Aberdeen Journals Ltd v The Office of Fair Trading [2003] CAT 11 (23 June 2003) [2003] CAT 11

CAT

Aberdeen Journals Ltd v The Office of Fair Trading [2003] CAT 11 (23 June 2003)

[2003] CAT 11 · 2003-06-23

[1]GENERAL ............................................................................................... 119a. A. The relevant law ................................................................................. 119

b. B. The burden and standard of proof ...................................................... 123

c. C. The Tribunal’s approach to evidence ................................................. 126

[2]THE CASE MADE BY THE DIRECTOR .................................................. 135a. A. The characteristics of the products in question ................................... 136 The Press & Journal ............................................................................ 137 The Evening Express ............................................................................ 140 The Herald & Post ............................................................................... 142 The Independent .................................................................................. 145 Analysis of the characteristics of the product ......................................... 148 — Paid-for versus free ........................................................................ 151 — Weekly versus daily ....................................................................... 155 — Other differences ............................................................................ 162 — “Substitutes” and “complements” .................................................... 166

b. B. The commercial strategy of the Independent ....................................... 175 The position in 1996 ............................................................................ 176 “The Scottish Opportunity – Aberdeen" ................................................ 177 The launch of the Independent against the Evening Express .................... 183 The Independent ’s activities after launch ............................................... 186 Mr Robins’ letter of 8 March 2000 ........................................................ 190 The OFT meeting of 25 April 2002 ....................................................... 192 Was the Independent launched as a “fireship”, primarily in order to force a sale to Aberdeen Journals? ........................................................ 194

c. C. The commercial strategy of Aberdeen Journals ................................ 220 The Herald & Post as a “fighting title” ................................................. 220 The evidence about bundling ................................................................ 227 Mr Farquharson’s evidence ................................................................ 229

d. D. The further documentary evidence supplied by Aberdeen Journals .............................................................................................. 233 Aberdeen Journals’ letter to the Director of 10 February 2000 ................ 234 Appendix 2 to the letter of 10 February 2000 ......................................... 238 Other documents ................................................................................. 239

e. E. Did the position change in March 2000? ............................................. 248

f. F. Conclusion on the Director’s case ....................................................... 256 ii Paragraph

[3]THE CASE MADE BY ABERDEEN JOURNALS ..................................... 257a. A. Criticism of the Director’s approach to economic evidence ................ 257

b. B. The RBB reports ................................................................................. 284

c. C. The Tribunal’s analysis ...................................................................... 288

d. D. Conclusion .......................................................................................... 301

[4]ALTERNATIVE MEDIA .......................................................................... 302

[5]DOMINANCE IN THE MARKET FOR ADVERTISING IN LOCAL NEWSPAPERS IN ABERDEEN ................................................................ 309

[6]THE DIRECTOR’S ALTERNATIVE PRODUCT MARKET ...................... 314 V ARGUMENTS OF THE PARTIES ON THE ISSUE OF ABUSE ................ 327a. A. Aberdeen Journals’ submissions ........................................................ 328

b. B. The Director’s submissions ................................................................ 341

c. C. Aberdeen Independent’s submissions ................................................. 348 VI THE TRIBUNAL’S FINDINGS ON ABUSE ............................................... 349 A. The relevant law ................................................................................. 349 B. Some comments on the Director’s methodology in the decision .......... 360 The negative contribution of the Herald & Post ........................................... 362 The issue of printing costs .......................................................................... 373 The time period for assessing predation in this case ..................................... 382 The costs treated by the Director as variable ................................................ 388 C. Analysis of Aberdeen Journals’ arguments ........................................ 403 Was the Herald & Post sold at below average variable cost during March 2000? ........................................................................................................ 404 Was the Herald & Post sold at below variable cost in the last week of March 2000? ............................................................................................. 411 Did the conduct of the Herald & Post change materially before the end of March 2000? ............................................................................................. 416 Did Aberdeen Journals intend to eliminate competition after 1 March 2000? ........................................................................................................ 423 Was there a distortion of competition or a threat to the Independent in March 2000? ............................................................................................. 433 The issue of recoupment ............................................................................ 436 Is the period of the predation too short to constitute an abuse? ...................... 447 Was the Independent an in efficient entrant? ................................................. 450 US and Australian cases ............................................................................. 451

d. D. Conclusion .......................................................................................... 452 VII EFFECT ON TRADE WITHIN THE UNITED KINGDOM ....................... 453 A. Arguments of the parties .................................................................... 453 iii Paragraph B. The Tribunal’s findings ...................................................................... 456 VIII THE PENALTY ........................................................................................... 464 A. Arguments of the parties .................................................................... 464 B. The Tribunal’s findings ...................................................................... 476 The duration to be considered ..................................................................... 476 The penalty imposed .................................................................................. 478 Intentionally or negligently ........................................................................ 484 The gravity of the infringement ................................................................ 489 IX CONCLUSION ............................................................................................ 501 iv I INTRODUCTION 1. By an application dated 18 November 2002 Aberdeen Journals Limited (“Aberdeen Journals”) appeals to the Tribunal 1 under section 46 of the Competition Act 1998 (“the 1998 Act”) against a decision of the Director General of Fair Trading (“the Director”) 2 no. CA98/14/2002 dated 16 September 2002 (“the decision”) taken pursuant to section 18 of the 1998 Act. Section 18 provides: “18. –

(1) Subject to section 19, any conduct on the part of one or more undertakings which amounts to the abuse of a dominant position in a market is prohibited if it may affect trade within the United Kingdom.

(2) Conduct may, in particular, constitute such an abuse if it consists in– (a) directly or indirectly imposing unfair purchase or selling prices or other unfair trading conditions; (b) limiting production, markets or technical development to the prejudice of consumers; (c) applying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage; (d) making the conclusion of contracts subject to acceptance by the other parties of supplementary obligations which, by their nature or according to commercial usage, have no connection with the subject of the contracts.

(3) In this section– “dominant position” means a dominant position within the United Kingdom; and “the United Kingdom” means the United Kingdom or any part of it.

(4) The prohibition imposed by subsection (1) is referred to in this Act as “the Chapter II prohibition”.” 2. Section 18 of the 1998 Act is closely modelled on the corresponding provisions of Article 82 of the Treaty establishing the European Community (“the Treaty”). So far as possible, the 1998 Act is to be interpreted and applied consistently with the principles of Community law: see section 60. 1 By virtue of Articles 2 and 3 of the Enterprise Act 2002 (Commencement No. 2, Transitional and Transitory Provisions Order 2003, S.I. 2003 no. 766, with effect from 1 April 2003 this appeal is deemed to be made to the Competition Appeal Tribunal established under section 12 of the Enterprise Act 2002. Both the Competition Appeal Tribunal and its predecessor, the Competition Commission Appeal Tribunal, are referred to in this judgment as “the Tribunal”. 2 Under section 2(1) of the Enterprise Act 2002, brought into force by Article 2 of S.I. 2003 no. 766, the functions of the Director were transferred to the Office of Fair Trading (“the OFT”) on 1 April 2003, a body corporate created under section 1 of that Act. As from that date, the OFT has stepped into the shoes of the Director and is now responsible for the contested decision by virtue of Section 2 and paragraph 6 of Schedule 24 of that Act. For convenience this judgment continues to refer to the Director as the person who took the contested decision and conducted the proceedings up to 1 April 2003. 1 3. In the decision the Director found that Aberdeen Journals held a dominant position in the market for the supply of advertising space in local newspapers (paid-for and free) in the Aberdeen area, and that it had abused that dominant position contrary to the Chapter II prohibition. The abuse identified is deliberately incurring losses on the Aberdeen Herald & Post (“the Herald & Post ”) in an attempt to expel the Aberdeen & District Independent (“the Independent ”), its only rival in the relevant market, over the period 1 March to 29 March 2000. For that conduct the Director imposed a penalty on Aberdeen Journals of £1,328,040, pursuant to section 36(2) of the 1998 Act. 4. Aberdeen Journals is a subsidiary of Aberdeen Journals Holdings Limited, which is in turn owned by Northcliffe Newspapers Group Limited (“Northcliffe”). It appears that Aberdeen Journals was purchased by Northcliffe from the Thomson Corporation in early 1996 for a consideration of £82 million. Northcliffe is owned by Daily Mail & General Holdings Limited, which is in turn owned by Daily Mail & General Trust plc. The ultimate holding company for all of these companies is Rothermere Continuation Limited, based in Bermuda. 5. Aberdeen Journals publishes three newspaper titles circulating in Aberdeen: a paid-for daily morning title, the Press & Journal , which is also available in different editions throughout the north of Scotland; a paid-for daily evening title, the Evening Express , which is focused on Aberdeen and its outlying areas; and a free weekly title, the Herald & Post , 3 which is distributed mainly in urban Aberdeen. 6. Until 1996, Aberdeen Journals enjoyed a monopoly of local newspaper advertising in the Aberdeen area, through its ownership of the Press & Journal , the Evening Express and the Herald & Post .

