 1 was to retain its status as approved, but De Clerk and Edmeston were clearly unhappy. De Clerk indicated he would be taking legal opinion and Edmeston indicated that the decision was not in the spirit of the SVR committee. At its August 2002 meeting, the perennial subject of standards was still on the agenda. The discussion on whether a change in technology constituted a change in product, and hence needed to meet the approval criteria de novo continued, with the meeting eventually resolving that this would be acceptable, if a professional engineer certified that the technology change would not degrade reliability. '°' In the same minute it is noted that an incident can be recorded as a recovery even if the unit is recovered without the car. At a later meeting the following caveat is added “This applies only if a member's own unit was recovered or was fitted to the recovered vehicle.” '°? The farce continues when at the next meeting this item is again referred to and is put on the agenda for the next meeting. In February 2003 a firm called Cell Stop wrote to the new VESA CEO, Henk Van Zyl, to apply for membership of the SVR committee. According to the letter the firm had been a member of the VESA electronic committee since 2000. (This is not the first time that Cell Stop has featured in the history of VESA. !n November 1999, at a meeting of the Vehicle and Tracking committee Caroline Da Silva informed the meeting that a firm called Cell Stop had approached SAIA to present its system to its motor advisory committee.“*° She raised the concern that Cell Stop’s advertisements state that its system does not affect 10 File E729 and 730. ‘1 File E763. 0? File £839. 103 File E239. Cell Stop later writes to join VESA claiming the approval criteria are a barrier to entry but states it is willing to lodge a financial guarantee File E 1162. This is in May 2003. and 1183. 40
[147]
[148]
[149]
[150]
[151]motor warrantees and that this was incorrect. Da Silva when asked about this extract at the hearing had no recollection of this." But if the Cell Stop experience is anything to go by, it makes it clear that SAiIA was unwilling to deal with a firm outside of VESA structures and not only that, reported back on these efforts to the firm’s competitors.) What is interesting about this application in 2003 is that Cell Stop claimed that it could not meet the 3000 unit installation and 100 recoveries requirements, because its product was not a mass market one. It currently had 1500 units installed, but just more than 500 were in South Africa. Cell Stop goes on to suggest that the current criteria inhibited innovation, as firms who were already conforming would be reticent to develop new products that call for re-evaluation as they understand the difficulty in achieving the numbers required to qualify.‘ Cell Stop’s application was considered by the SVR committee on 4 February 2003.‘ The application to waive the current application criteria did not find favour with the committee who stated that both new and current members had complied with the standard. The minutes note that Cell Stop had stated that it believed the current standard was exclusionary and that it was willing to have the legality tested “with regard to anti-competition laws”. The committee records that it considers the standards not exclusionary, and that lowering standards would lose them credibility with insurers. They record that it might be beneficial to have the current standard tested in a court of law. Cell Stop appealed to the VESA board. !n the letter of appeal it states, “... it seems that the sole interest of this committee is fo prevent new companies and products from entering the market.”"°” Cell Stop made the claim that other VESA technical standards (referring to other product committees) ail rely on rigid specifications whilst by contrast, '°4 Transcript 1857. '5 E954 letter dated 3 February 2003. 108 E 959-60. '°7 E 970 Letter is dated 7 February 2003. 41
[152]
[153]
[154]referring to the SVR specification, “...if is clear [that it] has been written specifically around keeping the competition out.” Recall that Cell Stop was a member of the electronic sub-committee. The Cell Stop issue was discussed at the VESA board in March 2003." The board evidenced some concern. It is clear that Cell Stop was taken seriously and was recognised to have both the means and the motive to take “the legal route.” The observation was made that: “The current perception may be that the SVR rules protect their own industry from newer members. This is due to the fact that their product is not only technical but also encompasses a service agreement.” Although there was discussion about whether the matter should be referred for advice, someone pointed out that this had been done and that the advice was that the specification was ‘marketing’ based and not a ‘technical’ specification. Rampton of Matrix, who represented the SVR committee on the board, was absent, so the board resolved the matter by asking him to give feedback on the re-writing of the specifications to technical standards. It appears that the board had previously made this request of the SVR committee in 2002. In the transcribed version of the discussions it emerged from the same meeting that various members of the executive had remarked inter alia that: [154.1] the SVR committee had drawn up rules which protected the industry and they were not going to allow anyone else to come into the arena very easily:(“the two biggies are trying to keep everyone out’); *°° [154.2] that the specification of 100 recoveries and 3000 installations was unrealistic; [154.3] The people who designed the specification had had the opportunity to be in the market before it was in place; '08 File E 1020. Meeting is held on 11 March. ©? File E1050. 42 [154.4] despite the board instructing them based on legal opinion to change the specification nothing had happened; [154.5] There were fears that the SVR companies, if pushed, might separate from VESA; [154.6] the board had sanctioned the criteria; ‘"° [154.7] the SVR respondents regarded the legality of the standards not as their
[155]
[156]
[157]
[158]
[159]problem, but that of the board.’*! The VESA board met again in April 2003 and Louis Green of Cell Stop was given an opportunity to address the board. Green made a presentation about his product and stated that without approval of his product as an SVR product he could not sell the product to insured vehicle owners, who were the largest target market identified." A suggestion by one member of the board that his product could be better housed in the fleet management committee was rejected by Green who explained that this did not suit his products characteristics. Whilst this was going on, the dispute with Tracetec, the complainant in this case, escalated. Some background on Tracetec is necessary. Tracetec was established in 2001. Its founders wanted to enter the SVR market because they believed that a radio transmitter technology that had been used in the retail and security sectors could be successfully applied in the SVR market. In October 2001, Tracetec submitted an application to join VESA." Tracetec was evaluated by Beattie on behalf of VESA and found not to comply with the performance specifications. Tracetec complained that without VESA "° File E1051 "tt Eile E1032-1037 and E1047 -1049 ‘12 File E 1152. Meeting was held on 16 April 2003. "3 43 p 756. Tractec statement of complaint para 72. 43
[160]
[161]
[162]
[163]accreditation, insurance companies were unwilling to back it so it was not possible for it to meet the performance criteria." Tracetec eventually found a home in another VESA sub-committee known as the “Electronic Accessory” category. Tracetec was later informed that this category had been terminated and that VESA wished to move them to the “Fleet Management” category. Tracetec did not consider that this met its needs and VESA then suggested it move into the tracking category. This proposal too, did not satisfy Tracetec, which complained to VESA in a letter dated 14 April 2003, requesting that a new category called “Vehicle identification and Recovery” be opened *... where we might fit in’.” 1" It did not want it said to be placed in the same category as the other tracking companies and “...nor for that matter would they be happy to have us there.” There is a curious ambiguity in Tracetec’s approach here. It does not want to be in tracking, yet both prior to and subsequent to this letter, it did. This apparent ambivalence was made much of by the SVR respondents. But it seems what Tracetec was seeking to achieve was to obviate the need to meet a hurdle it could not pass — the performance criteria of the SVR committee — but still somehow to get into the lucrative SVR market via a side entrance, through the creation of a committee that because of its similar name, looked and sounded like an SVR committee. Tracetec’s stratagem might be regarded as disingenuous, but it would be wrong to infer from this manoeuvring that it did not seek to get into the SVR market. Rather it shows that entrants struggling with the entry criteria sought to gain the VESA endorsement via other means, knowing that without it entry was well nigh impossible. Rather than being inconsistent with Tracetec’s case it is wholly consistent with it. In May 2003 the VESA board met again. The problem of the SVR membership criteria was still occupying its attention and it is worth quoting this extract in full: ‘14 Tracetec complaint supra paragraph 2.3. "? File E 1148. 44
[164]
[165]“The board has on previous occasions requested the SVR committee to review its membership criteria. In the light of current possible legal action, it was decided that VESA cannot fight a legal battle on this matter, and testing the process legally will be costly. The committee members will have to bear the costs of such a legal process. The question was raised again on the choice of 3000 units in operation and 100 recoveries being the correct and defendable numbers, and what those number (sic) represented in terms of company and product stability and performance.” On the day foliowing this board meeting, Van Zyl, VESA’s managing director, received a request from Rampton that the SVR committee meeting be postponed by a month. This might have seemed innocuous, but not to Van Zyl who wrote to his board members for guidance on whether to comply with it. His letter is another reflection of the ambivalent attitude of VESA towards its SVR committee. Van Zyl wrote that he was concerned that the postponement might be seen as an attempt to prolong the status quo. He noted that three companies, one of which was Cell Stop, had written to him wanting to join the SVR committee, stating that the current membership criteria was a barrier to entry into the market. Despite his suspicions of the motive for the postponement he notes, however, that he feels he has to comply with a request of a committee member, “...especially if backed by other members of the same committee. "77 The postponement request was successful and the next meeting was held in July. What is clear from what happened at this meeting was that the pressure was building on the SVR committee with threats from outsiders to litigate, coupled with the concerns being expressed by the VESA board. A discussion was held on reviewing membership criteria. First came a candid admission; it was observed that the current specification did not guarantee the sustainability of a new product nor did it provide criteria against which the VESA technical advisors could verify fitment quality, as they could with alarms and gear locks. 116 File E 1162. The meeting is held on 28 May 2003. "7 File E1176. Email is dated 29 May 2003. 45
[166]
[167]
