CC Computicket

The judgment

The court's own file

COMPETITION TRIBUNAL OF SOUTH AFRICA Case No: CROO8Apr10/DSM022May11 In the Dismissal Application matter between: COMPUTICKET (PTY) LTD Applicant And COMPETITION COMMISSION Respondent In Re: The Complaint Referral between: COMPETITION COMMISSION Applicant and COMPUTICKET (PTY) LTD Respondent Panel : Norman Manoim (Presiding Member) Anton Roskam (Tribunal Member) Andreas Wessels (Tribunal Member) 1 Heard on : 05 September 2016 Order issued on : 21 October 2016 Reasons issued on : 21 October 2016 Decision and Order Introduction

[1]

[2]

[3]

[4]

[5]In this case Computicket Limited (“Computicket’) seeks an order reviewing and setting aside a decision made by the Competition Commission (the “Commission’) to refer a complaint against it. The review relies on two self-standing arguments; the decision to refer has been made by the wrong person because it was made by the Commission not the Commissioner; and the referral, even if made by the competent person is reviewable, as it offends against the principle of legality. Computicket accordingly seeks an order from the Tribunal setting aside the referral. This matter comes to us on review following a lengthy history of prior litigation between the two parties relating to whether Computicket was entitled to certain documents in the Commission's possession which the former sought to ground its review. That litigation is now settled and its resolution has ultimately favoured Computicket as the disputed documents are now in the record. Since the documents are now before us a lengthy consideration of that history is no longer relevant. We therefore do so briefly. Background [6]During the period 2008 the Commission received complaints from five separate complainants, all rivals of Computicket, alleging that Computicket was engaged in anticompetitive practices by securing exclusive agreements with 2

[8]

[9]entertainment providers. The Commission consolidated these complaints into one case, investigated it, and eventually on 30 April 2010 referred the complaint to the Tribunal.‘ In the referral the Commission alleges that Computicket has contravened sections 8(d)(i), alternatively 8(c), and/or section 5(1) of the Competition Act, no 89 of 1998 (“the Act”). Computicket filed an answer to the referral on 30 June 2010 and the Commission filed its reply on 30 July 2010. The matter was then set down for hearing from 18-29 July 2011. By this stage pre-trial preparations were well under way. The Commission had made discovery and filed its factual witness statements; its expert witness statements were to follow. Then came a skirmish over discovery. On 04 April 2011 the Commission filed an application calling upon Computicket to produce further documents. Computicket did not answer this discovery application. Instead on 12 May 2011, Computicket filed an application to dismiss, what, with some mutations, has become the source of the present application. The history of the litigation becomes immensely complicated thereafter. Since it is fully set out in an earlier decision of the Tribunal there is little purpose in repeating it now.? What is relevant now is that as a prelude to the dismissal application, Computicket subsequently brought an application for discovery of the internal documents that had supported the Commission’s referral. The Commission opposed this application. That application for documents, inter alia, became the subject of an interlocutory hearing before the Tribunal. The Tribunal dismissed the application to produce but Computicket appealed it to the Competition Appeai Court (CAC). 1 Prior to receiving these complaints the Commission had, in 2007, received a complaint of a similar nature from another firm that has since been sequestrated. This complaint as we understand it was never referred. See record page 654. 2 See Tribunal decision in Computicket (Pty) Ltd v Competition Commission; case number; 20/CR/Apr10; pages 3-5; paragraphs 6-18. 3 The Tribunal also had to decide certain other points but they are not relevant in casu. 3

[10]On appeal the decision of the Tribunal was set aside and the CAC ordered production of documents that had served before the Commissioner, subject to a qualification which restricts internal deliberations of the Commission from the obligation to discover them.4 The Commission then sought to appeal this decision to the SCA. It was unsuccessful in obtaining leave from the CAC, and then appealed the latter refusal to the SCA, where too, it met with no joy. Thus the CAC’s discovery order stood. A dispute then raged over the ambit of the order. It did not, despite its emotion, lead to further litigation, but eventually - the Commission agreed to provide Computicket with the report that served before the Commission and which formed the basis of the referral. The Commission claims it did so even though it was not legally obliged to do so — the Commission views this document as falling within the class of exception carved out by the CAC order. However it contended that it volunteered discovery to avoid further prolonging the dispute.”

[11]We need not decide whether the Commission has made a virtue of what may have been a necessity or made a bold concession to stop the impasse. The point is that we now have the report and it forms the centre piece of the current debate in this matter as that was the document on which the referral decision was based.

