---

COMPETITION TRIBUNAL OF SOUTH AFRICA

Case No: 18/X/APR10

In the matter between:

MEDIA24 LTD ….....................................................................................First Applicant

ABRAHAM PETRUS VAN ZYL …..........................................................Second Applicant

And

COMPETITION COMMISSION OF SOUTH AFRICA …...................................First Respondent

COMPETITION COMMISSIONER OF SOUTH AFRICA …..........................Second Respondent

BERKINA TWINTIG (PTY) LTD, TRADING AS GOLDNET

NEWS …...........................................................................................................Third Respondent

HANS STEYL …..............................................................................................Fourth Respondent

Panel : Norman Manoim (Presiding Member), Yasmin Carrim (Tribunal Member)

and Andreas Wessels (Tribunal Member)

Heard on : 09 June 2010

Order issued on : 08 July 2010

Reasons issued on : 08 July 2010

Reasons for Decision

1. The applicants have brought this application to set aside a summons issued by the second respondent, the Competition Commissioner (the ‘Commissioner’) addressed to the second applicant Abraham Van Zyl (‘Van Zyl’) in his capacity as Chief Executive of the newspaper division of the first applicant, Media 24 Ltd, (‘Media 24’).¹

2. The attack on the summons is that it is ultra vires the powers of the Commissioner in that it is void for vagueness and contains impermissible interrogatories.

Background

3. The summons has been issued by the Commissioner in pursuance of an investigation by the first respondent, the Competition Commission (‘Commission’), into a complaint of an alleged prohibited practice perpetrated by Media 24. In January 2009, Hans Steyl, the fourth respondent, a director of the third respondent Berkina Twintig (Pty) Ltd (‘Berkina’), which published a newspaper called Goldnet News, lodged a complaint with the Commission against Media 24. Steyl alleges that Goldnet News, a weekly local newspaper, competed for advertising in the so-called Free State Goldfields area with two publications owned and managed by Media 24 called Vista and Forum.² In about 2004/5 he alleges Media 24 cut its advertising rates for Vista and Forum. Although Berkina tried to respond to these cuts by reducing Goldnet News’ rates it was not able to do so as the rates would have been below the costs of production.³ Eventually, it is common cause, Berkina was forced to close down Goldnet News in April 2009 and subsequently, Media 24 closed down Forum in January 2010.⁴

4. The Commission commenced investigating the complaint. In July 2009 the Commissioner issued a summons addressed to John Davis, the General Manager of Media 24. Although this summons, which we refer to as the first summons, is not the subject of these proceedings, it is relevant insofar as it is linked to the chain of events that led to the second summons which is.

5. In the introductory note to the first summons it is alleged that the Commission is investigating allegations of a contravention of section 8(c) and 8 (d)(iv) of the Competition Act (“the Act”).⁵ It goes on to state that the substance of the complaint is that Media 24 through Vista and Forum had abused a dominant position in the market by selling local and national advertising space at a highly discounted rate which is totally unrelated to the cost of production and normal overhead costs of a free local newspaper. It goes on to allege that the advertising rates for Media 24’s publications in Bloemfontein, Bethlehem and Kroonstad where they (Media 24) don’t face competition are market related.

6. A long series of requests for documents and information then follows. In respect of Vista and Forum the series of questions relates to costs, the attribution of costs into categories, revenue sources of income and so forth. The documents and information requested date back to 2001, and include strategic documents, minutes, research and reports to management. But the documents requested were not confined to these two papers. Documents were also sought of Media 24, inter alia requesting advertising rates and ‘achieved prices’ for all their community newspapers.⁶ The latter were defined as publications containing at least 40% editorial content and which were free to the reader. Again the requested documents and information date back to 2001.

7. Following a meeting between the applicants’ attorneys and the Commission the request was narrowed both in relation to time periods and in respect of some customer information.

8. The information was then supplied apparently to the Commission’s initial satisfaction.⁷ Subsequent to the first summons, the Commission wrote two letters requesting further information from Media 24. The second letter seems to have been the proverbial straw that broke the camel’s back. Media 24‘s attorneys complained that they did not understand why some documents were being requested. They also requested more time to compile the information.⁸

9. The Commission replied explaining its reasons for requesting the information. Firstly, it wanted to compare prices in regions where Media 24 has market power to regions where it does not, because it wanted to analyse whether community newspapers constitute a separate relevant market. It also wanted to analyse whether predation had taken place by examining whether Media 24 could recoup its losses incurred during the predation period. It also wanted to establish whether recent prices in the Welkom area had increased more than in other areas since the exit of Goldnet News. The Commission then granted an extension of time for the production of information. This elicited another letter from Media 24’s attorneys again debating the relevance of some of the information requested.⁹

10. The Commission’s response was, through the Commissioner, to issue the second summons on 30 March 2010, which is the subject of this application.