[168]Second came a dramatic alteration of the approval criteria; it was decided to offer new entrants the alternative of lodging a financial guarantee of R 2 million until such time as the 3000/100 client base/recovery level had been reached. The minute records that these criteria would be circulated to members for sign- off.'*® The meeting also noted that Cell Stop had indicated that it would be prepared to lodge a financial guarantee. It further noted that a firm called Global Asset Protection had expressed an interest in becoming an SVR member. Recail that this idea of the guarantee alternative was not new. It had been first mooted by Edmeston of Netstar in his letter to Bezuidenhout of Santam on 15 May 1998 and then mentioned by Da Silva at the meeting of the SVR committee on 8 November 1999.""° Why then did it take so long for the committee to endorse this suggestion when one of its own had initially proposed it, and SAIA, the supposed mandating body, had endorsed it? The overwhelming probabilities are that the SVR committee members, led by the three SVR respondents, appreciated that the financial guarantee alternative would have significantly lowered barriers fo entry into their committee and so they frustrated its adoption until the pressure placed on them eventually forced them to do so. The transcript of the meeting contains a long comment from one of Matrix’s representatives, presumably Joss, which makes a number of important observations: [168.1] That it was relatively easy to get Fleet Management approval because you don’t need the three elements necessary in the SVR requirements ( ie 3000/100 and one year of operation); [168.2] That installing 3000 units does not ensure the success of a business; [168.3] That he thinks the specification is anticompetitive in its current form, as two extracts indicate; 18 File E 182. 9 File E 76 and E 238 46
[169]
[170]“| am not sure what we are doing around this table. We are a bunch of competitors trying to mandate an industry. Everybody has got their own self —interest at heart and at the end of the day we are suppressing free competition. My own view is that if there are new entrants let them come;” '7° And “Matrix could not have launched in the market that exists today.”... “... we cannot provide the insurance industry the protection that they want, and that is why I'm revisiting in my mind what we are doing here.”"*' When the VESA board met again in July 2003 it noted that the SVR committee had amended its criteria to make provision for the financial guarantee. Whilst the record shows a resolution being adopted by the SVR committee there is no resolution ratifying or approving the change by the board - it is simply noted.'”7 This suggests that the specifications were a matter for the committee, not the board. The SVR minutes by contrast as we noted above had indicated that the criteria would be circulated to members for sign-off.'?* But whilst the SVR committee might be lowering criteria for outsiders it was proving equally adept at looking after its own. Tracker had over the years been the most vocal opponent of lowering standards to suit incumbents. The probable reason for this was that its product had not changed over the years unlike those of its rivals who, as we have seen, were launching new products. Tracker however was to abandon its orthodox views when it decided to introduce a new product, the Tracker Alert. At its meeting in August 2003 the SVR committee approved the Tracker Alert product, despite the fact that it had not even met the approval criteria for new products for existing members set at 30 recoveries; Tracker had only achieved 22. The rationale for this was that the product was considered an “enhancement of an existing product” and hence, as Tracker had proved the viability of its infrastructure, this requirement was waived, '*4 120 File E1185.13. "1 File E 118.25. ‘22 File E1191. The meeting is held on 10 July 2003. 19 File E1182. "4 File £1210. Tracker report this in a board minute dated 13 August 2003. E 1231.1 47
[171]
[172]
[173]
[174]
[175]On 14 August Van Zyl wrote to Pickering of Tracetec to advise him that the VESA board had decided that Tracetec must apply to the SVR committee. Recall that Tracetec had asked VESA to create a new category for it.’° On the following day Van Zyl again wrote to Tracetec to inform him of the financial guarantee option.’ He stated that this option was the result of a meeting of the SVR committee held on 1 July 2003. The rationale for the guarantee it was explained was to allow customers who bought the new entrant’s product to convert to another system in the event of either business failure or product failure. A definition of what business failure means was conveyed in the letter — it goes beyond insolvency and includes a failure to perform services to customers. Tracetec was not happy with this outcome and wrote back to the chairman of the VESA board to express its dissatisfaction. Tractec still wished to be defined differently to SVR, but also had qualms about the lack of detail surrounding the terms of the financial guarantee. Tracetec asked to address the next VESA board meeting on 20 August.**” Tracetec did so and in an acrimonious meeting both parties held to their original positions - VESA that Tractec must apply to SVR and Tracetec that it had been given an expectation that a new category would be created for it. The minute of the board records that they felt that the product was better defined under SVR, but that it would look at better defining the guarantee."7° In the same minute a draft SVR specification is tabled and Rampton is recorded as stating that it might be anti-competitive.**? The minute states that the draft will be forwarded to the board for review."*° The transcript of the proceedings on this point is not very coherent. It appears what is being referred to is a technical specification that was drawn up by the technical managers of the various members of the SVR’s but there was some 125 £4232. 126 £4233. 127 E 4234-5. 128 E 1377, 128 F 235 190 F 4378. 48
[176]
[177]
[178]
[179]
[180]concern that the technical specification was in Rampton’s words, “not going to let many people in.”**" In correspondence through August and in to September there is much wrangling between Tracetec and VESA over the terms of the deposit and then also on some technical issue not raised earlier. Given that VESA undertakes to let Tracetec do a draft guarantee to submit to it, it seems that Tracetec was wangling to improve its position and could, but for this posture, have supplied the guarantee and entered the committee.'*” But Tractec seems to have persuaded Van Zyi later of a need for its own category. Van Zyl wrote to Tracetec undertaking that following a demonstration to him of the product he would consider advising the board of having a non- categorised section, whilst conceding that this was a radical new way of operating for VESA."** : At the next board meeting Tracetec prevailed. The board approved “the start up of a vehicle identification category”. The board noted that Tracetec’s offering was not SVR and the option of helping them to VESA membership was noted and vehicle identification was seen as a possible solution to it. Tracetec was informed of this on 12 September 2003."*° But in October Chris Bezuidenhout, wearing his Santam hat, wrote to Pickering of Tracetec to complain that Tracetec was telling people that its system was comparable to that of Netstar and Tracker, and that Tracetec was also saying that Santam has accepted this, implying that Santam has accepted it as a tracking device. Bezuidenhout states that this was not acceptable to Santam and urged Tracetec to desist from making these claims. **° VESA held a meeting with SAIA and vehicle manufacturer association Naamsa on 19 September 2003. Tracking was one of the agenda items. Da Silva who 81 Eile E 1352. '82 See letter from Van Zyl to Tractec dated 1 September 2003. It seems that during an evaluation of Tracetec it was not possible to trace a vehicle tested on. 83 File E1400. ' File E 1401.2 "85 File E 1516. 86 File E 1633.1 49
[181]|
[182]
[183]
[184]opened the meeting mentioned that because of the success of the tracking industry, SAIA was inundated with calls every week from potential new entrants, “claiming to have superior products and threatening the Competition Board.”"*” Da Silva notes that SAIA would not be concerned abou this as long as it was just “minimum standards that were creating barriers to entry.” She recalls that the alternative of a financial guarantee had been proposed, but not adopted. Scott from SAIA mentions that it must be made clear that barriers to entry are based on high standards and not set simply to stave off the competition. Van Zyl, who represented VESA, mentioned that VESA had introduced the financial guarantee in August and that five new applicants were being considered who wanted to make use of this alternative. Despite this, VESA is recorded as saying that it did not support the market being opened up to anyone or even further than it is currently, but agreed that it needed to be transparent and in a position to defend its specifications in the face of an attack."°° lt does not appear that the insurers were convinced. Complaints were made about communication and Scott remarks that he does not know if the guarantees provided the necessary peace of mind. It is recorded that “SA/A expressed disappointment with the process to date.”** Tensions between SAIA and the SVR respondents are noted in the SVR committee’s next meeting on 7 October 2003. The minutes refer to SAIA’s Barry Scott and Chris Bezuidenhout’s being extremely forceful in their demands that they be given drafts of the new specifications and also the wording of the financial guarantee."*° The SVR committee notes that SAIA approval was not required, but that SAIA was entitled to send a representative to their meetings. (Note that while this is going on SAIA Approved, a standard-setting body 'S? File E 1561.2. 'S8File E 1561.4-5 Note that all three SVR respondents are represented at this meeting. "89 File E 1561.5. 40 File E 1625.3-4. 50
[185]
[186]
[187]
[188]
[189]directly controlled by SAIA, is being mooted and that this has been viewed by VESA as an initiative likely to threaten its existence.)'*" At the same meeting two new entrants experienced different fortunes. A company called Car Track was admitted as a new member and is recorded as being the first new member to utilise the financial guarantee option. Another company, Karroo Cell, was allowed to address the meeting to voice its concerns with the financial guarantee option. This plea was rejected and the committee decided that the mandate would remain unchanged.‘*? '° On 9 October 2003, Van Zyl writes to Tracetec to notify it that its cheque for payment of membership fees was being returned as the new membership category had not yet been formed according to the requirement of VESA internal procedures.'** This was explained as due to an “ambiguous mandate between the VESA management and the Board of directors.” Apologies were tendered and Van Zyl suggested the delay in sorting this out would not be long. On receipt of this letter Tracetec wrote to Barry Scott of SAIA asking for SAIA’s intervention on its behalf with VESA, failing which it would take legal action against VESA."*° The Tracetec saga continued at the meeting of the VESA board in October 2003. The board was concerned that Tracetec was misrepresenting its approval status and its product capabilities. The board believed that Tracetec was using its ‘Vehicle identification’ status to facilitate a move to SVR. The board considered that Tracetec’s motives were not genuine and decided to write it a warning letter. Interestingly in discussing why it had created this category the board notes: '41 File E 1563 -1624 Transcript of VESA board meetings held on 25 September 2003 and 2nd October 2003. 42 File E 1625.3. The Commission obtained a witness statement from Alvaro Tafur the company’s spokesperson at the meeting but could not call him as he is now resident in South America. (Transcript p959-960), 143 File E 1636.1 Meeting held on 16 October 2003. 4 File E 1726. 146 File E 1624.10. Although this letter is dated 6"° October it refers to Van Zy!’s letter which is dated 9 October, so one of these dates is incorrect. 51 “It was also stated that it [vehicle identification] was not yet in a position to threaten any existing VESA category or committees. it was seen to offer enhancement to certain security product offerings.’**°
[190]There followed a lengthy exchange of correspondence between VESA and Tracetec during October.