[7]In March 1996, Aberdeen Independent Limited (“Aberdeen Independent”) launched the Independent as a high-quality weekly free newspaper in the Aberdeen area. The proprietor of Aberdeen Independent is Mr Keith Barwell, who is an experienced owner of free local newspapers. In the 1970s and 1980s Mr Barwell set up a large number of free newspapers which he sold in 1988 to the Thomson Corporation, for whom he subsequently worked in a senior position until 1990. At that time the Thomson Corporation also owned Aberdeen Journals. While he worked for the Thomson Corporation between 1988 and 1990 Mr Barwell 3 The Herald & Post has now been relaunched as the Aberdeen Citizen . For convenience we continue to refer to the Herald & Post . 2 was involved in launching the Herald & Post in Aberdeen, alongside the Press & Journal and Evening Express .

[8]It is common ground that Aberdeen Journals reacted to the launch of the Independent in 1996 by significantly reducing the advertising rates charged by the Herald & Post , substantially increasing the pagination (i.e. the number of pages per edition), and also increasing the number of copies distributed. As a result, the Herald & Post incurred substantial losses in the period from 1996 to mid-2000. Throughout almost all that period, including the month of March 2000, the revenues earned by the Herald & Post were insufficient to cover the cost of newsprint (i.e. the paper on which it was printed) and the cost of distribution, let alone any of the other costs associated with producing the Herald & Post .

[9]Aberdeen Independent submitted a complaint to the Director on 28 May 1999 alleging that the Herald & Post was pricing its advertising space at “massively below market value”. Aberdeen Independent subsequently expanded this complaint to allege that the Herald & Post was being used as a “defensive free”. As we understand it, “defensive free” is a term used in the newspaper industry to denote a free title the principal purpose of which is to deter competitive entry in a particular local area, notably in order to protect the revenues of a paid- for title circulating in that area. Thus, in its report on the proposed transfers of newspapers between Regional Independent Media Limited and Gannett UK Limited/Johnston Press plc/Guardian Media Group plc, Cm 4887, November 2000, the Competition Commission said at paragraph 2.25:

“2.25. We heard evidence that publishers of local newspapers fought hard to protect their markets from new entry and would, on occasion, maintain a loss-making free newspaper where this supported a paid-for title as part of a layered or segmented market.”

[10]The Director initially investigated the matter under the Competition Act 1980. On 1 March 2000 the 1998 Act came into force. The Director’s investigation then became an investigation under the 1998 Act to determine whether Aberdeen Journals had breached the Chapter II prohibition.

[11]On 23 November 2000 the Director issued a written notice (“the first Rule 14 notice”) to Aberdeen Journals in accordance with Rule 14 of the Competition Act 1998 (Director’s Rules) Order 2000 S.I. 2000 No. 293 (“the Director’s Rules”) stating that he proposed to make a decision that Aberdeen Journals had infringed the Chapter II prohibition by charging excessively low rates for advertising in the Herald & Post with a view to driving the 3 Independent out of the market. On the information available to him, the Director considered that the infringement continued at least for the period from 1 March until the end of July 2000, but was likely to continue until September 2001.

[12]In accordance with Rule 14(7) of the Director’s Rules, Aberdeen Journals submitted written representations to the Director on 2 February 2001 and attended an oral hearing on 15 February 2001.

[13]On 16 July 2001 the Director adopted Decision No. CA 98/5/2001 entitled “ Predation by Aberdeen Journals Ltd ” (“the first decision”). The first decision found that Aberdeen Journals had abused a dominant position in the market for the supply of advertising space in local newspapers (paid-for and free) in the Aberdeen area in the period from 1 March until 29 March 2000, and imposed on Aberdeen Journals a penalty of £1,328,040.

[14]On 14 September 2001 Aberdeen Journals appealed to the Tribunal against the first decision (“the first appeal”).

[15]On 19 March 2002, the Tribunal set aside the first decision on the grounds that the treatment in that decision of the relevant product market was inadequate: see Aberdeen Journals Limited v Director General of Fair Trading

[2002] CAT 4

, [2002] CompAR 167 (“the first judgment”), at [186], [189] and [197]. The Tribunal, however, considered that it was in the interests of justice that the proceedings should continue (first judgment, [187] to [192]). Accordingly the Tribunal remitted to the Director for further consideration “the issue of the definition of the relevant product market, in particular which newspapers constitute the relevant product market” and gave directions as to the timetable within which any further decision of the Director should be adopted (first judgment, [194]).

[16]Following the first judgment, the Director reopened the administrative procedure. On 20 May 2002 the Director served a further notice (“the second Rule 14 notice”) on Aberdeen Journals under Rule 14 of the Director’s Rules stating his intention to adopt a decision that Aberdeen Journals had infringed the Chapter II prohibition in the period from 1 March to 29 March 2000. The matters put to Aberdeen Journals in the second Rule 14 notice were substantially the same as the Director’s findings in the first decision, with the important exception that the Director substantially expanded his analysis of the issue of relevant product market. The Director also included certain additional information on costs to reflect matters which had come to light during the first appeal. 4

[17]Aberdeen Journals replied to the second Rule 14 notice in writing on 19 July 2002, and orally on 29 July 2002. Those submissions on behalf of Aberdeen Journals were limited to the new matters relating to the relevant product market raised by the second Rule 14 notice. In relation to all other matters Aberdeen Journals relied on its previous submissions already made in the context of the first decision.

[18]The Director adopted the decision now under appeal on 16 September 2002. Paragraphs 19 to 143 of the decision deal with the issue of relevant product and geographic market and, in effect, replace paragraphs 14 to 52 of the first decision. The remainder of the decision, as regards the issues of dominance, abuse, effect on trade and penalty, is virtually identical to the first decision, save for certain additional information as regards costs which emerged during the first appeal.

[19]In its present appeal lodged on 18 November 2002 Aberdeen Journals has submitted detailed argument on the issue of relevant product market. Save for one additional argument relating to whether Aberdeen Journals priced below average variable cost in March 2000 (notice of application, paragraph 6.2), on all other issues Aberdeen Journals relies on the arguments it presented to the Tribunal during the first appeal.

[20]On 11 December 2002 the Tribunal ordered that the Tribunal’s file in relation to the first appeal form part of the record for the purposes of the present appeal. In the result, and with the consent of the parties, the Tribunal deals in this judgment with the arguments of the parties in relation to(i) the issue of relevant product market and the issue relating to below- cost pricing raised by Aberdeen Journals in its notice of application dated 18 November 2002, as supplemented by the oral hearing on 29 January 2003; and

(ii) all other issues raised by Aberdeen Journals in its application dated 14 September 2001, as supplemented by the oral hearing on 11 December 2001. II THE CONTESTED DECISION 4

[21]In the decision, which runs to 99 pages with 11 annexes, the Director first defines the product and geographic market(s) which he considers to be relevant to the issue of whether Aberdeen Journals has a dominant position (paragraphs 19 to 143), and then concludes that Aberdeen Journals is dominant in the market(s) as so defined (paragraphs 144 to 149). At paragraphs 150 to 209 the Director explains the reasons for his conclusion that Aberdeen Journals abused 4 Paragraph references in this section are to paragraphs in the contested decision, unless otherwise stated. 5 that dominant position by engaging in predatory conduct, namely selling below average variable cost, from 1 March to 29 March 2000. Such conduct is found to affect trade within the United Kingdom (paragraphs 210 to 212). At paragraphs 213 to 226, the Director deals with the imposition of the penalty of £1,328,040. The Director’s findings on dominance — Relevant product market

[22]At paragraphs 19 to 132 of the decision the Director gives his reasons for his primary conclusion “that the relevant product market within which Aberdeen Journals’ conduct must be assessed comprised advertising space in local newspapers concentrated on the Aberdeen area, namely the Evening Express , Herald & Post and Independent ” (paragraph 131).