11. The applicants have mounted several attacks on the second summons.

12. The first is that the summons contains impermissible interrogatories. The complaint here is that the summons contains questions requiring information from Van Zyl which the applicants allege would not be susceptible to answer by way of documents, but only by Van Zyl answering questions. This they say is impermissible because the questions are framed in the ‘document request’ section of the summons.

13. It does not seem to be the argument of the applicants that the Commission may not use a summons for the purpose of interrogatories; rather that because the interrogatories appear in a section that the Commission has defined as one pertaining to documents, the request is unlawful. Put differently what the applicants appear to be arguing is that once the Commission has said the schedule relates to a document request it may only be used to request documents and not solicit other information.

14. The structure of the summons is such that the first three schedules comprise an introductory note, a definition section and an instruction section, whilst the fourth and final schedule, contains the requests for documents and information.

15. The applicants point to the first page of the summons, where the fourth schedule is first referred to, and note that its language limits itself to a request for documents. To quote the relevant passage “The documents which the Company is required to deliver to the Commission are specified in Part IV of this annexure below.”

16. Whilst on a strict reading the Commission is asking the interrogatories in the fourth schedule which as we have seen it described earlier as relating to ‘documents’, it is worth noting that the fourth schedule is the only place in which questions are asked of the addressee and that the schedule is headed “Documents to be submitted and information requested”.(Our emphasis) Van Zyl who is the addressee of the summons does not complain that he was confused by the location of the interrogatories nor could he sensibly have done so. Nor has this point of complaint been raised on the papers but it only surfaced afterwards during argument. Nor did the applicants query with the Commission whether the questions were requests for documents or information by way of interrogatories. Indeed the summons’ instruction section permits enquiries to be made about its terms.¹⁰

17. This objection is without substance and based on a selective reading of the terms of the summons. Even if one characterises the questions as interrogatories they would be permissible in terms of the authorising statute which empowers the Commission both to request documents and submit the person summoned to answer interrogatories.¹¹ This leaves as the only point of objection that the interrogatories were not placed in some separate schedule headed interrogatories. This is a purely formal complaint about how the summons is structured and given that the architecture of the summons did not confuse the applicants this objection must accordingly fail.¹²

18. The more substantial attack on the second summons is that the information sought is void for vagueness.

19. It is not necessary to repeat the contents of the summons because the attack on it is thematic. What the Commission seeks through a series of questions is information on community newspapers in other geographic locations owned by Media 24. The information relates to:

• Circulation figures for certain community papers named in the summons for a period from January 2001 to the present date;

• Rate data in disaggregated form;

• Strategy documents and board minutes for community newspapers with similar weekly circulation figures to Vista and Forum (There follows a list of the names of those publications the Commission considers meet this definition although the list is not considered exhaustive of papers in this category); and

• Information on new entry of the five most recent Media 24 community papers that have entered the market. This information request relates inter alia to sunk costs, the time period for the publications to have broken even and the ‘scale’ at which they became effective competitors.

20. The applicants argue that the remainder of the second summons, i.e. those parts that don’t constitute the interrogatories, is unintelligible and overbroad. The nub of the attack is that information is being sought about publications outside of the Free State Goldfields area, cannot be relevant because the predation is alleged to have occurred in the Goldfields market.

21. The applicants argue that any rate comparison across geographic areas is meaningless given the variance in conditions. They attach to their papers a confirmatory affidavit from an economist explaining why comparisons across geographic markets suffer from limitations because of many variables that exist between geographic markets that relate inter alia to population demographics and differences in local competitive dynamics. The conclusion is that:

“Given these factors, comparing advertising rates in different areas without considering the full range of other factors that influence such rates will produce statistical results that cannot be competently used for any alleged reasons put forward by the Commission.” (Our emphasis)¹³ The applicants suggest that this conclusion is common cause and hence the request is unintelligible. This is not a fair reading of the Commission’s position. What the Commission concedes is that various reasons may explain variation in advertising rates across regions but that is the point of its summons – to enable it to have the necessary information to come to conclusion on this issue.

22. What Media 24 is contending for amounts to us coming to a conclusion that a comparison of rates for the same product in other geographic markets can never be relevant in case where the dominant firm is accused of predation in respect of that product in a particular geographic market. But the basis for suggesting that it is not relevant to the investigation of the complaint is the applicants’ expert's assumption that this information could not produce statistical results that could be competently used. This is the essence of this part of the objection; a methodological assumption about an exercise that has yet to be performed based on a factual premise that Media 24 expects us to accept without the benefit of trial and evidence.