[191]Tracetec’s approach to resolving the issue was not consistent. At one stage it proposed becoming a member of the yet to exist Vehicle Identification category, till the end of December 2003, to be followed in the following month (January 2004) by admission to the SVR category, on tender of a financial guarantee.'*” Later the financial guarantee was tendered, but in an amount of R 750 000 based on the fact that Tracetec’s product was not as costly as the standard SVR product on which the R 2million figure was originally calculated. Thereafter followed a threat of litigation including a threat to take the matter to SAIA."° An attorney's letter was sent on 31 October, but in this letter the demand was confined to the re-instatement of Tracetec to the Vehicle identification category.'*°
[192]On 3 November Chris Bezuidenhout wrote a letter to be circulated to a number of brokers concerning Tractec allegedy marketing its system country wide as a tracking and recovery system.’*' Bezuidenhout writes that Tracetec is claiming that their system justifies offering substantial discounts on premiums for policy holders who have the system installed. Bezuidenhout points out that the system has only been approved as vehicle identification and does not have approval as a vehicle tracking system. Bezuidenhout states that SAIA does not approve systems as this is handled by VESA. He states that it is unlikely that Santam would offer the discount on premiums on this product and that he has asked -VESA to request that Tracetec be given a final warning to halt its “... deceptive marketing’ . 46 File E 1636.2. "47 File £1731. 4° File E 1738. "49 File E 1742. '8° File E1742.28. The letter also asks for confirmation of Tracetec’s membership of VESA a seemingly superfluous demand. '®? File £1742.9 52
[193]
[194]
[195]
[196]On the following day, the SVR committee met. It discussed a request from Tracetec that the financial guarantee option be reviewed.**? The request was rejected. The minute also records Rampton as saying Tracetec is claiming it is VESA approved for recovery and that VESA has sent two letters to them warning them about this. At the same meeting members voted to boycott a magazine for brokers called FA News which had requested VESA to let it do an objective test on members’ recovery and service. After the lawyers’ letters, Tracetec adopied a different approach and proposed to furnish the R2 million guarantee and go into SVR. Van Zyl circulated this to all board members, who were decidedly cool. Rampton took the hardest line of all, saying the product did not conform to the technical evaluation specifications and hence, even if Tracetec were to furnish the guarantee, it would still not qualify for membership. He noted that Tracetec was causing havoc in the market place with the “illegal VESA certificate” and that to date VESA had done nothing about this. He suggested going to court to get a “restraining order” against Tracetec to stop it misrepresenting its VESA status." Meanwhile matters only got worse for Tracetec. On 17 November SAIA put out a statement to all its members from Da Silva noting the claims made by Tracetec that its systems had been approved by SAIA and VESA. She noted that SAIA did not approve systems as this was handled by VESA. She goes on to state that VESA had approved the system as Vehicle Identification only, and not as a stolen vehicle recovery system. The following day VESA’s attorney sent off a letter of response to the allegations, denying that Tracetec was a member of any committee, let alone obtaining recognition as a member of a vehicle identification committee, although in the final paragraph of the letter this was still kept as a possibility it wanted to obtain. "© 182 File E 1745. Meeting held on 4 November 2003. File E 1751, '4 File E 1759.1. ‘85 Eile E 1767. The letter ends “. we do however trust our client shalf be in a position to advise your client as to the establishment of the new VI membership category shortly.” 53
[197]
[198]
[199]
[200]On 24 November VESA’s attorney wrote to inform Tracetec’s attorneys that the Vehicle Identfication category had been created and invited Tracetec to make an application to become a member of this category. The Tracetec debate continued into January 2004, with lawyers at one stage appearing to come close to settling the issue and then the settlement failed. In its January 2004 minutes, the VESA board stated that Tractec was still advertising its products as VESA and ABA approved, even when they were not. Eventually on 18 February 2004 Tracetec referred its complaint to the Commission.’ At the same time in January 2004, the VESA board had to grapple with the claims of another would-be entrant to the SVR committee, over the burdensome nature of the guarantee. A firm called LST wanted a waiver to be able to raise the guarantee over six months, instead of having to provide it upfront. This request was referred back to the SVR committee to decide. But, as if to suggest it had not entirely passed the responsibility on to the committee, the board chose to record in the minutes that “... it may review any decision if it believes that decision to be wrong or not in the best interests of VESA.” Once again this is indicative of the VESA board engaging in an exercise of seif- delusion. It appeared reluctant to make the decision itself, despite the fact that determining whether the guarantee should be reduced was hardly a technical decision that only the SVR companies could determine, yet at the same time it wanted to make it clear to the SVR committee that if it did not like the decision it remained the superior organ in the association with the power to review it. If the threat of review was meant to incentivise the SVR committee not to put its own interests ahead of VESA’s, it did not succeed. The SVR committee considered the request from LST on 3 February 2004 and rejected it.'°”
[201]The fault lines within VESA were not confined to the relationship between the SVR committee and the board. The committee also clashed with the VESA officials. ‘86 File E 1937. ‘57 File E 1921. The minutes also record that Tracetec has referred its complaint to the Competition Commission. 54
[202]
[203]
[204]
[205]At the same meeting the members were informed, apparently by one of the VESA officials, that a firm called Digitec had now been approved under SVR. Indeed Digitec was present at the meeting. Rampton was clearly riled by this and stated that the committee needed to be informed with regard to a possible new member before the member was invited. Van Zyl’s response was that the committee has given VESA a mandate and when an applicant complied with the specification, the application was approved. Rampton’s riposte was to request a copy of this mandate. LST wound up back again as a VESA board problem at its meeting on 13 February 2004. The SVR commitiee’s decision was not ‘reviewed’ however, but referred to VESA’s attorney to deal with “...in order to facilitate their application for SVR membership .”"** It is not clear from the minute what the attorney’s mandate on this issue was, but had VESA wanted to take an opportunity to assert that the standard was exclusionary and to rescind the committee’s decision, it lost an opportunity to do so. By mandating the attorney to deal with the matter it left the standard to an issue of individual settlement with LST, and not one of broader application to other would be entrants. On 17 March 2004 Louise Dauth, a VESA official, wrote to all members of the SVR committee stating that a firm called Mobile Tracker had been evaluated and that it had met the technical standards and lodged the financiai guarantee."*’ She asked committee members for any comments by the next day in which case “...we will assume that we may confirm approval with the chairman Keith Rampton to finalise the approval.” Dauth’s deference to the wishes of the committee members indicates that Rampton had won the mandate argument with the VESA officials. Approval of new members was not something the officials could implement by interpreting their mandate. Both Edemeston and De Clerk wrote back to say that they could not give their approval. ‘°° Edmeston wrote that the committee needed to make sure it did not turn the approval into a rubber stamp. He also questioned whether the VESA staff had the expertise to do the approval, even though he admited he did not know the person who conducted the tests. De Clerk wrote 188 File E 1935. 188 File E 1977. 180 File E 1976. 55 that operators were now going to the insurance industry with VESA approval and insisting on support. He adds more ominously “... we need to seriously assess whether we want to be part of this.” {206] These two emails were circulated only to Van Zyl, Netstar, Matrix and Tracker - unlike Dauth’s, which had been sent to all members of the SVR committee.
[207]Rampton wrote to Netstar and Tracker on 2 April 2004 to inform them that VESA had stepped back from its position on certificates."*' He stated that with the possibility of SVR splitting from VESA this suggestion had been put on the table for discussion at the next meeting. Again the circulation of this is only to Tracker and Netstar and not the other committee members. The issue of the certificates may seem peripheral to the subject matter of this case, but its relevance is two-fold. The SVR respondents and in particular Joss of Matrix offered this as the principal reason for their resignation from VESA; the second reason is that it attests to the fact that VESA was unable to impose its will on this committee and hence is relevant to the issue of who was responsible for setting the performance standards for SVR as we consider later when we deal with finger pointing between VESA and the SVR respondents.
[208]On 6 April 2004 the SVR committee met again. It was to prove the last meeting any of the SVR respondents attended."®* For the first time they were outnumbered on the committee by the newer members. Discord in the new widened ranks is apparent.'®* The minutes record that “... some members feel that having the VESA brand associated with certain unproven products will 181 Eor sometime VESA had wanted installers of SVR products to issue certificates to the motorist concerned who would then pay for the certificate. The installer in turn would have to be VESA approved to be entitled to issue the certificates. VESA saw this as a method of ensuring standards, the SVR respondents viewed it as money generating scheme for VESA, by leveraging their client base for more expenses. Although alarms and other security products were managed in this way the SVR respondents considered the certificate superfluous because if a vehicle could be tracked it had been properly installed. Note that in a transcript of a VESA board meeting dated May 28 2003, the board discusses whether to impose certificates on the SVR industry. Mr.Jones, a member of the board, remarks that if SVR is imposed upon the SVR companies “they would just walk away”. He notes how big SVR is in the insurance industry’s eyes; “... they are just looking at SVR and think SVR there is only one thing in the world and that is SVR” He observes if you want to keep SVR from walking they need the insurance industry's support because the insurance industry does not want too many satellites. (File E2035-6)). 182 Tracker is not recorded as being present in the minutes but the transcript shows that De Clerk was present and very vocal at the meeting. "83 File E1989. 56
[209]
[210]
[211]
[212]dilute the VESA brand.” It is also noted that the introduction of the financial guarantee is seen as a significant relaxation of the approval criteria. At the same meeting VESA’s legal representative warns that the current SVR specification is anti-competitive and a restrictive trade practice. It is agreed that a new draft technical specification should be completed as a matter of urgency. (It is not clear what is being referred to here. The technical specification, the performance criteria or both since it is the performance criteria, not the technical specification that have been the subject of complaint). Subsequent to that meeting the three SVR respondents and Bandit met to discuss jointly terminating membership of VESA forthwith. We know this from an email circulated internally to Netstar staff by Edmeston.*** The reasons given in the email were that with the new financial guarantee new members could join without a track record and gain the same standing as the “likes of Netstar"® The email contains a frank assessment of the dilemma that Netstar and the other SVR respondents were now facing. They believed that the lowered entry criteria were allowing new entrants to win credibility by association with the industry leaders like Netstar instead of earning it the “hard way” themselves. At the same time there was an acknowledgement that they.may be a target of competition litigation. Thus faced with this Hobson choice of staying in with lowered entry barriers and allowing in entrants to ride on their coat tails or maintaining the status quo and facing a lawsuit, the firms had decided to opt out altogether and resign from VESA. Edmeston remarks that SAIA were not concerned about them resigning. He notes that due to tensions between SAIA and VESA the two organisations have had a fall out. The SVR respondents had also become unhappy with VESA attempting to regulate them more. The complaint about VESA wanting the SVR ‘6 File E 2006. Dated 26 April 2004. Edmeston refers to a meeting with certain other members where they “jointly agreed to terminate our membership forthwith.” 18 This assessment seems to have been shared by Henk Van Zyl of VESA. In a meeting between him and the three SVR respondents that was transcribed he is recorded as stating: “/ am afraid some of these smaller guys now is using VESA you know just to walk into this market and it doesn’t work like that...” File E 2017.22 12 May 2004 ) 57