[23]After setting out his general approach to the issue of relevant product market, including the issue of whether certain newspapers are to be considered as “complements” rather than substitutes (paragraphs 19 to 30), the Director sets out an overview of newspaper advertising markets, including different types of newspaper advertisement and different types of newspaper (paragraphs 31 to 44). He then considers the marketplace in Aberdeen (paragraphs 45 to 48) and the characteristics of the four newspapers involved in this case, the Press & Journal , Evening Express , Herald & Post , and Independent (paragraphs 49 to 70). At paragraph 70 the Director concludes that, having regard to its particular characteristics, the Press & Journal does not compete in the same relevant market as the other three newspapers.

[24]The Director then concludes that the relevant product market cannot be segmented into sub- markets by advertisement type (e.g. recruitment, property, motors, displa y/retail): paragraphs 72 to 74. Moreover, according to the Director, neither the fact that the Evening Express is paid-for, while the Independent and Herald & Post are free, nor the fact the Evening Express is daily while the Independent and Herald & Post are weekly, is sufficient to show that the Evening Express is in a separate market from the other two titles as far as advertising space in local Aberdeen newspapers is concerned (paragraphs 75 to 85). Similarly the Director does not consider that other media – such as the internet, local radio, recruitment agencies, the solicitors’ Property Register, direct mail, leaflets, advertising-only publications or other forms of local media – form part of the relevant product market (paragraphs 86 to 92).

[25]At paragraphs 93 to 99 of the decision, the Director sets out his position as regards economic and econometric evidence in this case. He questions the relevance for market definition of 6 purchasers’ behaviour when prices are far removed from competitive levels (paragraphs 93 and 94) and notes in addition the difficulty of obtaining meaningful pricing data (paragraph 95). These difficulties notwithstanding, the OFT conducted an econometric analysis using average yield data for the newspapers concerned, but were unable to obtain any reliable statistically significant results (paragraph 96). The Director also rejects the contention that a report submitted by Aberdeen Journals prepared by RBB Economics, based on invoice data from Aberdeen Journals, establishes that the Evening Express is not in the same market as the Independent and the Herald & Post (paragraphs 98 and 99 and footnote 98). He concludes that “in this case, the economic evidence available does not, in itself, provide sufficiently strong and compelling evidence of the existence of a sufficient competitive constraint between the Evening Express and the Independent to prove that both newspapers were active on the same advertising market in March 2000” (paragraph 98).

[26]The Director considers, however, that in this case the conduct of Aberdeen Independent and Aberdeen Journals provides strong evidence that the Independent was competing with both the Herald & Post , and the Evening Express , until at least the end of March 2000 (paragraph 101). The Director relies first on the launch strategy of the Independent (paragraphs 102 to 106); Aberdeen Journals’ response to that launch (paragraphs 107 to 115); and the advertising sales methods used by the parties concerned (paragraphs 116 to 119).

[27]In relation to the launch strategy of the Independent , the Director considers that since the monthly advertising revenues of the Herald & Post in 1996 were less than the planned cost per edition of the Independent , the Independent’s launch “would clearly have been an uneconomic prospect if its aim was limited to capturing advertising solely from the Herald & Post . On the assumption that [Aberdeen Independent] was acting as a rational, profit-maximising economic operator, the launch of the Independent can therefore only be understood as a means of taking business from the Evening Express ” (paragraph 103). The Director also refers to a pre-launch investigation commissioned by Mr Barwell, entitled “The Scottish Opportunity – Aberdeen” which, according to the Director, found that(i) all advertisers using Aberdeen Journals’ titles were potential customers for the Independent ;

(ii) the Herald & Post was at that time a low-quality newspaper with no motor or property advertising; and

(iii) the Evening Express was vulnerable to entry by a high quality weekly free newspaper (paragraph 104). According to the Director, the decision of Mr Barwell, an experienced proprietor, to enter, and then remain, in the market for four years supports the conclusion that advertisers from the Evening Ex press would see the Independent as a viable alternative (paragraphs 105 to 106). 7

[28]In relation to Aberdeen Journals’ response to the launch of the Independent , the Director refers to comments by Aberdeen Journals in its internal documentation to the effect that the Herald & Post was “a tactical tool in the company’s armoury”, to be activated as necessary to protect the revenues of more lucrative titles and then run down again once the new entrant had been expelled (paragraphs 107 and 181). According to the Director, the reduction in advertising rates and increase in pagination of the Herald & Post immediately following the launch of the Independent “can only be viewed as rational as a means of protecting the Evening Express ’s considerable revenues, rather than the Herald & Post’s much lower revenues” (paragraph 108), a conclusion which is supported by internal memos dated 12 July 1996 and 29 July 1998 between Northcliffe and Aberdeen Journals (cited at paragraph 181 of the decision). The Director concludes that Aberdeen Journals’ response to the launch of the Independent can only be understood as an attempt to use the Herald & Post as a “fighting title ” to defend the revenue of the Evening Express against competition from the Independent . That, according to th e Director, is further compelling evidence that all three titles were present on the same product market from 1996 until at least the end of March 2000 (paragraph 109).

[29]The Director further relies on evidence from earlier investigations by the Director which found that Aberdeen Journals was offering discounts to advertisers in return for exclusivity. Some of the discount offers were in relation to the Evening Express , whilst others “bundled” discounts across the three Aberdeen Journals’ titles (paragraphs 110 to 113). According to the Director, such discounting practices provide further evidence that the Independent was viewed by customers as being “a viable alternative to advertising in the Evening Express ” (paragraph 114). The Director considers that even if, as Aberdeen Journals claimed, it was no longer offering such discounts by March 2000, these practices provide evidence of an overall market strategy that persisted into that month (paragraph 115).

[30]In relation to advertising sales methods used by the parties, the Director relies on evidence that Aberdeen Independent staff regularly used editions of the Evening Express to identify potential customers for the Independent who would then be called by telephone, including in March 2000. The Director also refers to a witness statement from a former Aberdeen Journals sales employee to the effect that the Evening Express sales staff targeted Independent advertisers in a similar manner. The Director considers this evidence further supports his conclusion that the Independent and Evening Express were in the same market (paragraphs 116 to 119). 8

[31]Finally, the Director also relies on documentary evidence to support his conclusion that the Independent competes not only with the Herald & Post but also with the Evening Express (paragraphs 120 to 130). According to the Director, a submission to the OFT of 10 February 2000 from Mr Scott, Managing Director of Aberdeen Journals , demonstrates that Aberdeen Journals considered the Independent to pose a significant threat to the Evening Express , and there is no evidence to suggest that the position had changed by March 2000 (paragraphs 121 to 123). According to the Director, that conclusion is further supported by an internal memo of Northcliffe dated 21 May 1999 indicating a link between the conduct of the Independent and the strategy for the Press & Journal and the Evening Express (paragraph 124) , and a meeting note of 5 August 1999 showing that Aberdeen Journals knew that Mr Barwell saw the Evening Express as vulnerable to the Independent (paragraph 125). The Director further relies on a Northcliffe document of January 2000 entitled “Review of Aberdeen Independent” which, according to the Director, shows that the Independent and the Evening Express were clearly competing (paragraphs 126 to 128). Finally, the Director refers to evidence submitted in an appendix to Aberdeen Journals’ letter to the OFT of 1 August 2000 as indicating that there was demand substitutability between the Herald & Post and the Evening Express (paragraph 129). On the basis of this documentary evidence, the Director finds that Aberdeen Journals’ strategy for meeting the challenge posed by the Independent was prompted by, and based on, the view of Northcliffe and Aberdeen Journals that the Evening Express , the Herald & Post and the Independent were direct competitors; and that Mr Barwell shared that view (paragraph 130).