[12]The reason it is the centre piece of this review is because of two common cause facts. The referral decision was taken by the Commission and the report was the only document that served before the Commission when it was made. Both these facts found the two grounds of this review — was the decision made by the correct person, and if so, did the report and the manner in which it was 4 Computicket (Pty) Ltd Limited v The Competition Commission of South Africa; case number; 118/CAC/Apr12 (29 October 2012, Swain AJA). 5 See Competition Appeal Court decision in Competition Commission v Computicket(Pty) Ltd; case number; 118/CAC/Apr12(20 September 2013); also see Supreme Court of Appeal decision in The Competition Commission v Computicket (Pty) Ltd; case number; 853/2013. § This report is to be found in pages 652 to 719 of the record. 7 See Commission’s heads of argument paragraph 4.9. 4 considered constitute a proper basis for a decision to refer. We go on to consider these two arguments. First ground of review Whether the decision maker was the one authorised by the Act to have made the decision

[13]In terms of section 50(2) (a) of the Act a decision to refer a complaint by a complainant must be made by the Commissioner.’ We set out section 50(2) below : 13.1 50(2) Within one year after a complaint was submitted to it, the Commissioner must— (a) subject to subsection (3), refer the complaint to the Competition Tribunal, if it determines that a prohibited practice has been established.

[14]However many other provisions of the Act require decisions to be made not by the Commissioner i.e. an individual who is an office bearer, but by the institution which the Commissioner heads i.e. the Commission. The Act defines the term Commission in section 19(2) as follows: 14.1 “The Competition Commission consists of the Commissioner, and one or more Deputy Commissioners, appointed by the Minister in terms of this Act.

[15]The life cycle of a complaint that gets referred under the Act goes through two phases. Depending on its origin, and the phase, different legal actors are responsible for the execution of the complaint. In the first phase, what is termed ® In this section we underline the last letters in the word “Commissioner” to more easily distinguish that word from the word “Commission” to make the arguments addressed, easier to follow. 5 ‘initiation’, the complaint may be generated by either the Commissioner or any person, the latter is referred to as the complainant.®

[16]The second phase is the referral. With a complainant initiated complaint there are two possible outcomes that may lead to a referral. Either the referral is made by the Commissioner’? or if the Commissioner decides not to refer the complaint, by the complainant itself."

[17]Where the Commissioner has initiated the complaint the complaint can only be referred by the Commission.‘? (There is no error in our usage of these terms as the reader might think. Rather this is what the Act states and hence the interpretation issue we have to resolve in this case).

[18]It is common cause that the decision to refer in this case was taken by the Commission not the Commissioner acting alone. It is also common cause that when we refer to a decision made by the Commission, on the present facts, it was a decision made by the Commissioner and the then Deputy Commissioner." It is also common cause that the complaint being referred in this case was one that had been initiated by a complainant, not by the Commissioner. With these facts, all of which are common cause, we can now consider the legal argument made by Computicket.

[19]Computicket argues that a referral by the Commission as opposed to the Commissioner does not meet the requirements of section 50(2)(a) and is hence of no force and effect. As authority for this proposition Computicket relies on ® Section 49B(1) for the Commissioner and 49B(2)(b) for any person . Section 1(1)(iv) defines a complainant as the person who has submitted a complaint in terms of section 49B(2)(b). 10 Section 50(2)(a). 1" Section 51(1). 12 Section 50(1). ‘3 The Act allows for more than one Deputy Commissioner to be in office. ( Section 19(2)) However there is no evidence that there was more than one Deputy Commissioner serving in office at the relevant time. 6 several cases where courts have set aside orders or decisions where they have not been made by the official authorised by the statute to do so."

[20]The Commission argues that the point is without substance. It is clear it says that the reference to the Commissioner as opposed to the Commission is a drafting error. In support of this argument it relies on other sections of the Act which, when read together, suggest that the legislature had presumably contemplated a reference to the Commission, rather than the Commissioner, in section 50(2) (a).

[21]Further, the Commission argues, this is not a case where a junior official has usurped the prerogative of a more senior one to make a decision which the statute does not authorise the former to make. Understandably such a decision would be set aside. However this case is not a case of usurpation by one official, not authorised by a statute of another, who has been authorised; rather it is the Commission, as an institution, which includes the Commissioner that has made this decision. There is no dispute of fact that the Commissioner supported the referral.

[22]Of course the inconsistent use of the terminology in the statute is bewildering. Is it attributable to design or error? To answer this we consider the history of the section and the context of the other corresponding provisions to which it relates.

[23]The present section 50(2) (a) was not in the Act originally. It was inserted later as a result of an amendment made in 2000." Prior to that its predecessor was the then section 50(a), which combined both referral considerations into one section which read as follows: 14 See Molefe v Dihlabeng Local Municipality [2003] ZAFSHC 35; [2003] ZAFSHC 9 (5 June 2003) at paragraph 35; Shidiack v Union Government 1912 AD 642 at paragraph 648; Sigaba v Minister of Defence and Police 1980 (3) SA 535 (Tk) at paragraph 541. +5 Amendment of section 50 of Act 89 of 1998, as amended by section 15 of Act 39 of 2000. 7 23.1 50(a). “After completing its investigation, the Competition Commission must- Refer the matter to the Competition Tribunal, if it determines that a prohibited practice has been established’;