[213]
[214]
[215]
[216]
[217]fitment centres to join VESA and to impose a fee for fitment certificates is mentioned. Edmeston aiso predicts that once the majority (by which he means the four firms) have resigned, the SVR committee would disintegrate and he would expect the fleet management committee to disintegrate as well. This last observation indicates that the resignation was not simply an act of disenchantment with an organisation whose policies they could no longer support, but a calculated and well-considered move to discredit a committee that was now well-placed to lower barriers to entry to the SVR market. Only by resigning collectively and being able to suggest that over 90% of the market was now outside VESA, could they credibly do so. Netstar also indicated that there was no intention at that stage to create another body for representation in the industry. On 12 May 2004, all three SVR respondents sent a letter of resignation to VESA and announced that their resignation was effective forthwith." They enclosed with their formal separate letters of resignation, a joint letter explaining as they put it as a matter of courtesy, why they had resigned.'*” In the joint letter they explain that in pursuance of the original mandate from VESA to prevent another Datatrak debacle, the performance criteria had been drawn up. This however had led to concerns about the criteria being anti- competitive and they were concerned that VESA and “...even potentially its members” were exposing themselves to legal action in this regard. They also state that the financial guarantee alternative: “...could be deemed fo be imposing an unreasonable barrier to entry.” “ Remarkably they state that they “...don’t believe that either of these mechanisms provides a suitable level of comfort that complies with the original insurance mandate.” The reason is that even achieving these targets does not give one the critical mass to be anywhere near in a profitable trading position. '86 File E 2012, (Tracker), File E 2013 (Netstar), File © 2014 (Matrix) ‘87 File E 2015. 58
[218]
[219]
[220]
[221]
[222]
[223]VESA, they state, had been unable to come up with a mechanism that will both meet the original mandate and not be deemed anti-competitive. They go on to observe that: “The hard reality is that like with most service provider businesses, the test of time’ is the only mechanism that will provide a high level of comfort about a business’s ability to support itself over the long term.” Although the installation certificate issue is mentioned as a further reason for their resignation it is noteworthy it is given as the last reason. The final line is a curious one. The three SVR companies request that VESA desist with immediate effect from making any claims regarding its representation of the Stolen Vehicle recovery or tracking industries. Edmeston’s internal email observation that the SVR committee will disintegrate is being put into effect. On 16 May Edmeston put out a circular addressed ‘To whom it may concern’, explaining Netstars reasons for resigning from VESA, outlining essentially the same reasons offered to VESA, except there is no mention here of the certificates of approval.'®® He notes that to remain a member of VESA “imparts credibility to the approval system that it no longer can justify in terms of the diluted approval process. ... With the resignation of all the substantial members of VESA, VESA is no longer representative of the industry.” Van Zyl on 25 May informed VESA membership of the resignations and indicated the committee would continue with its remaining five members.'*° In June the SVR respondents put out a press release concerning their resignation. ‘”? The press statement repeats much of what has been said earlier around the resignation. But it has been clearly designed to ensure that readers know that the heart of the once representative SVR committee had been ripped out. It points out that the three firms represent 95% of the market. 188 File E 2019. 18 Eile E 2021. These are Bandit, Global Telematics, Digicore, Cartrack and Mobile Tracker. 70 File E 2182. 59 They go on to assure the public that they have received overwhelming support from the insurance industry since their withdrawal! and state that there is not a single insurance company that they can identify that will not support their products. But the three are not content to issue the statement simply to reassure the public that it is business as usual notwithstanding their resignations, they go on to attack the VESA brand in no uncertain terms: “VESA’s ‘solution’ was fo offer new entrants an alternative to the original standards whereby they could effectively ‘buy’ VESA approval by lodging a refundable financial guarantee with them. Although we do not claim that all new entrants being approved on this basis will not make the grade, we do believe that at least some of them will ultimately fail. We therefore believe that VESA may have lost sight of their original mandate and_we_no longer see any value in their brand equity. it is our opinion that ‘ VESA APPROVED’ status is no longer a meaningful qualification and as such we have decided not to dilute the value of our respective brands by further association with this process..... In the absence of meaningful regulations for approval we believe it is preferable to have the market decide for itself which SVR operators to support.” (Our emphasis)
[224]That support of insurers was vital is indicated by the fact that both prior and subsequent to their resignations, all three collectively went to meet with the industry. An internal memorandum from Netstar is evidence of this as well as a report at the next SVR meeting, where Davidson of Bandit reads out a press release quoting Da Silva as requesting all short term insurance companies to continue accepting the three firms’ products.'’’ Another internal email from Netstar records a meeting the three firms held with SAIA on June 23 2004 to discuss the way forward.'’? They discussed whether there should be a ‘SAIA Approved” to replace the VESA SVR once it was established that the three firms would not “reconcile”. De Clerk advocates a “free market system” with underwriters doing their own approval. This suggestion, according to the email, was rejected by Bezuidenhout who says SAIA does not approve “... this buying of business criteria” ‘71 Eile E 2181.2 Meeting of the SVR committee dated 3 June 2004. ‘72 File E 2227. Memorandum dated 9 July from Henry Smith to John Edmeston and Harry Louw. 60
[225]SAIA agreed to advise all underwriters to “...keep on supporting the members that withdrew from SAIA.”
[226]The rest of the history to date is short. After a brief lifespan of the entity which became known as SAIA Approved and which attempted to replicate the functions of VESA, no organisation exists at the time of writing that approves SVR devices and has the support of the insurance industry.‘
[227]De Clerk’s free market has been realised, ' PART E : LEGAL ANALYSIS
[228]The respondents are alleged to have committed the following contraventions of the Act. In the case of the SVR respondents it is alleged by both the complainant Tracetec and the Commission that the SVR respondents have contravened section 4(1)(a) and section 8(c) of the Act. In the case of VESA, Tracetec alleges that it has contravened section 4(1)(a) of the Act.‘”®
[229]For reasons that will become clear later, we deal first with whether the evidence discloses a contravention of section 4(1){a) by the respondents.
[230]Section 4(1){a) states: “4. Restrictive horizontal practices prohibited "3 Da Silva testified that after the resignation of Netstar, Tracker and Matrix from VESA, SAIA tried to set up SAIA Approved which was later registered as a private company. At that time she moved to Santam and left SAIA. She said that there were a lot of problems with SAIA Approved to such an extent that she stated SAIA Approved “.simply didn’t fly”. . (See Transcript page 1798) Later whilst questioned by VESA’s representative she states that she is not sure how the industry organises itself now (Transcript p1906). Joss stated that at one time Matrix had joined SAIA Approved but later it pulled out. (See Transcript p737) According to VESA in its heads of argument, as a result of the resignation to the best of its knowledge there had been “... a reduction in demand for new members becoming VESA approved due to the insurance industry. and the industry itself being unsure of the direction of VESA standards.” (VESA HOA paragraph 4.5.). 74 When being cross-examined by counsel for Tracetec, Joss admitted that it was his opinion back then that, in the absence of credible criteria of accreditation then the market should determine what happens in the market and to the market players (See Transcript p1735- 1736). (See also Transcript p1704-1708). > Recall that the Commission has not referred a complaint against VESA. 61 (1) An agreement between, or concerted practice by, firms, or a decision by an association of firms, ts prohibited if it ts between parties in a horizontal relationship and if _ (a) It has the effect of substantially preventing, or lessening, competition in a market, unless a party to the agreement, concerted practice, or decision can prove that any technological, efficiency or other pro- competitive gain resulting from it outweighs that effect” (a) Was there an agreement or concerted practice?
[231]It is common cause that the three SVR respondents are competitors in the SVR market. The agreement alleged to be a restrictive practice is the performance standard for membership of the VESA SVR committee. The SVR respondents do not concede that they are a party to an agreement in respect of these standards. It is hard to see how they can credibly make this assertion. The SVR respondents all joined VESA with a view to it setting performance standards. The early history of the Tracking and Recovery sub-committee showed that when the SVR respondents did not reach agreement no standard was determined. It was only when all three returned after being prodded by SAIA that a standard was agreed upon and that required consensus being reached by these three firms and in particular Nestar and Tracker. Although other firms had brief roles in the committee they played as extras. The real roles and hence the agreement on the standards came about when the three SVR respondents reached an agreement all three found acceptable. The agreement was finally determined sometime in September 1998 and then implemented in February 1999, with the press release. It was, as the chronology shows, the sub-committee which set the standard and then proceeded to make it public ~ prior to its endorsement by the executive of VESA — something it is unclear ever happened. At various times this standard was amended by the Committee to make it easier for the three firms to get approval for their new products whilst at the same time denying this latitude to new entrants. The SVR respondents then voted in favour of raising the admission threshold by abolishing provisional 62
[232]
[233]membership and then repeatedly resisting the suggestion of a financial guarantee until conceding to this in August 2003 at the eleventh hour. As the history has shown, when major decisions needed to be taken they often consulted outside of the committee and only amongst the three of them. VESA permitted the SVR committee to set the standard and the SVR respondents were the key protagonists in that consensus — hardly surprising as they represented at that time more than 90% of the industry. When the consensus in the SVR committee broke down at a time when the SVR respondents could be outvoted by the other members of the committee, they resigned, collectively endorsing the same reasons for their resignations and adopting the same press statement. As they had operated in setting the performance standard so they operated to prevent the standard that had been for once imposed upon them by outside pressure, from becoming a means of easing entry into the market. As we have seen, once they left VESA it ceased effectively fo become the approval body for the industry. There is therefore little doubt that the standard set for approval based on a performance standard was reached as a result of inter alia the agreement of the three SVR respondents. Since they did so as a VESA sub-committee, which VESA permitted to be held out to the world as its standard, it must be regarded as a decision of an association of firms as well. We deal more fully with VESA below. (b) Did the agreement, concerted practice or decision substantially, prevent
[234]or lessen competition in the SVR market? The theory of harm advanced in this case is that the standards operated to exclude rival firms from effectively entering in the market to compete with incumbents who had been approved in terms of the standard.
[235]Not all standard setting by rivals raises competition concerns. Unlike price fixing and market sharing by rivals, standard setting is not per se illegal and hence the case has been brought under section 4(1}(a) and not section 4(1){b). 63
[236]Standard setting may sometimes benefit consumers, for instance safety standards protect consumers against unsafe products.
[237]However when rivals collectively set a standard it poses the potential for danger to competition. As Hovenkamp in his treatise puts it succinctly: *“.. joint standard setting can facilitate collusion or impede progressiveness by excluding from the market firms that threaten their rivals with lower prices, higher quality, or other innovations that consumers would prefer if given the opportunity.”