23. It would require remarkable self assurance for us to adopt such a categorical approach. Furthermore, the relative strength of various types of evidence in reaching a final decision about a predatory allegation may significantly differ from case to case and therefore it is imperative that each individual case be assessed at the actual hearing and not on mere methodological assumptions on which more than one economic expert may very well disagree. It may well be that the comparisons ultimately prove odious, but that does not make the attempt to do this exercise illegitimate. The economist has no basis to conclude a priori that the exercise can never be statistically competent.

24. It is well established in antitrust economics that evidence in pricing cases, be they excessive or predatory is notoriously difficult because evidence on costs and their relationship to prices in a particular market is not always susceptible to precision as it depends largely on how a firm accounts for them. Therefore, no immutable rules exist in competition economics for analysing such cases, since most instances involve intricate issues of both economic and accounting judgment. Therefore comparisons are often made of the sale of the same or similar product in other geographic markets in order to assess whether conclusions can be reached by way of inferences drawn. Indeed this has been the approach to excessive pricing cases where too the argument can be made that variations occur across geographic markets. Yet notwithstanding this the Competition Appeal Court (CAC) in Mittal recognised this as a legitimate means of comparing prices to ascertain if a price was excessive.¹⁴

“Prices ordinarily charged locally in other markets by the same firm or by other firms with broadly comparable cost structures at comparable levels of output, may obviously serve as a measure of the ‘economic value’ of the same good or service in our market ...”¹⁵

25. In the same decision the CAC quotes the approach of the United Kingdom’s Competition Appeals Tribunal who observed in the Napp case that:

‘Measuring whether a price is above the level that would exist in a competitive market is rarely an easy task. The fact that the exercise may be difficult is not, however, a reason for not attempting it. In the present case, the methods used by the Director are various comparisons of (i) Napp’s prices with Napp’s costs, (ii) Napp’s prices with the costs of its next most profitable competitor, (iii) Napp’s prices with those of its competitors and (iv) Napp’s prices with prices charged by Napp in other markets. Those methods seem to us to be among the approaches that may reasonably be used to establish excessive prices, although there are, no doubt, other methods.’ ¹⁶ (Our emphasis)

The CAC went on to observe that:

Evans and Padilla, in their discussion of various policies towards the prohibition of excessive pricing by dominant firms, emphasise the ‘conceptual as well as practical difficulties’ of determining what constitutes an ‘unfair’ price for purposes of Article 82 of the EC Treaty. Due to the complexity of the exercise more than one method is employed under Article 82. Primarily, a comparison between the actual price and the costs of production is made but, where this is not possible, the price can be compared to prices in comparable markets In comparing prices the European Court makes use of different comparator prices. ¹⁷

26. Although predatory and excessive pricing are not the same thing, both entail coming to conclusions about the relationship of price to costs. Thus in pricing cases of both kinds a comparative analysis of related markets may be highly relevant.

27. In general terms information in regard to an alleged predatory firm’s prices and costs in different geographic markets relating to the same product or service could conceivably either allay fears of or reinforce a likely predation hypothesis. For example: a firm that is charging similar low prices in all of the geographic markets in which it operates is highly unlikely to be predating in any one geographic market. On the other hand a significant variation between the market where the alleged predation occurs and another series of markets may give rise to an inference that the applicant is charging below cost. Indeed this is precisely what the complainant Steyl alleges.¹⁸ Of course this is not a complete answer or conclusive evidence that a contravention has taken place, but this is not what we have to decide now. What we can conclude is that a comparison of pricing information and costs for a similar product in other geographic markets to the one in which the predation is alleged to have occurred is a legitimate investigative exercise for the Commission to perform. The actions of a predatory firm in markets related to the relevant market in which the alleged predation takes place (i.e. in the instant case other potential geographic markets of the same relevant product market) may be relevant to inter alia:

1. a better understanding of the overall pricing and profitability strategy of the alleged predatory firm, as well as the rationality and feasibility of a predatory strategy in any one particular market;

2. justification for the incumbent firm’s lower prices and/or costs in one market compared to others, considering for example the relationship between market demand (i.e. the relative size of the relevant market) and relevant costs and prices;

3. the allocation of costs relating to any multi-market activities of the alleged predatory firm; the financial constraints that the alleged predatory firm faces (i.e. the ability to sustain losses in the relevant market in question as a result of activities in other markets);

4. the financial constraints that the alleged predatory firm faces (i.e. the ability to sustain losses in the relevant market in question as a result of activities in other markets);

5. potential cross-subsidisation from other (more profitable) markets where the alleged predator may face lesser or no competition; and

6. the potential benefits flowing to other markets from any reputational effect in the market under consideration, i.e. the recoupment of profits in markets other than the one in which the predatory behaviour occurs.