[32]Paragraph 131 of the decision concludes: “that the relevant product market within which Aberdeen Journals’ conduct must be assessed comprised advertising space in local newspapers concentrated on the Aberdeen area, namely the Evening Express , Herald & Post and Independent . This conclusion is based on: · the fact that the three newspapers share certain characteristics, namely format, circulation area and general style, which are not shared with the Press & Journal , and, at least to some extent, similar advertising ratecard rates for display advertisements (except for the Herald & Post , whose rates were depressed by Aberdeen Journals’ predatory strategy); · the lack of viable alternative media for the majority of local advertisers; · contemporary evidence that the Independent was launched expressly to take business from the Evening Express ; · Aberdeen Journals’ reaction to the entry of the Independent , namely funding the Herald & Post in a manner that can only be rationalised as a means of defending the lucrative business of the Evening Express ; · evidence that Evening Express sales staff used copies of the Independent to target potential customers for their newspaper, and vice versa; and 9 · the contemporary documents of Aberdeen Journals and Northcliffe personnel, which refer to the Independent as a competitive threat to the Evening Express and evidence the companies’ use of the Herald & Post as a “defensive free” to protect that newspaper.” — The alternative product market definition

[33]If, contrary to his view, there are separate markets in Aberdeen for advertising in free local newspapers, and in paid-for local newspapers, respectively, the Director considers that the Evening Express would operate alone on the latter market (paragraphs 133 to 135). Even if this were the case, there are, according to the Director, strong “associative links” between the markets for, on the one hand, paid-for and, on th e other hand, free, newspapers in the Aberdeen area. These associative links are constituted by Aberdeen Journals’ strength in both sectors, the common customer base for advertising space, the common readership and the common production methods for the titles (paragraph 136). On that hypothesis, applying the decisions in

Case T-65/89

BPB Industries v Commission

[1995] ECR II-389

, at paragraphs 92 to 97 and

Case T-83/91

Tetra Pak v Commission

[1994] ECR II-755

at paragraphs 112 to 122, on appeal

Case C-333/94

[1996] ECR I-5951

at paragraphs 21 to 31 (“ Tetra Pak II ”), the Director considers that Aberdeen Journals’ conduct of pricing below average variable costs on the market for free local newspapers in Aberdeen constituted an abuse of its dominant position in the market for paid-for local newspapers in Aberdeen (paragraph 137). According to the Director: “… the loss making strategy of the Herald & Post can only be rationalised either as an attempt to prevent the Independent from attacking the Evening Express ’s revenues directly (on the basis that both titles were on the same market – as the Director argues) or as an attempt to eject the Independent from the free newspapers market before it could become a threat to Aberdeen Journals’ position on the separate but associated paid-for market, on which it enjoyed a monopoly” (paragraph 138). — The geographic market

[34]Finally in relation to market definition, the Director concludes that the relevant geographic market is Aberdeen or the circulation area of the Herald & Post (paragraphs 140 to 143). — Dominant position

[35]As regards the issue of whether Aberdeen Journals had a dominant position, the Director finds that Aberdeen Journals’ market share of advertising in local newspapers in the Aberdeen area in the period from January to March 2000 was 78 per cent by value and 67 per cent by volume (including all Evening Express sales). If the Evening Express sales are apportioned, 10 so as to include only sales made in the urban area of Aberdeen where the Herald & Post and the Independent are distributed, then Aberdeen Journals’ market share of advertising in local newspapers in the Aberdeen area in the period from January to March 2000 falls to 73 per cent by value and 63 per cent by volume. On the basis of those market shares, coupled with what the Director finds to be significant barriers to entry, the Director considers that Aberdeen Journals had a dominant position in the supply of advertising space in both paid-for and free local newspapers in Aberdeen or the circulation area of the Herald & Post (paragraphs 144 to 149). The Director’s findings on abuse

[36]As to whether Aberdeen Journals has abused its dominant position, the Director relies on the matters set out at paragraphs 150 to 212 of the Decision. He considers that the legal principles to be applied in considering a possible predatory abuse where a dominant undertaking deliberately incurs losses to expel rivals from the market or to deter new entry are those to be derived under Community law from

Case C-62/86

AKZO Chemie v Commission

[1991] ECR I-3359

(“ AKZO ”), and Tetra Pak II (cited above). Within that framework, the Director undertook a financial analysis of the contribution generated by the Herald & Post , taking into account the policy followed by the Herald & Post in relation to advertising rates, pagination and distribution (paragraphs 155 to 174); the relationship between the revenue of the Herald & Post and its costs (paragraphs 175 to 180); the intentions of Aberdeen Journals as disclosed in certain internal documents (paragraphs 181 to 183); and certain arguments advanced by Aberdeen Journals by way of defence (paragraphs 184 to 206). — Advertising rates, pagination and distribution of the Herald & Post

[37]According to the Director, the key features that Aberdeen Journals can vary in producing the Herald & Post are(i) advertising rates;

(ii) pagination (i.e. the number of pages in each edition); and

(iii) distribution (i.e. the number of copies distributed of each edition, also known as circulation): see paragraph 164. According to the Director, Aberdeen Journals took the following action on each of these aspects.

[38]The average advertising rate of the Herald & Post was cut shortly after the launch of the Independent from £3.03 per single column centimetre (“sccm”) in April 1996 to £1.43 per sccm in May 1996. The rate was further cut in October 1998 to below £1 per sccm. In March 2000, the average rate was £1.25 per sccm. That was raised to £1.52 per sccm in April 2000. 11 The average rates for May, June and July 2000 were £1.73, £1.66 and £1.52 per sccm respectively (see paragraph 165, and Annex 2, Graph 1).

[39]Pagination of the Herald & Post was increased from under 100 pages a month (i.e. no more than 20-25 pages a week) in the period up to April 1996, to 148 pages in May 1996. Subsequently pagination rose, until it was mainly between 350 and 480 pages per month in the period between September 1998 and October 1999 (i.e. weekly papers containing 84 to 96 pages). Between November 1999 and March 2000, pagination declined to between 300 and 380 pages per month. Pagination in March 2000 was 388 pages. Pagination was cut back in April 2000 to 220 pages per month, and remained at about that level until July 2000 (see paragraph 167 and Annex 2, Graph 2).

[40]From October 1995 to April 1998 the distribution of each weekly edition of the Herald & Post was between 80,000 and 100,000 copies. Distribution increased to above 120,000 per edition in November 1998, but between October 1999 and March 2000 distribution declined to between 100,000 and 108,000 copies. In March 2000, average distribution was 102,600 copies for each weekly edition. This was cut to 83,974 copies in April 2000, and remained at about that level until July 2000 (see paragraph 169, and Annex 2, Graph 3). — Costs and revenues of the Herald & Post

[41]Monthly figures for the costs and revenues of the Herald & Post were supplied to the Director by Aberdeen Journals for the period October 1995 to July 2000. The figures supplied were those reported in the management accounts of the Herald & Post , which set out the “directly attributable costs” and “directly attributable revenue” allocated to the Herald & Post by Aberdeen Journals. The principal costs so allocated were(i) the cost of newsprint (i.e. the cost of paper);

(ii) editorial costs;

(iii) costs relating to sales of advertising; and

(iv) distribution costs. The net balance between such “directly attributable costs” and “directly attributable revenue” constitutes the Herald & Post’s “contribution” (positive or negative) to Aberdeen Journals’ financial results, as shown in the management accounts (paragraphs 158 to 160).

[42]On this basis, paragraph 171 sets out the Herald & Post’s contribution to Aberdeen Journals’ financial results for the period October 1995 to July 2000. In summary, paragraph 171 shows that before the launch of the Independent, the Herald & Post’s contribution was positive. However, after the launch of the Independent in 1996, the Herald & Post incurred substantial “losses” (i.e. negative contribution) which continued throughout the period up to July 2000. 12 According to the management accounts of Aberdeen Journals, the “loss” incurred by the Herald & Post for March 2000 was £34,700. Further “losses” were recorded in the months of April, May, June and July 2000.

[43]In the contested decision, the Director considered that the “losses” of the Herald & Post shown in the management accounts were “very significant in the context of the revenue being earned” (paragraph 172). During the period from November 1998 to September 1999, the “losses” of the Herald & Post were over 75 per cent of its revenue. In March 2000 the “losses” of the Herald & Post represented 42 per cent of its revenue. The Director infers in the contested decision that “these were not losses incurred as a result of a miscalculation: they were too large and durable for this to be the case” (paragraph 173). He noted further that the “losses” of the Herald & Post since April 2000 have been less significant, but were still 12 per cent of revenue in July 2000 (see paragraphs 171 to 173, and Annex 2, Graphs 4 and 5).