[238]How then do we differentiate between benign and malign standard setting? There seems to be no magic answer to this question. The best one can do is to have regard to the following factors; [238.1] Does the standard setting body have market power? If the answer is no, it is unlikely to have an exclusionary effect. [238.2] Who drove the standard? A standard set by rivals will prima facie be suspect for that reason alone as they will be presumed to be intended to limit the entry of other rivals. However a standard set by customers would not attract this conclusion as customers would not be considered to be setting standards with the exclusionary intent.'”” [238.3] What is the effect of the standard? At one end of the continuum may be a standard without adherence to which a competitor is unable to enter the market. At the other end may be mere communication of a standard. The less its effect the less restrictive it will be considered to be. [238.4] Is the standard reasonable? United States courts have often shown a reluctance to determine whether a standard is reasonable or not since, as Hovenkamp put it, an expert can be put up for either side of the case.*” Despite this caution a standard may be considered unreasonable by means of ‘76 Herbert Hovenkamp, “Antitrust Law’, paragraph 2230 page 343. 7 Wovenkamp op cif paragraph 2230c page 348-9. 8 See Hovenkamp.op cit, page 372. 64
[239]
[240]
[241]other indicia -— is it consistent with its rationale, is there evidence that a reasonably efficient firm or a firm that is at least as efficient as the respondent firms could comply?"”* Market power: \t is quite clear that the SVR respondents in this case did have market power. Collectively by their own admission they represented over 90% of the industry."®° The claim was not mere self-serving hype. An internal VESA document from February 2002 showed the relative market shares of its members at the time; Tracker with 45% Netstar with 44%, Matrix 9% and Bandit 2%.**' Nor have the SVR respondents put these market share figures in issue. Before they put their efforts into VESA no standard could be drawn up for the industry and when they left the SVR committee ceased to practically exist. The VESA board as we have seen was unable to exert its authority over the sub- committee until legal threats from outside made them exert some belated pressure. By then it was too late. There can be little doubt that the SVR committee had market power, because it had as its members the three SVR respondents who represented over 90% of the market for SVR. It follows that since the SVR committee was a VESA organ, so did VESA. Who drove the standard: James Hodge, the economics expert testifying on behalf of the SVR respondents, argued, relying on Hovenkamp, that standard setting should not be considered anti-competitive when the standards are driven by consumers, as opposed to competitors. He argued that the facts of this case suggested that the consumer here was the insurance industry, which had wanted the standard and had approved it. In this scenario the SVR committee is seen as a mere agent of the industry translating the insurance companies wishes into performance standards. Hence the frequent use of the term the ‘mandate given by SAIA’ that has pervaded the SVR respondents’ submissions throughout this case. Even if we accept that Hodge’s theoretical submission is correct, which we need not decide now, there are two reasons for not applying the theory to the facts of this case. 8 See Hovenkamp op cif 358 to 372. '80 In the press release referred to earlier at the time of their resignation they claimed to represent over 95 % of the industry. E2182. 8" File E 593.1. 65
[242]
[243]
[244]First, the evidence shows that although SAIA might have driven the concept of having a standard it was not instrumental in formulating its content. When Da Silva attended the meeting during the provisional approval controversy she suggested the financial guarantee option (indeed not an original idea either it had first been proposed by Edmeston a year earlier) but it was ignored from the time she suggested it in November 1999 till the adoption of the guarantee in July 2003. !n the late stages of the SVR committee’s interaction with SAIA, the latter, as we have seen, expressed concern with the exclusionary nature of the criteria and Da Silva had queried why the financial guarantee had not yet been adopted. '®? At the October 2003 SVR committee meeting Scott and Bezuidenhout of SAIA are recorded as demanding that they be given drafts of the new specifications and the text of the financial guarantee. These were not demands they had made earlier. The probabilities are that at this time SAIA, like the VESA board, was becoming worried about threats of competition litigation and wanted to assert some control over the wayward child it had created. But this was done only after the standard had !ong been in effect, and points to SAIA’s lack of involvement in the content of the standards. This is not conduct analogous to that of consumers driving the creation of a standard. Indeed SAIA’s lack of involvement in the standard setting phase of the committee was pertinently pointed out by Tractec’s counsel during cross examination." Second, even if one is still willing to give to the SVR respondents the benefit of SAIA having driven the need for a performance standard that would exclude a number of firms, the reason SAIA was doing so was not based on any consumer welfare agenda. Rather the facts show that SAIA had its own agenda for wanting VESA to contro] access to SVR market and this was not rooted in concerns for the consumer but in concerns for its own business interests. ‘82 We refer to the meeting between VESA, Naamsa and SAIA on 19 September 2003.0f course it had been adopted at this time, but only in July/August of that same year. 183 At the hearing, as we noted earlier, counsel for Tracetec pointed out to Da Silva that she had only attended 7 out of the 45 meetings of the SVR Committee or its predecessors. (Transcript p1896). 66
[245]
[246]
[247]
[248]Bezuidenhout had suggested at one time freezing SVR membership so that the incumbents could become profitable. He had also expressed concern that the proverbial fly-by-night operators were confusing brokers by offering incentives to sell their products. We see this sentiment come up later when he writes a letter concerning Tracetec’s claims that its systems justify offering a discount. He states that his firm i.e Santam will not be offering any discount.’ In other words what SAIA is concerned about is that an uncontrolled industry could lead to pressure on insurers to reduce premiums by firms offering special deals. By only recognising a handful of companies that had SAIA approval this downward pressure exerted by new entrants is averted. Indeed lowering the cost of SVR was not a primary concern that SAIA had, as Da Silva conceded during her testimony.'® Once the SVR committee had agreed a standard it was difficult even for SAIA to do much about. As Hovenkamp puts it: “depending on the power and authority of the standard setting organisation, its members are generally encouraging and sometimes effectively forcing others to refuse to deal as weil.” *° We thus conclude that the approval standard was not consumer driven, because the role of SAIA in its genesis was both limited — it did not determine the content of the standard and secondly, it was conflicted -its interest in the standard was not that qua consumer, but driven by the insurers’ self -interest. Nor was there any other ‘customer’ other than SAIA who drove the standard - this was conceded by VESA during fina! argument."®” Effect of the standard: The standard in this case was not an absolute or de iure standard — a firm could legally operate in the market without having VESA approval. The case for the Commission and Tracetec was that it was a de facto 184 File E 1742.9. See also paragraph 192 of this decision supra. '85 See Transcript page 1828. Da Silva is asked by the Tribunal whether pricing for SVR products and service might come down if there were new entrants. Her answer is that although the more competition the better the prices the price at the time was not “seen fo be prohibitive” and then ‘“/ think it was a competitive price that we could afford in terms of discounting, but price would have been an issue.” This rather limp wristed response, despite her last remark to the contrary, shows that price was not seen as a concern to SAIA 186 Liovenkamp op cit page 343. 187 Transcript page 2205. 67 barrier to entry. On their version without VESA approval one could not succeed in the market. The SVR respondents argued that a new entrant could still achieve the target of 3000/100 by selling to the uninsured market. The figure cited was that the uninsured market represented 70- 75 % of the total vehicle park.'®® On this argument a new entrant could target this uninsured sector of the market as well as other clients such as car fleets, which self insure, and thus generate the necessary installations and recoveries to gain admission to membership.
[249]At first blush this argument seems plausible. After all, faced with the overwhelming number of uninsured vehicles, self insured or non-SAIA insured, why could a new entrant not make it in the market, indeed without ever having to apply to VESA for membership? The short answer is that during this period no one ever did, despite the fact that there were many potential entrants. Let us examine why.
[250]The SVR respondents have made much of the fact that Tracetec was in no position to enter the market as it did not have a viable network to perform recovery and its product was better designed for other purposes such as asset security but not vehicle recovery. Tracetec vigorously disputed these allegations. It is not necessary for us to determine whether Tracetec could have successfully entered the market at the relevant time, but for the VESA approval criteria. Rather the question is posed in more general terms — could a reasonably efficient firm or a firm at least as efficient as one of the SVR respondents have entered the market without VESA approval or if not, could it have obtained VESA approval within a reasonabie time period?
[251]There is no evidence that non-insured motorists generated any significant demand for tracking equipment. If this was the case one would have expected to see at least one of the SVR firms targeting this market given their first mover advantage the strength of their brand names and infrastructure. Yet they produced no evidence that they had considered a market which logically they would have entered it if was viable. Nor do we see any reference to this as an alternative in the long record of evidence that we have in this case. '88 This claim is made by the SVR committee in it minutes. See File E 487. Meeting of 3 May 2001. 68
[252]There was evidence that at the time VESA wanted to create its SVR committee in 1998 that there were a number of firms wanting to enter this market. A number of firms attended the first workshop on 16 July 1998."*? Oosthuizen in his press release in February 1999 when announcing the formation of the tracking and recovery committee, stated that there were 120 companies operating in tracking and fleet management and that of these 30 were capable of doing a good job."
[253]Steven Klagsbrun, who was VESA’s legal advisor at the time, mentions in a letter to VESA that he had been given to understand there were 15 firms wanting to join SVR industry. This letter was written in February 2000.'*"
[254]There is also theoretical support for the fact that a nascent industry, as SVR was at this time, would be characterised by high entry.
[255]Michael Holland of Price Metrics who was called as an expert economics witness by Tracetec testified to this effect and referred us to the work of Steven Klepper who writes: “When industries are new, there is a lot of entry, firms offer many different versions of the industry's product, the rate of product innovation is high, and market shares change rapialy.”"**
[256]Thus both the factual and theoretical evidence suggests a large number of would be entrants at the relevant time. Given this situation it seems probable that if the non-insured market was an option that some of those firms would have identified this as a viable business area and entered successfully — yet none did.