[24]However part of what is now in section 50(2)(a) appeared at that time in the then Commission rules 19(2) and 19(4). We quote these two rules: 24.1 19(2) Subject to sub-rule (3), the Commission must either refer a complaint to the Tribunal in Form CT(1), or issue a Notice of Non-Referral in Form CC8, no more than one year after the date on which the complaint was initiated or accepted, as the case may be, as reported to the Tribunal in terms of sub-rule (1). 19(4) If the Commission has not referred a complaint to the Tribunal, or issued a Notice of Non-referral, within the time allowed by sub-rule (2), or sub-rule(3) if different, the Commission will be deemed to have issued a Notice of Non- referral on the expiry of the relevant period. (Our emphasis)

[25]What is notable about both the old section 50(a), and these prior rules, is that when a complainant's complaint was referred the designated decision maker was the Commission not the Commissioner. This rule was then removed from the Commission’s rules in 2000 when section 50(2) (a) of the Act as it is now, was introduced. The old section 50(a) of the Act, quoted above, was deleted. The outcome of these changes was that the Commissioner replaced the Commission as the decision maker for the purpose of referring a complaint from a complainant. Was this change in designation deliberate, as suggested by Computicket or an error that came about as a result of the redrafting of the amendments, as contended by the Commission?

[26]Let us look at the logic of the Act to try and answer this question. As noted earlier, where a Commissioner generated complaint is referred, the referral decision in terms of section 50(1) is made by the Commission. (27] But with a complainant's complaint section 50(2)(a), as noted, designates the Commissioner for this function. Is there any basis for this distinction — that one should be the prerogative of the Commission and the other the Commissioner? If it was intended it is hard to see a policy rationale for it. A decision to refer a complaint, whatever its genesis — requires the functionary charged with making that decision to exercise the same discretion conforming to the same legal standard. There is no policy reason why the one decision should be that of the Commission and the other of the Commissioner.

[28]The absence of an argument based on design as opposed to error is illustrated when we consider other sections in Part C of the Act which sets out the procedure for complaints.

[29]The most obvious is section 50(3) which provides as follows: 29.1 50(3) When the Competition Commission refers a complaint to the Competition Tribunal in terms of subsection (2) (a), it— (a) may— (jrefer all the particulars of the complaint as submitted by the complainant; (i)refer only some of the particulars of the complaint as submitted by the complainant; or (iij)add particulars to the complaint as submitted by the complainant; and 9 (b)must issue a notice of non-referral as contemplated in subsection (2) (b) in respect of any particulars of the complaint not referred to the Competition Tribunal.

[30]This subsection provides for four different options open to the decision maker all of which have a consistent theme — they are consequent upon the decision by the Commissioner to refer a complainant’s complaint. Logically, according to the argument based on design, all four of these decisions should be made by the Commissioner not the Commission. Yet all four options are given to the Commission not the Commissioner. Again divining design as the rationale for the distinction between who makes the primary decision and makes the consequential decisions pursuant to the primary decision makes no sense. Again this points to the likelihood of a drafting error.

[31]The same point is illustrated with regard to sections 50(4) and 50(5) which we set out below: 31.1 50(4) In a particular case— (a) the Competition Commission and the complainant may agree to extend the period allowed in subsection (2); or (b) on application by the Competition Commission made before the end of the period contemplated in paragraph (a), the Competition Tribunal may extend that period.

[32]Then there is section 51(1) which states: 32.1 “If the Competition Commission issues a notice of non-referral in response to a complaint, the complainant may refer the complaint directly to the Competition Tribunal, subject to its rules of procedure.” 10

[33]Here the likelihood of a drafting error pursuant to the 2000 amendment becomes even more probable. A notice of non-referral only arises in the context of a complainant's complaint. These two sections refer to the subsequent management of that complaint.

[34]Logically all these decisions in the argument based on design should have been given to the Commissioner since they are interwoven with the decision to refer or not to refer contemplated in 50(2). Yet the text makes the Commission the design maker. Clearly there is no rationale for this and it points to a drafting error. {35] But a case for error is best illustrated by section 50(5) which states: 35.1 “If the Competition Commission has not referred a complaint to the Competition Tribunal, or issued a notice of non-referral, within the time contemplated in subsection (2) or the extended period contemplated in subsection (4), the Commission must be regarded as having issued a notice of non-referral on the expiry of the relevant period.”

[36]Here there is a direct reference back to section 50({2) yet the functionary referred to is not the Commissioner, as the argument for design must require, but the Commission.

[37]Again, only error, not design, can sensibly account for this anomaly.

[38]That a drafting error has crept in is also illustrated by an aberrant “ft” that has crept into section 50(2). Why would the Commissioner be referred to as if? Computicket suggested that this was a drafter’s attempt to avoid resort to having to use a personal pronoun (it being considered a better option than the 11 undesirability of choosing between ‘he’ or ‘she’, or the inelegance of using both). But if the adoption of the impersonal pronoun was simply the drafter’s convention to avoid this Hobson’s choice, one would have expected to encounter it elsewhere in the Act - but one does not. This despite the fact that the term Commissioner appears frequently in the Act (See for instance section 22, where despite extensive reference to the term Commissioner the drafter has not resorted to either it or the use of a personal pronoun.)