[44]The Director further notes, at paragraphs 161 to 163 of the decision, that certain costs, including “printing and ink/plate costs”, are not allocated to the Herald & Post in the management accounts of Aberdeen Journals. During the course of the first appeal, further unallocated costs were identified, namely “electricity” (i.e. the cost of the power consumed by the presses printing the Herald & Post ), “transport – fuel” (i.e. the additional fuel consumed by Aberdeen Journals’ delivery drivers when delivering the Herald & Post ), overtime (i.e. the costs of printing staff working overtime in the months concerned) and “production – pre-press” (i.e. the cost of staff involved in setting up the newspaper’s pages before printing). According to the Director, adding these costs to the “loss” of the Herald & Post in the month of March 2000 increases the negative contribution of the Herald & Post from £34,700 to £48,038 during that month. Similarly, in the period April to July 2000 the negative contribution of the Herald & Post is significantly increased when those costs are included (see the note to the table following paragraph 171, and the table following paragraph 174 of the decision).

[45]Turning to the question of the fixed and variable costs of the Herald & Post , the Director cross-refers to the definitions of “fixed” and variable costs set out in OFT Guideline 414, namely that “fixed costs” are those that do not vary with output, whereas “variable costs” are those that do vary with output. For the purposes of this case, the Director has assessed “variable” costs as those that could be varied within the period of a single month, on the basis notably that one month is the period “over which short term planning for the Herald & Post 13 might be determined”. The Director, however, avers that such a short period “errs against a finding of predation” (paragraph 175).

[46]On that basis, the Director accepts that, over one month, there was a fixed element to the costs of the editorial staff and the advertising team of the Herald & Post , but he considers that the costs of newsprint and distribution are variable (paragraph 176). According to the Director, in the 45 months between July 1996 and March 2000, the revenue of the Herald & Post exceeded the costs of newsprint and distribution on only five occasions, most recently in April 1998. In March 2000, the revenue of the Herald & Post was £82,397, which did not cover the newsprint and distribution costs of £89,736. However, in April, May, June and July 2000 the costs of newsprint and distribution were covered by revenue (paragraphs 176 and 177 and Annex 2, Graph 6).

[47]In the contested decision, at paragraphs 178 to 180, the Director also carried out a further analysis of costs for the months of March to July 2000, in order to determine the variable costs of the Herald & Post on the basis of what costs could be avoided if the Herald & Post were not produced for a month. On the basis of this further analysis, the Director considered that Aberdeen Journals had failed to price above average variable costs regarding the Herald & Post in March, May and June 2000. Specifically as regards the month of March 2000, the Director found that the revenue of the Herald & Post was £82,397 and that its variable costs, including an allocation for common costs and incorporating the further information on costs provided during the first appeal, were £100,067, giving a negative contribution of £17,670 for that month. In April 2000, there was a surplus above variable costs so calculated of £3,461. In May 2000, there was a negative contribution of £3,781, in June 2000 there was a negative contribution of £3,647, and in July 2000 there was a surplus of £6,364 (see paragraph 179). — Aberdeen Journals’ intentions

[48]At paragraphs 181 to 183 of the decision, the Director found that it was Aberdeen Journals’ intention to use the Herald & Post strategically to expel the Independent from the market by means of incurring substantial losses on the Herald & Post . In support of that conclusion, the Director set out, in the table following paragraph 181 of the decision, extracts from a number of internal memoranda between staff of Aberdeen Journals and Northcliffe supplied to the Director by Aberdeen Journals. The citations in the contested decision are as follows:

‘Memo dated 12 July 1996, Mr Alec Davidson (Managing Director of Northcliffe Newspapers Group Ltd) to Mr Alan Scott (Managing Director of Aberdeen Journals): 14 Under the heading ‘ Herald & Post ’: ‘You view the Herald & Post as a tactical tool in the company’s armoury. Barwell’s [Keith Barwell owns the Independent ] move to Aberdeen has caused you to increase your efforts on this and if and when he goes away you will leave a three month gap between that happy event and running it down again.’ ‘Next year’s figure [the annual budget] would include the £500,000 investment we are making against Barwell. Whilst you thought it possible that Barwell would cease publication by Christmas this cannot be built into the budget.’

Memo dated 1 April 1997, Mr Davidson to Mr Ian Lovett (Commercial Systems Manager at Aberdeen Journals): After references to whether the Independent is making a profit, or loss, ‘Finally, please keep your foot on their neck!’ Memo dated 12 May 1998, Mr Davidson to Mr Scott:

‘You perceive the Independent to be less of a threat to you and therefore propose to fight it with the Herald & Post , not the paid-for titles. We authorise an additional £50,000 to be invested into that and this will be taken into account when calculating your strive payments at the end of the year. I would be tempting fate if I recorded that you think the Independent may cease publishing by the end of this financial year but here goes anyway!’ ‘You also proposed to place greater separate focus on the Herald & Post so that it is our only title pitched against the Independent . Again, this is agreed for this could be an area where we could make substantial profit progress over the next 18 months to 2 years, given that we are successful in closing them down.’

Memo dated 29 July 1998, Mr Michael Pelosi (Deputy Managing Director at Northcliffe) to Mr Scott:

‘the closure of the Aberdeen Independent would allow you to reduce gradually investment in the Herald & Post , resulting in additional profits of between £0.5 m and £1m.’

Memo dated 6 December 1999 from Mr Davidson to Mr Scott:

‘You agree to produce 2 scenarios as far as the Independent is concerned. The first assumes that we acquire them. The second assumes that you are given a sum of money to neutralise them.’

Memo dated 5 January 2000 from Mr Davidson to Mr Taymour Ezzat (then Northcliffe’s London Financial Controller):

‘The purpose of your visit is to help Aberdeen construct three operational and financial scenarios relating to the Aberdeen Independent . These can be summarised as: 2.1 To continue with the existing policy 2.2 To purchase the Aberdeen Independent ; and 2.3 To considerably enhance our existing activity with a view to denying the Independent all commercial oxygen.’

Review of Aberdeen Independent by Mr Ezzat (undated, but responsive to Memo dated 5 January 2000 from Mr Davidson to Mr Ezzat): After reviewing competition between Aberdeen Journals and Independent : ‘The current position is one of stalemate’ (page 2). 15 Recommendation (page 5) ‘to purchase the Independent from Barwell and merge the title with our own free title ... Unfortunately Barwell is currently on a roll and may feel that he can demand a higher price. NNG will therefore need to move forward by developing the Herald & Post and increasing the pressure on Barwell. ... Our response to Barwell was very vigorous and most publishing entrepreneurs would not have been able to fund these losses over four years. Assuming the OFT risk is minimal, open negotiations with Barwell as we need to bring his price expectations down. On the basis that he will not accept our views, we need to continue with the development of the Herald & Post at the same time. NNG have to be prepared to maintain this approach for a sustained period (6 to 12 months) in order to convince Barwell that we will not allow the Independent to break even. I believe that maintaining the pressure on Barwell by attacking the Independent more aggressively and satisfactorily resolving the OFT queries will eventually ensure he will accept our offer.’”

[49]On the above evidence, the Director presumed what he describes as “predation” by Aberdeen Journals in March, May and June 2000, contrary to the Chapter II prohibition. The Herald & Post ’s prices were below average variable cost during that period, even taking into account only costs which were regarded as variable over a reference period as short as one month (paragraph 184). — Aberdeen Journals’ counter-arguments

[50]According to paragraphs 185 to 205 of the decision, Aberdeen Journals raised three counter arguments to rebut the Director’s presumption of predatory conduct, namely that:(i) Aberdeen Journals was merely meeting competition;

(ii) by 1 March 2000 Aberdeen Journals had changed its strategy, so no predation could be established after that date; and

(iii) in any event, there was no predation in May and June 2000 by Aberdeen Journals since extra costs were incurred in those months because the threat of industrial action in Aberdeen compelled it to print the Herald & Post in Leicester.