[257]One can appreciate why this happened. 8° File E66. '90 File E169. '' File E 275 at 276. 182 Steven Klepper, “Entry, Exit, Growth, and Innovation over the Product Life Cycle”. American Economic Review, Volume 86 no 3, page 562. See also Holland’s expert report, paragraphs 58-61. 69 [257.1] [257.2] [257.3] [257.4] The demand for tracking equipment was driven by insurers’ insistence or preference that cars priced over a certain value had to have tracking equipment installed. The practice varied in that it was sometimes an express requirement and other times it was merely incentivised by premium reductions. Insurers were asked by tracking companies to extend this policy to vehicles below this threshold value but insurers at that stage appeared reluctant to do so. '? When the SVR respondents resigned from VESA they made sure that they had insurance industry support prior to their resignations and then made this fact public in their joint press release. This again suggests strongly that demand was driven by insurers and hence demand from the broader car park could not be generated without insurers inducing it. Motorists who are reluctant to insure vehicles are not a likely category to install a tracking device as an alternative. De Clerk remarked at an SVR meeting 6 April 2004 “Addressing the insurance industry the importance of formalising alliances with them and 90% of the business comes from the industry” 14 Insurance endorsement was relied upon in marketing. Netstar in its marketing materials dated 1999 claims that “ Netsfar enjoys the endorsement of the major insurance companies and brokers in SA who will offer you substantial discounts on your monthly premiums of your vehicle has been fitted with a Netstar Vehicle Tracking and Recovery system” '® When Matrix was moved from SVR to fleet management it fought tooth and nail to get back, threatening litigation. Matrix as an incumbent firm thus considered that if it was not in SVR, despite remaining in VESA, its business was threatened. 493 At an SVR meeting on 8 November 1999 Da Silva was pressed by members to say why insurers won't make fitment of tracking systems mandatory for all vehicles. She says it had been shown that the “crime component” of the premium would be too small to cover the cost of tracking for vehicles under R 70 000. This seems to indicate that demand for SVR devices is driven by insurer policy and not motorist demand. Indeed the comment recorded is that: “Members conveyed concerns regarding the level of support for their systems from the insurance industry.” File E 239. '%4 Transcript of meeting File E 1991.29. ' File E 164.3. 70 [257.5] [257.6] [257.7] [257.8] All three firms, when faced with threats that their new products might not be approved, fought hard to ensure that they were, leading eventually to the lowering of standards for incumbents but not new entrants. Conflict over clains of VESA approval by firms or products not VESA approved and the saga over the VESA best SVR firm award show how seriously VESA endorsement and a fortiori insurance endorsement was taken by incumbents. If there was a ready market in non-insured vehicles would these bitter battles have been fought? The complaint that one could not get insurance support without having one’s product approved by the VESA SVR committee was not confined to Tracetec and as we have seen was a concern for other new entrants as well. The fact that some firms allegedly misrepresented their VESA status e.g. Bandit, Global Telematics and Tracetec, is an indication of its importance in marketing. Firms who were not members of the SVR committee found themselves being targeted by SVR members when they tried to market themselves as SVR providers. The reaction to Orbitech, which operated through a brokerage called Dominium, is a case in point. Despite not misrepresenting that it had VESA status, its efforts to market itself among brokers, as opposed to insurers, was met with a response from incumbents that this firm was marketing products not approved by VESA. When Matrix was briefly re- assigned to Fleet management it received similar treatment at the hand of Tracker who wrote to brokers fo point out that none of Matrix’s products were approved in SVR.'° Thus VESA SVR members not only operated to ensure their membership was an advantage but also to ensure that it was a disadvantage for non-members marketing efforts. By assuming its role as the industry regulator, as it claimed, VESA and the SVR committee made outsiders appear to be firms outside of a system of regulation.'*” '98 See paragraph 112 of this decision supra. ‘97 File E 512 Letter from Edmeston dated 16 August 2001, circulated to Oosthuizen of VESA, Scott of SAIA and Matrix, Tracker and Bandit. Edemston encloses the brokers letter from Dominium brokers and states “Secondly you may also wish to investigate it further as it is quite contrary to the VESA approval system for SVR, is not indicative of the high recovery rates achieved by VESA approved operators and could be damaging in the market place.” \n the letter Dominium query the accuracy of the claimed recovery rates by Netstar and Tracker and 71
[258]
[259]
[260]
[261]
[262]
[263]Even if a few uninsured motorists did want SVR products installed in their vehicles , is it likely that after making diligent inquiries and establishing that an industry body existed that approved products, that they would choose a product without any industry endorsement? The non-insured market was thus not a viable alternative for firms wishing to enter the SVR market. Nor indeed was the lower end of the insured market either at that time. If the non-insured market was not a viable alternative we must now consider whether it was possible to enter the insured market without VESA approval. It is common cause that SAIA represents the bulk of the short term insurance industry. This was conceded by Da Silva.’ It was thus not viable for a firm to enter into this market backed only by insurers who were outside of SAIA. No evidence was led that any insurer broke ranks to do so except for the efforts of de Meillon of Auto and General. Even Auto and General had, whilst interested at one stage in Tracetec, relied on Matrix as a partner. Thus Tracetec was not looked at in isolation but through the lens of an incumbent. No more than uninsured motorists would non- SAIA insurers want to consider non-VESA firms unless they were considering a joint venture of the sort contemplated by Auto and General, which in any event did not work out. It is common cause that the SAIA insurers would not deal with any non-VESA approved firm. Nor did they approve of any diluted form of approval, as the history of the ill-fated provisional approval category demonstrated. We conclude that it was not possible for a firm to expand in the SVR market at the time without having its product approved by VESA in the SVR category. Thus the issue narrows down to whether the performance standard was reasonable. We are mindful of the fact that we should, as a non-expert body, be cautious about pronouncing on the reasonableness of an industry standard. “other providers” as being approximately 90%.Dominium suggest that from their experience the recovery rate would have been closerto 20%. (File E 514). '88 Da Silva states that there are 110 licences issued in the industry and that SAIA has only 50 members however those 50 represent the main traditional ‘man in the street’ insurers, Transcript 1806-7. 72
[264]
[265]
[266]1267] However various factors point to the fact it was unreasonable. In the first place during the period that the approval standard was in place only two firms acquired full approval, they being Bandit and Global Telematics. Bandit did so in circumstances that would make its position unique. It was already in the market prior to the standard being adopted, was a member of the committee as we noted despite being only provisionally approved and was alleged to have used its provisional approval status to hold out that it had full approval. All this suggests that the experience of Bandit is insufficient to show that any efficient competitor could enter. Global Telematics like Bandit we know was around at least in 1999 as it is mentioned in VESA’s first press release in February that year as being approved in the fleet management category. It applied to join SVR in May 2001 but fell short of the standard and was eventually only admitted in about June 2002. It too was accused at one stage of misrepresenting its status as its sin was to have mentioned VESA approval pending in its advertising. Given the time it took it to become a member and the fact that like Bandit it was in the market prior to the standard being adopted it hardly serves as an example of successful entry into the market. On the contrary this is evidence of a firm whose entry was delayed considerably by the enforcement of the standard. Contrast that with the position when the financial guarantee alternative was offered. It may have taken some time after this was approved before firms took this up, but when they did the contrast was significant.'** Holland says in his witness statement that in 2005 six companies entered the insurance sector, substantially increasing competition.” In VESA board minutes the chairperson of the meeting is quoted as stating after the adoption of the financial guarantee alternative that there were three new firms that had been admitted since August 2003 and that four more were in the pipeline for 2004.2°" 189 At the SVR meeting on 7 October 2003 Car Track is recorded as being the first to enter membership by way of guarantee. 200 Holland witness statement paragraph 56. He names the three new as Cartrack, Digicor and Mobile Tracker. Knocking on door are LST, Augtech and Mobile Vision. LST and Augtech are described as falling short on number of 13
[268]
[269]
[270]
[271]
[272]
[273]The performance standard was also self-serving as it was designed to suit the SVR respondents’ business models and not necessarily those of other entrants. Cell Stop as we noted had specifically marketed a niche product which it considered viable as a higher end product Cell Stop’s refusal is significant because the only firms present at the meeting where its application was refused were the three SVR respondents.* But the most damning evidence against the reasonableness of the standard comes from the SVR respondents’ own experience of getting new products approved. As we have seen, all three struggled to get their new products approved in terms of the standard and sought special exceptions for themselves. Because new products for existing members had to be approved it is significant that even these firms with their advantage as incumbents, the strength of their brands and existing approvals for other products, could not climb their own self- constructed hurdles without tinkering with the standards to suit themselves. Interesting too is the statistics we have in the record of the two strongest firms’ new products attempts to reach the target. Netstar’'s Sleuth had its evaluation done in Oct 2000. Its first product was installed in December 1999. At the time of the application for approval a total of 3371 units is recorded as having been installed but recoveries are only at 23 out of total loss of 29.7 In July 2003 Tracker wrote to VESA asking it to distinguish in its statistics between its Alert and Retrieve products. At the time of writing it had installed 6000 Alert products, although it is not clear how long this had taken it to units in operation but are technically sound and in process of raising the guarantee. File E 2008 .19-20 Mobile Tracker attends its first SVR meeting on 3 June 2004. E 2181.1 Mobile Data (not to be confused with Mobile Tracker) is admitted at meeting of 3 August 200. File E 2238. 202 File E 959. 203 File E 389-90. 74 achieve. However it had received 26 activations of which 24 had been recovered.” Thus way short of the 100 recoveries imposed on new entrants.
[274]Statistics Matrix and Bandit produced of the progress in the market of their new products show how difficult it was to get to the three thousand units installation and 100 recoveries threshold.?°°
[275]lronically had they applied the standard rigidly to themselves their newer products would have struggled in the market. This indicates how in an innovation market the standard existed to exclude new technology and benefited older technology -- something that could never redound to the benefit of consumers.
[276]The underlying rationale for the performance standard was that it would prevent so called fly- by- nights from entering the market and discrediting the industry if it was littered with the corpses of failed start ups so that consumers no longer thought these products credible. The mantra of the SVR firms and VESA was that the criteria ensured that firms were able to survive in the market. Yet their own critique of these standards later undermines them.
[277]Joss stated that the standards would be exclusionary if VESA’s brand equity was strong - something he was not willing to accept. *°° Joss had to make this concession in his evidence because he had already questioned the rationality of the standard in the correspondence we considered earlier thus he was forced to argue that VESA did not enjoy strong brand equity. But VESA considered that its brand was strong and stated this on numerous occasions. 207 So too did Da Silva.? 204 File E 1192.106 . 205 For Bandit see File E 283 where it figures for September 1999 to March 2000 show a move from 1785 to 2883. The total incidents during this period number 40 and there are 35 recoveries. For the same periods the new Maitrix products MX3 goes from 1306 to 1576. There are 13 incidents and 12 recoveries. For the Matrix Eco the figures for this period are from 1533 to 1989, of these there were11 incidents and 9 recoveries. File E 283-4. 206 Transcript page 1450, Joss stated: “! believe that | expressed a view that | believe that they had the potential under certain conditions, which were never realized to become anticompetitive. VESA’s brand equity was never strong enough. They never had enough support from the insurance industry for my concern to become a reality..” See also 1576. 207 (Note at this meeting a minuted comment that VESA brand equity is currently perceived to be strong1189). 75
[278]Joss’s view on VESA ‘s brand equity is belied by the conduct of his Matrix colleague Rampton, who at one stage suggested that VESA interdict Tracetec from misrepresenting its VESA approval status — hardly the remark of someone 209 who did not believe VESA enjoyed brand equity.