[39]The Commission’s argument that this is an error is the only plausible explanation which is supported by — i) The history of the amendment and its predecessor which refers to the Commission; ii) The related sections referred to above, all pointing to the likelihood that this power was intended to be bestowed on the Commission, not the Commissioner acting alone, and where no sensible explanation has been offered to suggest why it could have been intended that like functions should have been granted to different functionaries to perform; and iii) the pronoun “it” used in section 50(2) (a) which is more suggestive of an intention to have used the term Commission rather than Commissioner.

[40]We are not bound to interpret statutes literally. Case law permits the interpreter of legislation to have regard to the possibility of a drafting error.'® [41]We are satisfied that the reference to the Commissioner in section 50(2) is a drafting error and that a referral by the Commission (which as an institution in any event includes the Commissioner) is legally competent. This point of review is therefore dismissed. 16 Bothma-Batho Transport (Edms) Bpk v S v Bothma en Seun Transport (Edms) Bpk 2014 (20 SA 494 (SCA), paragraph 10. 12 Second Ground of Review Whether the decision maker applied its mind to the matter of referral

[42]The second part of the review challenges the decision itself.‘” The challenge is both to the process i.e. how the decision was made and the factual basis on which it was made. The dispute here between the parties is one of both fact and law. We consider the legal issues first.

[43]There is agreement between the parties on at least the following legal principles as they arise from previously decided cases in the SCA. e First that a decision by the Commission to refer a complaint does not constitute administrative action reviewable under the Promotion of Access to Justice Act No. 3 of 2000 (“PAJA’).'8 e Second, that despite not being reviewable under PAJA, a decision to refer can be reviewed under the principle of legality.19 e Third, that the Tribunal in terms of section 27(1) (c) of the Act, has the powerto review decisions of the Commission, including a decision to refer a complaint.2°

[44]However apart from these Issues there is no consensus on the extent of the applicable legal standard. According to the Commission because the decision is not reviewable under PAJA but only under the principle of legality, the 17 For ease of reference, given the fact that the Commission and not the Commissioner was the decision maker and that we have found that this was competent, we will fom now on, refer in this section to the decision as having been made by the Commission. This despite the express language of section 50(2)(a), which, as we have seen, refers to the Commissioner as the decision maker. 18 See Simelane & Others NNO v Seven-Eleven Corporation (Pty) Ltd and Another 2003 (30 SA 64 (SCA); Competition Commission of South Africa v Telkom SA Ltd and Others [2010] 2 All SA 433 (SCA) paragraph 11; Competition Commission v Yara SA (Pty) Ltd & Others 2013 (6) SA 185 (26 November 2014) paragraph 18. *8 See Competition Commission v Computicket (Pty) Ltd (853/2013) [2014] ZASCA 185 (26 November 2014). 20Sibanye Gold Ltd v Competition Commission [2015] 1 CPLR 324 (CT). 13 grounds for review are a fortiori narrower and are confined to whether the decision was ultra vires, irrational, unconstitutional or taken in bad faith. 21

[45]Computicket disagreed. ft argued that a legality review can incorporate both rationality and reasonableness. It relies for this on a passage from the decision by the CAC in the discovery case. [In that decision Swain AJA writing for the court states: “However if the decision maker’s opinion is challenged on the basis that it was irrational, the decision maker must show that the subjective opinion it relied on for exercising power was based on reasonable grounds.” 22

[46]The learned judge relied on the decision of the Constitutional Court in the Walele case in this regard.?? It is worth setting out in full what the Constitutional Court stated in Walel/e in the passage quoted: “In the past, when reasonableness was not taken as a self-standing ground for review, the City’s ipse dixit could have been adequate. But that is no longer the position in our law. More is now required if the decision-maker’s opinion is challenged on the basis that the subjective precondition did not exist. The decision maker must now show that the subjective opinion it relied on for exercising power was based on reasonable grounds”

[47]Computicket relies on this passage of the CAC decision and its reliance on Walele as support for the adoption of the application of the reasonableness standard to the present case. However it is by no means clear that the CAC was doing so. Wale/e concerned a decision by a City Council which was subject to PAJA. 21 Here the Commission relies on President of the Republic of South Africa v South African Rugby Football Union 2000 (1) SA 1 (CC) paragraph 148; Masetlha v the President of the Republic of South Africa and Another 2008 (1) SA 566 (CC) paragraphs 76-81; Democratic Alliance v President of South Africa 2013 (1) SA 248 (CC) paragraph 27. 22 Computicket CAC decision (29 October 2012, Swain AJA) ibid paragraph 15. 23 Walele v City of Cape Town and others 2008(6) SA 129 (CC) at paragraph 160 A-C. 14