[51]As to those arguments, the Director considered(i) that Aberdeen Journals did not react proportionately to the Independent’s entry, but rather initiated and maintained a strategy designed to expel the Independent from the relevant market, using the Herald & Post as a “fighting title” (paragraphs 186 to 189 of the contested decision);

(ii) that there was no significant change in the situation after 1 March 2000 so as to rebut the presumption of predation during the month of March (paragraphs 190 to 200), since Aberdeen Journals did not take any decisive action to reduce its variable costs until the end of March 2000 16 (paragraph 199);

(iii) Aberdeen Journals must have know there was a serious risk of pricing below average variable cost in March 2000 (paragraph 201) and had sufficient time to moderate its exclusionary policy so as to comply with the Chapter II prohibition from 1 March 2000 (paragraphs 202 to 204); but

(iv) that the cost increase in May and June 2000, caused by printing the Herald &Post in Leicester as a result of the threat of industrial action in Aberdeen, amounted to an objective justification displacing the presumption of predation in those two months (paragraph 205). On that basis, there remained only one month in which predation was found, namely March 2000. The Director’s conclusion on infringement

[52]On the basis of the evidence before him, the Director found, at paragraph 207 of the decision, that there was “a clear, unbroken chain of conduct” linking Aberdeen Journals’ predatory conduct before 1 March 2000 with its conduct once the Chapter II prohibition had entered into force on that date. There was no objective justification for that conduct. In particular, there was insufficient evidence to rebut the presumption that Aberdeen Journals had engaged in predatory conduct by its failure to cover its average variable costs from 1 to 29 March 2000 (paragraphs 207 to 209).

[53]The Director further considered that trade within the United Kingdom was likely to be appreciably affected by Aberdeen Journals’ conduct (paragraphs 210 to 212). Accordingly, the Director concluded that Aberdeen Journals had infringed the Chapter II prohibition (paragraph 212). The penalty

[54]The Director further held that the infringement had been committed intentionally or negligently (paragraphs 217 and 218) and imposed a penalty of £1,328,040, pursuant to his powers under section 36(2) of the 1998 Act (paragraphs 225 and 227). The Director’s calculations of the penalty are set out at paragraphs 219 to 226 of the contested decision. At paragraph 214 of the contested decision the Director found that no directions were necessary under section 33(1) of the 1998 Act “[s]ince Aberdeen Journals has provided evidence to the Director that it has ceased predating.” 17 III ARGUMENTS OF THE PARTIES ON RELEVANT MARKET AND DOMINANCE Introduction

[55]Aberdeen Journals contends, principally, that it is not dominant in any relevant product market as alleged by the Director. Aberdeen Journals’ principal submissions are(i) although the free weekly Herald & Post and Independent are in direct competition with each other, the Director has failed to prove that the paid-for daily Evening Express is in the same market as those two free weekly newspapers;

(ii) in consequence, the Director has failed to prove the relevant product market on which he relies, namely the market for advertising in local newspapers (paid-for and free) in Aberdeen;

(iii) since the Independent outsells the Herald & Post in the market for free weekly newspapers in Aberdeen, there is no question of Aberdeen Journals’ dominance in that market;

(iv) the Director’s alternative product market, namely the market for paid-for local newspapers in Aberdeen, in which the Evening Express is the only supplier, is not open to the Director to advance and is , in any event, insufficiently proved;

(v) in defining the relevant product market the Director has failed to take account of competition from other media; and

(vi) in any event, whatever the relevant product market, the Director has insufficiently analysed the question of whether Aberdeen Journals is dominant in the market or markets alleged.

[56]We note that, according to the decision (paragraph 144), in January to March 2000 Aberdeen Journals had a market share of all advertising in local newspapers in the Aberdeen area of 78 per cent by value and 67 per cent by volume, or slightly less if the Evening Express ’s sales outside the urban area of Aberdeen are disregarded. From those market shares and other factors, says the Director, dominance can be established. By contrast, Aberdeen Journals’ share of advertising in free local newspapers in Aberdeen was only 35 per cent by value (46 per cent by volume) during the same period i.e. the Herald & Post was being outsold by the Independent . It is therefore critical to the Director’s primary case on dominance that the Evening Express is properly included, together with the Independent and Herald & Post , in the same relevant product market, i.e. the market for advertising in local newspapers (paid-for and free) in the Aberdeen area. If the Director’s primary case is not established, his alternative argument is that the Evening Express is dominant in the market for advertising in local paid-for newspapers in the Aberdeen area and that Aberdeen Journals has abused its dominant position in that market.

[57]The arguments summarised in this section are principally addressed to these issues. The arguments on a subsidiary issue, namely whether local advertising media other than 18 newspapers form part of the relevant product market, are also briefly dealt with. In this case there is no material dispute about the relevant geographic market, namely Aberdeen or the circulation area of the Herald & Post . Aberdeen Journals’ submissions The Director’s primary case is not proved

[58]Aberdeen Journals’ main submission is that the Director has failed to adduce sufficient strong and compelling evidence to show that in March 2000 the Evening Express was competing in the same market as the Independent and the Herald & Post . In particular, the Director has failed to produce any, or any sufficient, evidence, notably from the demand side, that advertisers regarded advertisements in the Independent or Herald & Post as substitutes, rather than complements, for advertisements in the Evening Express , or that in March 2000 the advertising business of the Independent was a sufficient constraint on the advertising business of the Evening Express to be considered as part of the same market for local newspaper advertising.

[59]Aberdeen Journals argues, in particular, that the Director wrongly failed to carry out a proper survey of advertisers, for example of the kind done by the Competition Commission in Johnston Press plc and Trinity Mirror plc , Cm 5495, May 2002 (“ Johnston Press/Trinity Mirror ”); ignored the results of a survey he did carry out which tended to support the applicant’s case; found his own econometric analysis to be inconclusive; wrongly placed undue reliance on the views of Aberdeen Independent, an interested third party; and placed undue weight on subjective evidence relating to conduct and statements by the parties, instead of relying on objective, verifiable economic evidence.

[60]Moreover, says Aberdeen Journals, the Director has wrongly ignored the objective economic evidence prepared on behalf of Aberdeen Journals by RBB Economics in their first report of July 2002 (“the first RBB report”), which is now corroborated by a second report prepared by RBB Economics in November 2002 (“the second RBB report”). The second RBB report is in turn confirmed by the evidence of Professor Yamey before the Tribunal contained in a commentary dated 8 November 2002.

[61]According to Aberdeen Journals, the Tribunal should recognize a hierarchy of evidence on issues such as product market definition. The most directly relevant evidence is statistical analysis of the extent of any competitive relationship, based on primary raw material. 19 Secondly, there is survey evidence which, although subjective, can be given probative value through appropriate methodology. Thirdly, there is subjective evidence which may emanate from the alleged infringer, the complainant or another source, from documents or observed conduct. However, unless given in direct evidence before the Tribunal, such evidence constitutes multiple hearsay and must be treated with considerable circumspection: a document may not represent the current opinion of its author, it may have been corrected subsequently, or it may be based on errors. The motivation of the author is also relevant: for example, unverified evidence from Aberdeen Independent should be given very little credibility, given that it has a clear interest in the outcome of the case. Further, internal documents reflect the supply side, whereas it is the demand side which is the litmus test for product market definition. According to Aberdeen Journals, the Director’s approach in the present case, which is to rely almost entirely on past conduct and statements by the supplying parties, as distinct from evidence of the attitudes of consumers and users, is without precedent in EC law.

[62]Turning to the specific matters relied on in the decision, Aberdeen Journals considers that there is little of evidential weight in paragraphs 19 to 30 (introduction to product market definition), or paragraphs 31 to 44 (overview of newspaper markets) where the Director himself draws no conclusions, or paragraphs 45 to 48 (the Aberdeen marketplace). The discussion of the characteristics of the newspapers in question in paragraphs 49 to 70 of the decision produces no clear result. The Director’s review of copies of the newspapers for March 2000 (only two weeks of that month for the Evening Express ) provides no evidence of substitutability, but rather the reverse, since it demonstrates considerable disparity in average advertising yields between the Press & Journal , the Evening Express , and the free newspapers, respectively. The Director produces no evidence about differences in the readership base or “quality” of the newspapers in question.

[63]Similarly, the Director’s discussion, in the decision, of “paid versus free” and “weekly versus daily” (paragraphs 75 to 85) yields nothing of evidential weight. No mention is made of the Competition Commission report in Johnston Press/Trinity Mirror , cited above, which tends to show that paid-for dailies and weekly free titles are complements, not substitutes. Again the views of advertisers on the importance of such matters as cover price and periodicity have not been sought. Mr Alan Scott’s witness statement of 1 February 2001 indicates that those differences between the newspapers in question are crucial. 20

[64]Similarly the Director has taken no account of Mr Scott’s evidence to the effect that readers spend under half the time reading a free paper as they spend reading a paid-for paper; advertisements in free papers generate substantially poorer responses than advertisements in paid-for papers; 27 per cent of the Evening Express readership is outside the Aberdeen area; the Evening Express had a target readership aged 15-44, while the Herald & Post had a target readership of 44+; the Herald & Post had a higher A/B profile than the Evening Express ; and that advertisers used paid-for papers in preference to free papers because of their higher quality.