[279]The fact that the three SVR respondents sought to undermine VESA when they left shows that they considered it a credible threat to their interests because it would now enable people to enter the market more easily than if there were no VESA i.e. VESA’s approval standard had metamorphosed from a barrier to entry to a facilitator of entry.
[280]Rampton also mocked the 100 recovery standard, suggesting at one VESA board meeting that was he required to steal the vehicles to achieve this target. 210
[281]Members of the VESA board were also on record querying the logic of the standard. At the dawn of the standards committee the statistician quoted in the minutes had suggested that statistical sample of 500 would suffice. Thus undermined there was no proponent for why the figures adopted were a rational considered standard. Joss himself said on the other hand even complying with the standard was no guarantee a business would succeed and pointed to Datatrak as an example of a firm that would have met the standard but failed in the market.2"' Yet the Datatrak debacle was given as the reason for the need for a standard. These contradictions were not answered by the SVR respondents who did not call their witnesses. De Clerk and Edemeston were poised to testify but at the 11 hour did not. Given their central role in the (Note that at a meeting of what is termed the Bureau Service Workshop which takes place on 3 November 1999 and includes VESA staff the SVR members, except Tracker, other unknown companies and members of the police, Ms Steed of VESA indicates that VESA approval is greater than ever with the insurance industry and that insurers will only now use VESA members. (File E235). During final argument Mr Bromley for VESA stated, “VESA was a very substantial organisation, it had a huge brand, it still has a huge brand in terms of the vehicle security market.” Transcript page 2202. 208 File E 648 and File E 1884. 29 File £1751. 210 File E 649.9 transcript of meeting quotes Rampton about waiting for 100 thieves to steal the vehicles so | can go and recover them do | go and pay them so it goes quicker. 211 Transcript 1575. ADV BERGER: Yes you've said that the criteria would certainly have prevented a fly by night but wouldn't have prevented a Datarak. JOSS: Yes. 76
[282]
[283]
[284]committee their evidence to defend the standard would have been crucial. It no doubt would have been difficult for them to do so and hence they were not called. Their post-resignation damning of the VESA standard would have been difficult to explain. De Clerk’s preference in the final moments before the committee’s denouement for letting the free market decide would have not sat easily with his earlier adherence to maintaining the standards. Even the reliability of the VESA statistics, the basis on which firms’ performance criteria were assessed for the purpose of approval, were put into doubt. 7" It is also questionable whether the standard VESA chose by measuring performance quality by the statistics chosen were not self-serving as these were the figures that incumbents could muster to their advantage. Yet informed constituents such as brokers questioned their accuracy and in one case advocated a rival non-approved firm because its systems monitored a vehicle at all times, not only once notified as being stolen.?"* Others suggested other deficiencies with the approved products e.g. Tracetec suggested that its own product was superior because it did not depend on the stolen vehicle’s battery for power.2"* We do not choose to evaluate the truth of these claims, only to suggest that the choice of performance criteria was by no means uncontroversial. The SVR respondents might contend that they still believe in stronger standards but that they were not allowed to prevail because of the VESA board’s pusillanimity in the face of threats of litigation. However the time for their rethink of the standard occurred when the financial guarantee option was allowing new firms to enter. Prior to that as the authors of the standards they had no need to consider them inadequate. The three SVR companies did not publicly assert that the standards were irrational prior to the adoption of the 2'2 In a Tracker minute it was stated that as there is no formal and reliable review system De Clerk expressed his concern regarding the accuracy of VESA figures. At an SVR meeting on 3 February 2004 the issue of statistics was discussed again and much scepticism expressed with regard to reliability. Van Zyl records that at meeting with insurers the latter had felt that recovery rates were closer to 30% and not industry claimed rate of 80%. Van Zyi wanted VESA to randomly check insurance members’ recovery details. (File E 1922). Dominium Insurance brokers, as we have seen, suggested that claims of 90% were not accurate and considered the figure closer to 20% (E 514). 213 Claim made by Dominium Brokers. File E 512. 214 Transcript 448. FileA 662. v7
[285]
[286]
[287]financial guarantee alternative. After that they publicly criticised not only the provision of the guarantee but also the rest of the standards. Joss in his testimony was no defender of the reasonableness of the standard. Da Silva at best can be described as lukewarm. She testified “. it might not have been perfect, but it was certainly rational to our minds.” *° But Da Silva never defends the choice of standard since she was not able to do so. She made it clear during her testimony that SAIA never got involved in the technical details of the specification because as she put it “.. we trusted VESA to handle those issues for us.” 7‘° What she regards as rational is the idea of having a standard but she cannot be relied upon to defend the choice of standard. She was neither an expert on these matters nor as we have seen an active participant in the committee’s deliberations. It was left to James Hodge to be the only defender of the standards’ reasonableness — and as an economics not an industry expert he was in no position to credibly do so. We conclude that the standards had an exclusionary effect, and that the allegations that they were self-serving and irrational have been convincingly made. The agreement thus led to a substantial prevention and lessening of competition in the SVR market. We reject the argument that the standards were set by consumers, finding instead that they were the product of agreement between competitors. Once competitors have been found to set a standard there is at the very least an evidential onus on them to justify that they had not set an exclusionary standard. This the three SVR respondents have manifestly failed to do. For reasons set out above the SVR respondents having failed to defend the reasonableness of the standards have been unable to show any other technological efficiency or other pro-competitive gain from the standard. On the contrary the effect of the standards was to condemn consumers to higher prices and deny them the benefit of new technologies that would otherwise have entered this innovation market far earlier than they did. 215 Transcript page 1802. 218 Transcript 1898. 78 Liability period
[288]The Commission as we noted considered that once the financial guarantee alternative had been offered the standard was no longer exclusionary. This conclusion is largely premised on the entry that takes place post August 2003 when the financial guarantee becomes available. Tracetec persisted in their view that this alternative was still exclusionary.
[289]The origin of the financial guarantee being set at R2 million until the firm met the performance standard, was the proposal by Edemeston of Netstar in May 1998, later endorsed by Caroline Da Silva at an SVR meeting the following November as an alternative to provisional membership. *” ,
[290]The motivation for the guarantee is not consistent. It was suggested by some that it was intended to show that the guarantor was financially sound. Others had suggested it was there to compensate customers of the firm providing the guarantee if it failed. On this version the guarantee would be used to pay for the installation of a new system and the payment of the balance of instalments paid in advance.?*® However the guarantee amount was uniformly applied to all applicants regardless of the cost of installation of their product and instalment fees, costs which we know were not uniform in the market.
[291]Tracetec, we have seen, objected to paying the guarantee and was concerned that its finer details had not been properly thought through. Others, like LST, had asked for the guarantee to be set lower or to be payable over a longer _ period, but the SVR members decided not to grant that request.2"° 2° See File E 73 and File E237. 218 This rationale for the 2 million is given at a VESA board meeting in 5 May 2004, when the chairman was being questioned on this point by an unidentified, but it seems new board member. The chairman said the guarantee was to pay the new installer the costs of new installation if the guarantor failed; this included as well the monthly instalments paid for the balance of the contract. Interestingly he mentions who those firms might be who would be paid to install the new equipment for the unlucky motorist: “say a Netstar Sleuth, Tracker Retrieve or Matrix MX1”. _. (File E 2008.31). The chairman also stated that there would be no pro rata refund of the guarantee - a firm had have met all criteria before the guarantee could be withdrawn. (File E 2008.24). 219 File E1920. 79
[292]If the guarantee requirement was still exclusionary in nature it is certainly arguable that it was arbitrary and hence not reasonable and thus not a justifiable limit on competition. However Tracetec has failed to demonstrate that notwithstanding the guarantee the standards for approval were still exclusionary. For instance it failed to show why a firm could not make this payment and do an analysis on what the costs were for a new entrant firm. But the strongest evidence of its now non-exciusionary nature is the number of firms who then entered after the guarantee was offered as an alternative. It was also the view of the SVR three, informed analysts of the market as we have seen that the new alternative had considerably lowered barriers to entry. On this issue in the absent of better evidence to the exclusionary nature of the guarantee requirement, we conclude that it was not exclusionary.
[293]What then was the period for which the standard was exclusionary? In February 1999 the first approvals were announced and the existence of the standards and their endorsement by the insurance industry was made public. Whilst it is clear that the standards had been adopted by the SVR committee sometime earlier, probably September 1998. All of this is academic however. The reason is that the Competition Act only came into force from 1 September 1999 and hence this makes that date the earliest date on which the prohibited practice can be regarded as being in existence.
[294]The existence for a brief period of time of the provisional membership standard did not mitigate in any way the standard for full membership, because as we have seen provisional membership was not something meaningful and when Bandit tried to rely on it for marketing it was chastised.
[295]The exclusionary nature of the standard persisted at times becoming more drastic in its effect, until the end of August 2003 when the financial guarantee was Offered. For this reason we regard the period in which the standard operated in an exclusionary manner as extending from 1 September 1999, till August 2003. Liability of VESA |
[296]Tracetec as we mentioned earlier has sought to hold VESA liable as well. 80
[297]This is not surprising given that the SVR respondents in their answers to the Commission’s referral, which were filed before the Tracetec statement of complaint, placed the blame for the standards on VESA, not themselves. This was thus one of their defences. VESA however has played the same game and blamed the SVR sub-committee. In its heads of argument VESA states: “It is further clear from the evidence that the VESA Board attempted, following the lobbying by new entrants who have subsequently proven themselves to be more than fly-by- nights in the SVR market, to encourage the SVR committee to review their specifications. {It is further clear from the evidence that the First, Second and Third Respondents refused to act on request or the concerns communicated by the Board of the Fourth Respondent.””?°
[298]Whilst to some extent VESA hangs on the coat tails of its erstwhile members to defend the standards, to the extent that these are found to be anti-competitive, VESA’s defence is that the sub-committee was responsible for the standards and that it did not accede to this standard.
[299]Whilst one can have sympathy with VESA which throughout the relevant period demonstrated ambivalence towards its wayward SVR _ committee, it nevertheless failed to act to rein in the committee and assert its executive authority over it. VESA wanted to be associated with the tracking industry as it recognised its centrality over the other technologies which it had traditionally been associated with. Yet it was too institutionally weak to control the powerful and sophisticated companies who populated SVR. it also saw _ ‘their membership as financially beneficial to itself and a means to sirengthening its brand. Its ill-fated attempt to get SVR companies to have VESA certificates of approval is an indication that whatever VESA’s reservations about the legality of the approval standard it saw material benefits arising from having the SVR companies in the organisation despite the risk of exposure fo litigation.