[48]Indeed what the CAC states on the debate if anything suggests otherwise. After discussing the law on the appropriate test. — which is the context in which the reference to the Wale/e decision arises, Swain AJA says the following: “The test to be adopted in the kind of application which appellants seek to launch may well be that of rationality rather than reasonableness. “+

[49]Swain AJA went on to quote from the Constitutional Court decision in the Democratic Alliance case where that court had explained that rationality entails asking whether the steps in the process were rationally connected to the end sought to be achieved. After referring to the Democratic Alliance case Swain AJA went on to observe that: “On the basis of this test, the question remains: to what documents are the appellants entitled to review the respondent's [the Commission’s] decision for lack of rationality.”25

[50]Again the learned judge here refers to rationality.

[51]However even if we are wrong in interpreting what the CAC meant here, what is clear is that Swain AJA did not think much turned on this debate for the purpose of this case. This much is made clear in the following two passages: 51.1 At paragraph 19 Swain AJA remarks: “The distinction between the jurisdictional facts which have to exist based on these opposing submissions are in my view, more apparent than real.” 51.2 Later on the same point is emphasised again at paragraph 20: 24 CAC Computicket (29 October 2012, Swain AJA) ibid paragraph 20. 25 Democratic Alliance v President of South Africa 2013 (1) SA 248 (CC). 15 “l can accordingly see no difference in substance between the competing submissions of the parties, as to the jurisdictional facts which have to be present, to constitute a valid determination by the respondent[ the Commission] that a prohibited practice has been established, which has as its consequence, a valid referral to the Tribunal.’26

[52]If the learned judge is correct that in the context of this case the differences in the tests are more “... apparent than real ...” we do not need to decide whether the grounds for review extend beyond rationality to reasonableness. We do so because we have decided that the review does not succeed notwithstanding which standard is applied. Facts on which the review is based

[53]It is now common cause after the lengthy and fractious discovery disputes that the Commission's decision makers relied on a single report prepared by its investigation team as the basis for its decision to refer the complaint.2”

[54]Since the review attack requires a distinction to be made between those who formed part of the investigation and those who took the referral decision, we will not use the term “Commission” but instead “investigators” and “decision makers”.

[55]The review attack on the decision can be summarised as follows: 55.1 The decision makers acted as a rubber stamp as they only considered this report prepared by the investigators and none of the underlying documents accumulated during the course of the investigation and which were available to it to consider. ?6CAC Computicket (29 October 2012, Swain AJA) at paragraph 20. 27 This report is to be found in the record at page 652, 16 55.2 Related to this first point is that because the decision makers only had reference to this document and not any of the others they failed to consider relevant facts and took into account irrelevant facts. 55.3 The third attack is one of bias. The investigators were determined to only consider those facts prejudicial to Computicket and that the tone and selection of facts underlies this. Since the decision makers were entirely reliant on the report for their decision this bias by extension taints their decision to refer. Analysis

[56]The essence of the Commission’s case as made out in the report is that « Computicket is dominant in the national market for “..outsourced ticketing services for entertainment events ...” and has abused this dominant position by engaging in the practice of requiring exclusive contracts from inventory providers.2° This exclusivity had prevented rivals from obtaining the necessary economies of scale to enter the market. The Commission considers that no credible pro-competitive or efficiency gains have been advanced to justify the conduct. It recommends, that the matter be referred and that as remedies, inter alia, the conduct be declared a prohibited practice, the exclusivity clauses in existing contracts be voided and a penalty be imposed.??

[57]As far as the process is concerned there is no dispute on the facts. The report was first presented to a meeting of the Commission decision makers on 27 October 2009.°° This body is known as Exco and comprises of the Commissioner, the Deputy Commissioner and the managers of the various divisions of the Commission. The report was presented by Liberty Mncube (“Mncube”) who headed the investigation team. He states that he presented the report and was questioned about its findings. Nevertheless no decision to refer 28 See report, record page 672. 29 Report ibid, record page 693. 30 Minutes of the Commission record page 775. 17 the complaint was made then as the Deputy Commissioner was not present at the meeting (the Commissioner and several others staff members were) and so the meeting was not considered quorate.*’ Nevertheless discussion of the report took place; the minutes show that the meeting suggested that the remedies be reworked and some put in the alternative and the report be resubmitted for a final decision.°?

[58]The final decision was made at a meeting held on 8 December 2009. At that meeting all the minute records is that a decision was taken that the case be referred. Both the Commissioner and Deputy Commissioner were in attendance.

[59]Mncube states that subsequent to the October meeting the investigating team reworked the report. The structure was changed and further thought was given to remedies. *4 This report, as now restructured, is the one that appears in the record. Since the remedies constitute the end for which a complaint referral is made, the fact that the Commissioner required more work to be done on the remedies is inconsistent with Computicket’s contention that the decision makers simply rubber stamped what the investigators had done.