[65]As to the evidence of conduct relied on by the Director (paragraphs 100 to 119 of the decision), Aberdeen Journals submits, notably, that the launch strategy of the Independent “however misguided or ill-advised” is not itself evidence of market definition. Mr Barwell’s views of what was a profitable launch strategy do not establish how advertisers viewed the matter, e.g. Mr Barwell considered that his targets included the Press & Journal , excluded by the Director from the relevant market. Mr Barwell’s views are also contradicted by Mr Robin’s letter of 8 March 2000 and an OFT meeting note of 25 April 2002. Moreover, the evidence as to Aberdeen Journals’ response to the launch of the Independent (e.g. memoranda dated 12 July 1996 an d 29 July 1998, and correspondence with advertisers relating to May 1996, November 1998 and January 1999) is not relevant to the position in March 2000. The call logs of the Independent in March 2000 relied on by the Director only show that the Evening Express was a potential source of business; Mr Barwell’s evidence is that this type of activity is no more than routine. Mr Farquharson, whose evidence is also relied on, is a disgruntled employee who left Aberdeen Journals in 1998. No reliance should be placed on his evidence.

[66]As regards the documentary evidence relied on by the Director (paragraphs 120 to 130 of the decision) Aberdeen Journals’ position is that this evidence is inconclusive and does not indicate what the market position was in March 2000. According to Aberdeen Journals, evidence about competition between titles at point X in time is not relevant to the analysis at point Y in time, if the titles’ characteristics have changed in the meantime. Businessmen’s perceptions of what a particular strategy or market situation may have been at a particular point in time cannot be safely relied on to determine issues such as relevant product market.

[67]Aberdeen Journals makes detailed comments on each of the documents relied on by the Director, namely th e letter of 10 February 2000, Appendix 2 to that letter, Northcliffe’s memorandum of 21 May 1999, the meeting note of 5 August 1999, and the Review of 21 Aberdeen Independent prepared by Mr Ezzat dated 18 January 2000. Aberdeen Journals points out that the Director has failed to mention Northcliffe’s memorandum to Aberdeen Journals of 20 July 1999, which led to an increase in the cover price of the Evening Express . That is not the reaction of a newspaper that is competing with the Independent . Aberdeen Journals also relies heavily on the reduction in the Independent’s distribution from 125,000 copies a week to 107,000 copies a week in February 2000 as showing that the Independent was not targeting the Evening Express in March 2000. Mr Ezzat’s memorandum properly analysed shows that any impact by the Independent on the Evening Express was de minimis, as confirmed by the fact that average yields of the Evening Express have been unaffected by the entry of the Independent : see the first RBB report, figure 1 at page 9, and Mr Scott’s witness statement.

[68]More fundamentally, argues Aberdeen Journals, little reliance can be placed on the actions of Aberdeen Independent in launching the Independent , or on Aberdeen Journals’ response to that launch, because the Independent was an “inefficient market entrant” which entered the market at prices which were unsustainably low and with a distribution level and editorial quality which were unsustainably high. Since the Independent was, says Aberdeen Journals, competing on the basis of a product which was uncompetitive over any sustainable period of time, it cannot be argued that it was competing with the Evening Express in any normal way. The reaction of Aberdeen Journals to what was essentially an abnormal situation is therefore no guide to whether the Independent and the Evening Express could be said to be in the same market in normal competitive conditions, as Professor Yamey points out. In support of its contentions, Aberdeen Journals relies on Appendix 2 to the letter of 10 February 2000, notably where Aberdeen Journals expresses the view that “the objective of the Independent was to create a business that, over a relatively short period of time, would create a lot of damage to AJL’s business rather than build a long term viable newspaper” and that “it is very difficult to understand how the Independent can build a viable long-term business with this cost base”. Aberdeen Journals also alleges that the Independent could target the Evening Express only by making misleading claims as to the circulation of the Evening Express and misleading comparisons between the circulation of the Independent , Evening Express and Press & Journal , as the Audit Bureau of Circulation and the Advertising Standards Authority found on several occasions.

[69]In fact, argues Aberdeen Journals, Mr Barwell’s objective was not to enter the market in any normal way but to create a “fireship” calculated to wreak so much havoc that Aberdeen Journals would be forced to buy Mr Barwell out at a premium. At the hearing on 29 January 22 2003, Aberdeen Journals relie d on a note of meeting at the OFT on 25 April 2002 where Mr Barwell is stated to have said that he had not been approached by Aberdeen Journals offering to buy the Independent , which statement is allegedly contradicted by a confidentiality agreement with Northcliffe entered into by Mr Barwell on 19 April 2002. This contradiction, submits Aberdeen Journals, undermines the credibility of Mr Barwell’s evidence. Aberdeen Journals also refers to a note of a meeting on 5 August 1999, internal Northcliffe memos of 6 December 1999 and 5 January 2000, and Mr Ezzat’s review of January 2000, all of which refer to the possibility of a purchase of the Independent by Northcliffe. Since the documents before the Tribunal, taken as a whole, indicate that the motivation of Aberdeen Independent was all along to act artificially to force a sale , the evidence of conduct invoked by the Director cannot be relied on to show a normal competitive relationship between the Independent and the Evening Express . The Director’s lack of economic evidence

[70]According to Aberdeen Journals, the Director’s reasons for not relying on economic or econometric evidence (paragraphs 93 to 99 of the decision) are wholly insufficient. According to Aberdeen Journals, the Director’s survey of advertisers in March 2000, although inadequate, tends to suggest that advertisers saw the Evening Express and the free weeklies as complementary, not substitutable, products. Moreover, the Director’s survey of advertising content in newspapers for the purposes of the second Rule 14 notice, briefly summarised at Annex 7 to the decision, does not show whether there is any substitutability between the newspapers in question. Professor Yamey confirms that the RBB analysis throws serious doubt on the Director's product market definition.

[71]As regards the Director’s internal statistical analysis, (see paragraph 96 of the decision) Aberdeen Journals considers that th e Director could have obtained actual invoice data for March 2000 from Aberdeen Independent by asking for it , especially since that company should have kept records for VAT purposes. More importantly, says Aberdeen Journals, if prices were below the competitive level in 1996 to 2000, the free weeklies would have been at their most attractive vis-à-vis the Evening Express at that time, so an observed lack of switching over that period would be a very firm and useful conclusion. The Director’s letter of 8 August 2002, which disclosed details of the Director’s methodology, but not his detailed workings, is inadequate. The fact that, according to the Director’s letter of 8 August 2002, his analysis showed that volumes in the Independent fell if the price of the Evening Express increased, is entirely consistent with advertising in the two titles being purchased as complements. In fact, the Director’s analysis could well constitute exculpatory evidence 23 indicating a lack of any competitive relationship between the Independent and the Evening Express . The Tribunal should draw negative inferences from the fact that the Director has not disclosed his workings to the Tribunal in support of his contention that his statistical results were unreliable. The RBB reports

[72]By contrast, Aberdeen Journals relies on the first RBB report, the second RBB report, and Professor Yamey’s commentary. The RBB reports are stated to be based on an analysis of the behaviour of over 90 per cent of the advertisers who advertised in the Evening Express in the period September 1999 to June 2000, relying on Aberdeen Journals’ invoice data.

[73]The first RBB report concludes that there were significant differences – about 2.5 times – in the advertising revenue yields obtained by the Evening Express and the free weekly newspapers. Furthermore, there appeared to be no correlation between changes in yields for the Evening Express as compared with the free weekly newspapers. These factors, argues RBB Economics, show that the Evening Express operates in a separate market to the free weekly titles.