[300]Whether VESA institutionally approved the standard in the sense that it received board ratification is irrelevant as an issue in competition law. The 220 VESA Heads of Argument paragraph 4.1 81
[301]
[302]
[303]organisation was aware of its sub-committees decisions, allowed them to hold them out as those of the organisation and furthermore allowed its secretariat to participate in the enforcement of the standards. That the dog allowed its tail to wag it is no defence open to the dog. © In certain circumstances associations of competitors are liable in competition law precisely because they are used to enforce horizontal agreements amongst competitors. By allowing the three SVR respondents through first the T&R committee and later the SVR committee to set a standard of approval that was anti-competitive and to enforce that standard using VESA’s name and organisational structure, VESA is liable. Section 4(1)(a) has an effects based test. It matters little whether the association followed correctly its internal procedures in reaching a decision. What matters is the effect on the market of its decision. In European law, from which the language of this subsection is borrowed, the term ‘decision’ has even been held to apply to a recommendation.2" VESA made a true lawyer’s point in its heads of argument despite being unrepresented.?”” It argued that Tracetec in its statement of complaint had contended that VESA was liable because it was a party to an agreement or concerted practice. Recall that section 4{1){a) lists three bases for liability under this section, an ‘agreement’ , a ‘concerted practice’ or a ‘decision by an association of firms’. VESA argued that at no time was it alleged by Tracetec that it was a party to a decision by an association of firms or a concerted practice. 72° However in its heads of argument, Tracetec made the basis of VESA’s liability that its performance criteria were “a decision of an association of firms”.224 VESA argued that there was no evidence that it entered into any agreement of concerted practice concerning the standards and hence could not be liable on this portion of section 4(1)(a). Further that it could not be held liable on the 221 See Richard Whish Competition Law 6th Edition page 103. 222 ESA was initially represented during the first part of the hearings by attorney and counsel but later advised that for financial reasons it could no longer afford this legal representation for the balance of the hearing and so was represented by its staff members. 223 VESA heads paragraph 2.4. See Tracetec statement of complaint paragraph 215.1 where the conclusion of law is pleaded which VESA seeks to rely on. 224 Leads of argument paragraph 189.2. 82 basis that it had taken a “decision as an association of firms” as that portion of section 4(1)(a) had not been specifically pleaded.**°
[304]We do not think much turns on this distinction. A decision by an association is
[305]
[306]by its very nature a decision made by a collective of firms and whilst constitutionally, depending on the nature of the association, this may manifest itself as a decision of a single entity, the point of significance for competition law liability is that it has been arrived at by the aggregation of interests of individual firms who compete. In that sense the distinction between a decision of a firm and an agreement is such a fine one as not be of any moment. It should also be noted that the factual basis on which VESA was sought to be liable was made out in the statement of complaint and this has not changed during the course of the proceeding. 7° VESA, notwithstanding this technical objection, nevertheless also denies that there was any ‘decision’ by it. As we have shown the law does not require the formalism that VESA seeks to imply a ‘decision’ must meet to qualify to be one. As Whish puts it commenting on the understanding of the same language as it has been interpreted in the United Kingdom: “The term ‘decision’ has a broad meaning, including the rules of trade associations, recommendations, resolutions of the management committee and rulings of the chief executive; the crucial issue is whether the object or effect of the decision is to influence the conduct or co-ordinate the activity of the members.’”2”
[307]What one is interested is in the effect. VESA as we have shown meets this test. The standards: [307.1] were set as a result of decisions of its sub-committee which, on its own version, it allowed to set the standard;””® 225 VESA does not make this point in these express terms, but this is what we understand it to be arguing. 228 See for instance paragraph 13-14 of the statement of complaint. 227 See Whish op cit page 333. 228 See VESA heads of argument paragraphs 1.27-9. 83 [307.2] became its by- laws by which members’ admissions were assessed; [307.3] were implemented as we have seen by its officials; and [307.4] were, apparently, endorsed for some period of time by its executive, who even when it, at the eleventh hour, expressed concerns about the standards, did very little to see to their removal.
[308]Despite all this evidence VESA chose not to lead any witnesses nor did it put to Joss, who was the only one of the SVR respondents to testify, what its version in final argument was, namely that if the standard is found to be anticompetitive it [VESA] *... was not a knowing, willing participant to such agreement or practice but at the utmost was an instrument in the hand of the other three Respondents.”*”°
[309]Nor is it true that the VESA was entirely innocent of what the standard setting was meant to achieve. In April 2002, the VESA board heard a report back from a VESA official on his department’s activities. The following is noted in the minutes: “Members agreed that the admittance of too many stolen vehicle recovery companies could be detrimental to consumers. VESA should be extremely circumspect when considering granting membership to new applicants.”?°°
[310]Thus, as this extract shows, VESA knew that the standards were restricting entry. It simply chose to delude itself into believing that this restriction was in the consumer interest.
[311]VESA also seeks to make much of the fact that it is a non-profit association. This does not absolve it of liability for that reason alone. Our Act makes no 229 ESA heads op cif, paragraph 1.33. Curiously at the same time VESA contend that it is not its case that the V&R committee did manipulate, mislead or defraud it. It’s difficult to follow what it is trying to state here- that if the standard is lawful it has not been manipulated, but if it is it has? 739 File E 602. 84
[312]
[313]
[314]exclusion for non-profit associations in section 3 the applications section. Nor do comparable systems exempt non-profits from the ambit of their anti-trust laws.7*" VESA for these reasons is also liable in term of section 4(1)(a) for the anticompetitive effects of the SVR standard. Having found that the respondents are liable in terms of section 4(1)(a) it is not necessary for us to consider the argument that the respondents are also liable in terms of section 8(c).752 Nothing turns on us deciding in favour of one contravention rather than the other. If the same contravention is proved in terms of one section of the Act it seems superfluous to make a finding on the same factsin terms of another section of the Act, unless different relief is consequent. But the relief would not be any different if we had found in terms of section 8(c) not section 4(1{a) and vice versa. Both the Commission and Tracetec seek only a declaratory order and neither argued why it was necessary to make a finding in terms of both sections for that declaration to be effective. On the contrary bar the recitation of the section, the terms of the declaration would have been the same. The Commission argued that it wanted to create jurisprudence that collective dominance is recognised under our Act. The SVR respondents argued vigorously against its applicability stressing the manner in which the Act’s section 8 language differs from its counterpart in European law, from where the concept of collective dominance is derived. We will decline the Commission’s invitation and leave the issue undecided. We make no finding of liability of any of the respondents under 8(c), given that we have found the same conduct to 231 See California Dental Association v Federal Trade Commission, No. 97-1625, 1999 U.S. LEXIS 3606 (May 24, 1999) and Cimenteries CBR SA v Commission of the European Communities (T-25/95) [2000] 5 C.M.L.R. 204. In the latter case the court held; “it is not necessary for trade associations to have a commercial or economic activity of their own for Article 85(1) E.C.to be applicable to them. Article 85(1) applies to associations in so far as their activities or those of the undertakings belonging to them are calculated to produce the results which it aims to suppress. To place any other interpretation on Article 85(1) would be to remove its substance. [1320]. 232 Section 8 (c) states that: “It is prohibited for a dominant firm to engage in an exclusionary act,....if the anticompetitive effect of that act outweighs its technological, efficiency or other pro-competitive gain”. 85 have contravened section 4(1)(a) and that this would have no effect on the nature of the remedy. PART F: CONCLUSION
[315]All four defendants are found to have contravened section 4(1)(a). They are liable for the costs of Tractec jointly and severally including the cost of two counsel which given the length and complexity of the matter is justifiable. ORDER
[316]In terms of section 58 of the Act the Tribunal declares: [316.1] That the actions of the first, second and third respondents in concluding an agreement and/or engaging in a concerted practice, amongst themselves, inter alia, which had as its effect the setting and subsequent implementation of a performance standard for admission of firms to membership of the SVR category of VESA during the period September 1999 to August 2003 is found: [316.1.1] to be exclusionary in its effect and hence fo substantially prevent or lessen competition in the market for SVR products by preventing competitors from entering into or expanding in the market, denying consumers the benefit of lower prices, greater choice and technological development; [316.1.2] Not to be shown to be outweighed by any technological efficiency or other pro-competitive gain; and [316.1.3] To be a contravention of section 4(1)(a) of the Act. [316.2] That the decision of the fourth respondent, being an association of firms which had as its effect the setting and subsequent implementation of a 86 performance standard for admission of firms to membership of the SVR category of VESA during the period September 1999 to August 2003 is found: [316.2.1] to be exclusionary in its effect and hence to substantially prevent or lessen competition in the market for SVR products by preventing competitors from entering into or expanding in the market, denying consumers the benefit of lower prices, greater choice and technological development; [316.2.2] Not to be shown to be outweighed by any technological efficiency or other pro-competitive gain; and [316.2.3] To be a contravention of section 4(1)(a) of the Act. 316.3] That the respondents, jointly and severally, are liable for the costs of the intervenor (Tracetec), on a party and party basis, including the costs consequent on the employment of two counsel. / 19 April 2010 Norman Manoim DATE Tribunal Member Yasmin Carrim and Lawrence Reyburn concurring. Tribunal Researcher Romeo Kariga For the Intervenor Adv Chris Loxton (SC), assisted by Adv Robin Pearse, instructed by Knowles Husain Lindsay Inc For the Commission Adv Danny Berger (SC), assisted by Phillip Mokuena, instructed by the State Attorney For the First Respondent : Adv John Campbell (SC), assisted by Adv Anton 87 For the Second Respondent : For the Third Respondent For the Fourth Respondent Gotz instructed by Bowman Gilfillan Attorneys Adv Jerome Wilson, instructed by Werksmans Attorneys | Adv Mike Van der Nest, assisted by Adv Alfred Cockrell and Adv Gretha Engelbrecht, instructed by Cliffe Dekker Hofmeyr Attorneys Adv Sarel Wagener (SC) instructed by Koekemoer Attorneys - for the first part of the hearing in 2008. Then Louise Taljaard from VESA for the rest of the hearing 88
Read from the court's own list. Judgments are official texts; verify against the court's file before you rely on this page. Firms on AttorneyOS get judgments like this one against the matters they touch, in the Monday edition.