[60]The next point of criticism was that Mncube, who had presented the report at the October meeting, was not present at the second meeting when the decision was taken in December. This fact only affects the Deputy Commissioner who had not been present at the first inquorate meeting; the Commissioner was present at both. [61]At the second meeting when the referral decision was taken the report was presented by Keith Weeks, then the head of the enforcement division, the one responsible for the investigation of prohibited practice cases.°° 31 See supplementary answering affidavit of Liberty Mncube, record page 734 32 See Minutes, record page 776 supra, and also Mncube affidavit, supra, record page 734-5. 33 See Minutes, record page 779. 34 Mncube affidavit record page 735. 35 Ibid record page 735. 18

[62]Computicket argues that Weeks could not have given a comprehensive briefing as he was not part of the investigation team. Computicket is in no position to state this. It does not know what Weeks knew or did not know and why it mattered that for the purpose of making the referral decision that because he was not part of the investigation team — he was in fact the head of the division for which they worked - he could not have dealt at the second meeting with the report.

[63]This point seems to have been taken without any factual basis to support it and arises solely because Weeks’s name is not one of those described as the investigators on the front of the report. The Commission has no formal procedure of who should submit a report or indeed if anyone needs to at alll. This point is without merit.

[64]The third process point is that the decision makers made the decision solely on the basis of the report and without having sight of the full record obtained during the investigation. The argument made by counsel was that any decision not made on the underlying documents was, for this reason alone, subject to criticism on the basis that it was dependant on the assessment of others.

[65]As a general proposition this is not necessarily always true. There is an equally convincing counter argument that for a busy institution like the Commission it is impractical for the decision makers to have regard to every document yielded in an investigation. Indeed a better decision might be arrived at if its staff competently sift through all the material received during an investigation and present only the relevant facts to the decision maker accompanied by an analysis of them.°¢

[66]As a specific proposition this criticism might have had cogence if Computicket could point to some deficiency in the report which a reader of the full record would have detected. Despite the fact that we asked counsel on two occasions during the hearing for such a ‘big fact”, save for the omission of the comments 36 See Mncube affidavit where he states that the Exco meets every week to consider investigation reports tabled before it “..among other tasks...”Paragraph 30, record page 734. 19 of their external expert Professor Motta which we deal with below - they were unable to do so. Thus as a specific proposition this point has no salience either.

[67]There is thus no basis for Computicket to assert that the decision makers failed to consider any relevant matter or considered irrelevant matter in coming to their decision.

[68]The fourth process point was to accuse the Commission of having made no fair and objective assessment; specifically an exhibition of bias or preconceptions.*”

[69]The basis of this attack seems to be that the Commission has disagreed with various submissions that Computicket made during the course of the investigation. As a threshold issue the Commission as a prosecutor is not bound to accept the respondent's version. However it is manifest from the report that the respondent’s arguments have been considered and evaluated, albeit that the investigators rejected them. Nevertheless where the arguments were rejected the report explains why.

[70]Computicket in its supplementary affidavit, testified to by its attorney, criticises the Commission for coming to conclusions on certain issues which are detrimental to Computicket. But here Computicket is in effect elevating the decision to refer, to an administrative decision under PAJA. It is unable to point to a relevant material fact omitted for consideration which would be favourable to Computicket or an irrelevant fact which would have been material for coming to the decision included in the report. The best Computicket can come up with is that it disagrees with some conclusions.

[71]But these are all conclusions of fact, economic theory and law that are properly matters for the hearing not the basis for a legality review. 37 Computicket’s further Supplementary affidavit, record page 638. 20

[72]We deal with one example of this as it is explored at some length in the supplementary affidavit; it deals with the reason for the exit of a competitor which for reasons of confidentiality we will refer to as A.

[73]The report identifies the fact that A had exited the market. The report goes further to state that the “primary reason” for A’s exit was that it could not reimburse customers who had bought tickets for a concert that got cancelled. However the report also quotes a representative of A, who claimed the exclusivity contracts with Computicket were the reason why it (the competitor) could not expand in the market or get new inventory providers.

[74]Computicket, whilst conceding the investigators had included this other fact in the report, nevertheless complains that the investigators should not have accepted the possibility that the exclusive contracts may have contributed to the demise.

[75]There are several problems with this approach by Computicket which are illustrative of the danger of deciding this kind of issue on review as opposed to resolving disputes of fact at trial. In the first place the investigators draw to the decision makers’ attention that that the “primary” reason for A’s demise was the reimbursement problem. This is inconsistent with suggestions of selectivity or bias as this is a fact favourable to Computicket. It is true the report also quotes the response of A to this suggestion. But this is a perfectly proper and fair approach. Nor is the investigators’ conclusion evidence of bias. The two reasons offered for A’s demise may not be mutually exclusive. It is not unreasonable to assume that a firm struggling in the market due to foreclosure, as was suggested to the Commission, would be less able to absorb commercial risks such as reimbursing customers for a cancelled performance.