28. Therefore, an analysis of prices and costs across potential separate (i.e. localised) geographic markets where the alleged predator faces varying competitive constraints, seems a reasonable and relevant exercise in a predation context. We cannot at this stage categorically assume that the information required in the summons is not statistically competent.

29. Of course this assumes that the Commission has as yet determined the boundaries of the market. It emphasises that is has not – that both the product and geographic market boundaries are issues it must still investigate. For instance it would not be clear if community newspapers of the kind in issue constitute the relevant product market or whether the so-called Goldfields area is the relevant geographic market.

30. The information sought by the Commission in the second summons is relevant to both these exercises. The Commission’s rationale for seeking the information is both intelligible and within an orthodox approach to investigation of such cases.

31. The remaining attack on the summons is that it is overly broad as to region, time and particularity. Here the same argument is repeated that because the focus is on local competitive dynamics in the Free State Goldfields area any enquiry into publications nationwide “leads nowhere”. We have already explained how such an enquiry may lead “somewhere” and hence it is not necessary to repeat these arguments. It suffices to emphasise that it is not common cause, as the applicants would have it, that the relevant enquiry for information should be confined to the Free State Goldfields area.

32. The Commission’s response to the criticism about the lengthy time periods contemplated in the summons is also reasonable. Although some questions require information dating back to 2001, (the circulation figures for three newspapers circulating in the Western Cape) others date back to 2003 or only from the date of the summons (July 2009) to date. The Commission argues that since the alleged predation occurred in 2004/5 it is necessary for it to have information for the periods before this in order to come to conclusions on whether pricing was predatory. The information about rates required up to the present date is necessary to evaluate the possibility of recoupment occurring, as the complainant has exited the market as has one of the Media 24 publications in the Goldfields area. Again the justification for the selection of the various time periods appears well reasoned.

33. That of course does not mean that the request for documents and information - even accepting that a comparison with other regions, where the applicant circulates similar publications is legitimate – can be unbounded. The applicant suggests rhetorically in its heads of argument that the link between the change in pricing of its publications since 2003 in say Langa, Gugulethu and Athlone is too far removed to merit a resort to summons.

34. Taken out of context of the investigation that might sound reasonable, however the Commission did not come to this selection randomly as an examination of the course of the investigation shows. In the first summons the Commission asked Media 24 for:

“List prices (as published in a rate card or otherwise) containing advertising rates for all Media 24 community newspaper publications applicable between January 2001 and the present date. Community newspaper publication means any newspaper publication containing at least 40% editorial content, which is free to the reader.” ¹⁹

35. We were advised from the Bar by Mr Peterson appearing for the Commission that the list of publications contained in the second summons was yielded by this question. In other words the list of publications constitutes products the Commission considers comparable to the ones allegedly used to perpetrate the predation strategy in the Goldfields area. We were advised further that similar information sought was already given by Media 24 in respect of papers in the Western Cape Province, Eastern Cape, Northern Cape and some parts of the Free State.²⁰

36. Thus there is a rational connection between the information sought and the analysis the Commission seeks to undertake.

37. The final category of question in the summons requires the applicants to disaggregate information supplied pursuant to the first summons.²¹ What the Commission seeks is specific financial information at a publication level which is relevant to its comparison exercise as opposed to regional totals, which are not meaningful for the purpose of the assessment. This request is both reasonable and relevant given the exercise which the Commission is performing in benchmarking. Aggregated information is not helpful in this regard.

38. Of course one cannot be without sympathy for the applicants given that the request for information is extensive. The applicants complain that it will take as many as 500 personnel hours to procure it.²² Whilst it is unfortunate that firms that are the subject of an investigation are often put to considerable inconvenience to assemble documents to comply with a summons that burden does not invalidate the request contained.

39. We find that the application is without merit and it is accordingly dismissed.

40. There is no order as to costs.

_________________________________ 08 July 2010

DATE

Norman Manoim and Andreas Wessels.

Yasmin Carrim concurred.

Tribunal Researcher : Mahashane Shabangu

For the Applicants : Advocate Schalk Burger S.C. instructed by Werksmans

Incorporating Jan S. De Villiers

For the Respondents : Hylton Petersen of the Competition Commission

---