[76]lf this was the best example of bias or lack of fairness that Computicket could offer it was unconvincing.

[77]Computicket also alleges that the Commission has failed to properly identify the relevant market. It alleges that the Commission’s relevant market is too 21 narrow and does not include firms which self-provide (i.e. book tickets for their performances themselves) or travel bookings. However the report reflects that there were these other candidates for market definition, but gives reasons for rejecting them.*® The reasons are argued and not made as categorical rejections. The decision makers were thus alerted to Computicket’s version and given reasons why its version should be rejected.

[78]Again this debate over market definition, often a central point of dispute in a competition case is properly the subject of a hearing not a review. Computicket will in due course be able to file both factual and economic expert witness statements and lead evidence in support of its view on market delineation. It will furthermore be able to cross-examine the Commission’s witnesses on matters related to market delineation.

[79]Finally, Computicket sought to make something of the fact that the report identifies a Professor Motta as an external consultant but the report makes no mention of what he may have told the Commission. Computicket offers this as an example of the failure by the Commission to “...conduct a fair and objective assessment of facts gathered in good faith and reporting them in a balanced way to the decision makers.”?°

[80]It is common cause that Professor Motta is an internationally renowned expert in competition economics. Mncube states that he had discussed the case with Motta and sought input from him “.. particularly regarding the formulation of the theories of harm’.4° [81]We fail to understand this point of objection at all. The Commission was not obliged to consult with an expert, nor if they did, to follow that advice. If Computicket's point is that having done so the decision makers should have been told of Motta’s advice that too does not make the decision reviewable. If Motta’s advice was consistent with the approach adopted by the investigators, 58 See pages 19-21 of the report, which can be found at pages 670-672 of the record. 39 Ibid record page 644. * Mncube affidavit, paragraph 27.4, record page 733. 22 mentioning that point might have been a matter of comfort, but presumably redundant.

[82]If Motta had given advice contrary to an approach contained in the report and this was suppressed from the decision makers’ attention by the investigators, would that constitute a point of review? We consider not.

[83]First, we have no evidence that this was the case and without it Computicket does not get out of the starting blocks.

[84]Second even if he did, the decision makers were not bound to follow his advice. The cover page of the report signals that he was a consultant so his role was not suppressed from the decision makers’ knowledge. Presumably if the decision makers wanted to know if he was of a different opinion to the investigators they could easily have asked. Indeed this may have happened. Mncube is silent on this point but there is no reason to expect him, deposing several years later, to recall if he was asked this, nor do the minutes, which merely record decisions not discussion.

[85]This point too is without substance.

[86]Computicket’s next point is closely linked to the previous point. Mncube states in his affidavit that Motta had given input on the formulation of theories of harm.*1

[87]Computicket however accuses the Commission of bias for having said through the statement of Mncube that it has theory of harm.

[88]Computicket appears to suggest that having a theory of harm amounts to coming to a conclusion of guilt before one has assembled the facts or assembling only those facts which fit the theory and ignoring those that don’t. 4 Mincube ibid. 23

[89]This is to misconstrue what is meant by a theory of harm. The identification and development of a theory of harm provides guidance to a competition authority as to how specific conduct may be analysed based inter alia on economic theory and principles. It leads to considering all the relevant facts including the facts both for and adverse to the respondent firm. It does not amount either to an a priori conclusion of guilt or innocence.

[90]As a prominent competition economist has explained having a theory of harm is considered best practice in competition analysis: “The last ten years have seen an increasing focus from competition authorities on articulating a theory of harm behind competition concerns. A theory of harm should be a) logically consistent, b) reflect the incentives that various parties face, c) be in line with the available empirical evidence, and d) articulate how consumers have been/will be harmed. This establishes a rigorous standard of proof, which has substantially improved the quality of enforcement across all areas of competition law. The requirement to present a theory of harm makes it much harder for internally inconsistent and speculative competition concerns to survive the process of assessment and highlights genuine competitive problems.” 4 (Our emphasis)

[91]Understood in this way the Commission's approach in adopting a theory of harm is indicative of its rigour not its bias. This point too is therefore of no substance.

[92]Finally, Computicket complained about the style of the report. It was written it complained, in a manner suggestive of its outcome; namely to recommend a referral. Given that this is the report of an investigative team whose task is to make a recommendation to the decision makers who have to decide on whether to refer the complaint or non-refer it, it is hardly surprising that the report adopts a point of view. ” Theories of harm - Hans Zenger 27/10/2011 Charles River Associates CRA. 24

[93]But the fact that the report adopts this approach does not mean that it lacks rigour or does not consider arguments raised by Computicket during the course of the investigation. [94]A fair reading of the report shows that the investigation team has considered each element that would have to be proved in a case, whether under section 8 or 5 viz. - e« what evidence the Commission has in its possession and from whom it emanates; e what the relevant market is, and whether Computicket is dominant in it; this conclusion is arrived at after consideration of whether the market can be defined more broadly as suggested by Computicket; the investigators give reasons why in their opinion, the broader definition is not justified; e why foreclosure is relevant, and then a consideration of the extent and duration of foreclosure; e whether the practice had anticompetitive effects and their extent; e why an efficient rival would be foreclosed; e Any justification that might exist for exclusivity. For instance was the exclusivity linked to a claim based on efficiency; and e The remedy sought is rationally connected to the harm apprehended.

[95]The report also considers comparative jurisprudence and explains why a case in a similar industry where the firm was not found to have abused its dominant position differs from the present case on the facts.*?

[96]Even though the report was the only document that served before the decision makers we are satisfied that it set out facts and conclusions that constituted a proper basis for reaching a determination that a prohibited practice had been established. 43 See reference to an Irish case, report pages 32-33, which can be found at pages 683-684 of the record. 25

[97]In so doing the Commission acted rationally. However even if the threshold test on review is reasonableness, which we don't believe on the present case law it is, we nevertheless conclude that the Commission also acted reasonably.

[98]The review is therefore dismissed. CONCLUDING REMARKS

[99]Before concluding we wish to make certain remarks about parties that use reviews too hastily instead of availing themselves of the opportunity to defend themselves at the trial. (100] In the first place deciding certain factual disputes on review under the guise of a test of reasonableness can easily lead to an elision between these supposedly distinct processes. The facts of this case highlight that what emerged ultimately as the review points were in fact disputes of fact, and what conclusions can be based on those facts. These are issues that should be decided through the hearing process not by way of review.

[101]A review is decided on papers. A hearing is not; it involves inter alia, the hearing of viva voce testimony, cross examination and full discovery. The hearing is therefore the superior process for resolving the disputes of fact, and conclusions about them that inevitably arise in competition matters.

[102]The Act was designed to separate the functions of prosecution and adjudication and have them undertaken by separate bodies, independent of one another. Little is to be gained except delay, expense and obfuscation, by an overly solicitous review regime that makes prosecutors function as adjudicators and adjudicators sit as review courts.

[103]Second the principle of legality, operates not just to afford protection to respondents. Complainants and customers have rights too, which must also be considered by a reviewing tribunal, including the right to have disputes run 26 to their conclusion, and not be terminated prematurely on untested facts in a review. Further, they are entitled to have them heard with expedition. Reviews lead to a proliferation of process further delaying the final resolution of matters. (104] Similar sentiments were expressed by the Tribunal in an earlier case of Novartis which dealt with the question of whether a decision to refer required the respondent to be given audi alteram partem. There we expressed the concern that administrative review can seriously compromise the efficiency of the system.” [105} The Tribunal’s approach to reviews of complaint referrals expressed in that matter was endorsed by the Supreme Court of Appeal in another matter, Simelane and others v Seven Eleven where Schutz JA observed: “I cannot do better than refer to what is said in the Novartis case. For the reasons stated there it is clear that ina case such as the one we are concerned with the function of the Commission is investigative and not subject to review, save in cases of ill-faith, oppression, vexation or the like. Seven Eleven should husband ifs powder for the contest before the Tribunal.”#5

[106]This should be good advice to Computicket in this matter as well.

[107]We have decided this review on the facts put forward as a basis for the review by Computicket. In doing so we have avoided considering whether we have strayed across the boundary from a strict rationality test into one of reasonableness. Sometimes that border may be blurred. The reason for adopting what may seem an overly permissive standard — if that is what it is — is the length of time this matter has already taken without reaching finality on 44 See Novartis SA (Pty) Ltd and others v The Competition Commission and others; Case number CT 22/CR/B. In paragraph 48 of the decision the Tribunal stated: “the administrative efficiency of the Commission in rendering its duties could be severely affected if, in exercising its discretion in terms of section 50(2), it every action would be subject to scrutiny under the principle of administrative review in the manner suggested by the applicants in this matter. 45Simelane NO and others v Seven-Eleven Corporation SA (Pty) Ltd and another [2003] 1 All SA 82 (SCA). 27 the merits. There seemed little profit in narrowly construing our remit only to have this already extensive litigation prolonged further. This matter had it gone to trial would have been heard five years ago.

[108]The case is therefore not authority for the proposition that the test for a legality review of a complaint referral has now been broadened to include consideration of whether the decision was reasonable. ORDER

[109]The application is hereby dismissed.

[110]There is no order as to costs. / 21 October 2016 Mr HloRMAN MANOIM Date Mr Anton Roskam and Mr Andreas Wessels concurring. Tribunal Researcher : Caroline Sserufusa For the Applicant : Mr J. Gauntlet SC and Ms G, Engelbrecht instructed by Werksmans Attorneys For the Commission : Mr G. Marcus SC, Mr J. Wilson SC, Ms. | Goodman and Mr. P. Ngcongo instructed by the State attorney 28