[74]In addition, in its first report RBB Economics studied the advertisements placed by advertisers in March 2000. Of those advertisers that advertised in the Evening Express in that month, according to RBB Economics 61.2 per cent of the total advertising volume in the Evening Express was placed by advertisers who did not advertise in a free newspaper during that period. RBB Economics also studied the behaviour of the ten largest advertisers who advertised in the Evening Express and at least one free weekly newspaper between September 1999 and June 2000. In six out of the ten cases studied, the pattern of advertising as between the Evening Express and the free titles was not judged by RBB to have changed in response to changes in the rates or distribution of the newspapers in question during the period. RBB Economics concludes that these six advertisers, representing some 23 per cent of the total volume of advertising in the Evening Express in March 2000, treated the free newspapers as complements to, rather than substitutes for, the Evening Express . In relation to the other four principal advertisers, representing 15.6 per cent of the volume of advertising in the Evening Express in March 2000, RBB Economics was unable to reach any firm conclusions. From these results the first RBB report concludes that at least 84 per cent by volume of advertisers in the Evening Express in March 2000 did not regard the free titles as substitutes, but only as complements to the Evening Express . 24

[75]In the second RBB report, RBB Economic s sought to take account of certain points made by the Director in footnote 98 of the decision, notably by expanding the sample size. Using essentially the same reasoning, in its second report RBB Economics came to the conclusions that: ¾ Advertisers accounting for 46.4 per cent of advertising volume placed in the Evening Express in March 2000 did not use a free weekly newspaper in the period between September 1999 and June 2000. ¾ A study of 89 advertisers who advertised in the Evening Express in March 2000 and in at least one free title during the period between September 1999 and June 2000 showed that the behaviour of 45 advertisers (representing 34.7 per cent by volume of advertising in the Evening Express in March 2000) was not consistent with them viewing the free weekly titles as effective alternatives to the Evening Express . This conclusion is based, essentially, on the fact that changes in rates or distribution of the newspapers concerned were not judged by RBB to have caused any of these advertisers to switch their advertising. ¾ Of the remaining 44 advertisers (representing 11.7 per cent by volume of advertising in the Evening Express in March 2000) the results were not determinative of either substitutability or non-substitutability although, accordin g to RBB Economics, a case of non-substitutability could be justified. ¾ In the result, advertisers representing a total of some 81 per cent of advertising volume placed in the Evening Express in March 2000 did not regard the free newspapers as an alternative to the Evening Express .

[76]Aberdeen Journals refutes the criticisms made by the Director of the RBB reports by pointing out that advertisers representing 70 per cent of the volume of the Evening Express have been included. The lack of switching is confirmed by the evidence of Mr Scott, who points out that advertising rates for the Evening Express are not set by reference to advertising rates in the free newspapers, and that he did not detect switching from the paid-for titles when the rates for the Herald & Post were low. The increase in rates for the Herald & Post since March 2000 has not led to any switching to the paid-for titles. A ten-month sampling period is adequate, since many advertisers operate on a short time frame. The assertion that advertisers are not sensitive to distribution changes is unsupported and contrary to other aspects of the Director’s case. There is no evidence of ‘bundling’ at the material time, no evidence to support the Director’s assertion that the results may be affected by a “time lag”, and the size of the RBB sample is perfectly adequate. The fact that the RBB reports show a 25 lack of switching as between the free titles is irrelevant since the work was measuring switching as between the paid-for titles and the free titles. It is apparent that considerable switching in fact took place between the free titles, as the increase in the Independent’s market share shows.

[77]In a short further response dated 24 January 2003, RBB Economics responded to criticisms by the Director and Aberdeen Independent about the reliability of the database used, namely an internal database compiled by Aberdeen Journals’ staff of advertising in the Independent . Following a further audit, RBB Economics has come to the conclusion that 41.9 per cent by volume of advertisers advertising in the Evening Express in March 2000 did not advertise in a free title in the ten months between September 1999 and June 2000. Professor Yamey’s evidence

[78]Professor Yamey criticises the Director’s approach to statistical analysis, and questions why average yield data rather than invoice level data, was used. As regards the Director’s letter of 8 August 2002, Professor Yamey states that the material in this letter does not allow any assessment of the methods used or the appropriateness of the results arrived at, nor can he be satisfied that the exercise was carried out in a thorough, robust or impartial manner. In so far as the OFT’s results apparently show in various ways negative cross-price elasticities, described by the OFT as “anomalies” or “counter intuitive”, Professor Yamey considers that such results could support the view that the newspapers in question are not substitutes. In Professor Yamey’s view it would not, inherently, be contrary to nature or to economic theory for there to be a relationship, other than one of substitutability, between the Evening Express and the Independent ; alternatively, the demands for the two products could be unrelated.

[79]Professor Yamey also points out that the OFT’s work was based on actual yields and not on prices at normal competitive levels, which are the prices relevant to an analysis of relevant product market. Two products that are effective substitutes at one pair of prices need not be effective substitutes at a different pair of prices. It is reasonable to suppose that advertisers who might regard advertising in free newspapers as an effective substitute for advertising in a paid-for newspaper when the former is available at a below-competitive price, would not do so when it was available only at the (higher) competitive price. Professor Yamey suggests that Aberdeen Journals could have been legitimately apprehensive about the effect of the launch of the Independent on the resources of the Evening Express if the former were priced well below the competitive level. Apprehensions to this effect expressed by Aberdeen Journals are not evidence that the two newspapers are in the same product market (i.e. that 26 they are or would be effective substitutes if they were both priced at their respective competitive levels) if, in fact, the two newspapers were not priced at their respective competitive levels.

[80]As regards the RBB reports, Professor Yamey considers that this evidence throws serious doubts on the hypothesis that since 1996 the Independent exerted any material competitive control on the advertising business of the Evening Express . In particular, there is no clear evidence of switching by advertisers between the free newspapers and the Evening Express induced by changes in the relative price for advertising. This is equally true, says Professor Yamey, of changes in “the mix” , when an advertiser changes its newspaper mix in response to changes of relative prices. There is no evidence that this kind of change in “the mix” is occurring. Professor Yamey concludes that the evidence analysed by RBB Economics does not support the conclusion that there was a sufficient degree of substitutability to support a finding that the Evening Express and the free newspapers were competing in the same market, even when the free newspapers were being priced below their competitive level. The impact of other media

[81]In relation to other media, Aberdeen Journals considers that the Director has not supported his conclusions at paragraphs 86 to 92 of the decision with any evidence from advertisers in Aberdeen, or any information on relative costs. His reasoning is confused on the issue of complements versus substitutes. The Director has overlooked material in his file relating to the internet, local radio and directories.

(C) (iii)):

“for these media to exert a competitive constraint on newspaper prices, it is not necessary for all advertising to be able to switch between the two: it is only necessary for some advertisers at the margin to be able to switch, should a change in price occur.”

However, says Aberdeen Journals, the Director has not carried out this analysis. Moreover, Professor Yamey observes that any constraining factor on the Evening Express might not have been the Independent , but the availability of alternative media. Here the relevant question would have been whether alternative media, in combination, amount to a sufficient competitive constraint rather than whether any one alternative medium amounted to such a constraint, taken alone. 27 Dominance in the Director’s primary market

[82]On the issue of dominance in the primary market alleged by the Director, Aberdeen Journals argues (application in the first appeal, paragraphs 5.39 to 5.45), that the decision does not analyse whether the criteria set out in Case 27/76 United Brands v Commission [1978] ECR 207, at paragraph 128, are met in this case. In fact, from October 1999 onwards the Herald & Post was not behaving independently, to an appreciable extent, of its advertisers, but was renegotiating contracts and increasing its advertising rates, following the intervention of the Director. Mr Scott’s witness statement identified the loss of many major advertisers, notably in property and motors, as a result. Similarly a number of advertisers whom Aberdeen Independent alleged, in a letter of 2 June 1999, were being “picked off” by the Herald & Post , had either ceased advertising in the Herald & Post before March 2000, or had reduced their advertising spend. In these circumstances, argues Aberdeen Journals, it was incorrect for the Director to presume dominance on the basis of market shares alone. In effect, there were ‘exceptional circumstances’ present in March 2000 which required the Director to determine whether the Herald & Post did in fact enjoy a position of economic strength within the meaning of the case law of the Court of Justice, but the Director failed to do so. The alternative product market

[83]Aberdeen Journals contends that the Director’s alternative market definition (paragraphs 133 to 139 of the decision) is inadequately argued and unsupported. According to Aberdeen Journals, he has not shown that there are two separate product markets on the basis of strong and compelling evidence. In particular, he has not considered the extent of competition from the Press & Journal or alternative media; he has not shown that there are strong associative links between the two product markets (indeed, RBB’s evidence shows that there is no such competitive linkage between the two markets); and he has failed to link the alternative market hypothesis to his findings on abuse and penalties.

[84]In any event, Aberdeen Journals submits that the Director’s argument on the alternative product market is inadmissible. In the first appeal, the Director did not seek to support his alternative product market hypothesis because, he said, it had not been properly put to Aberdeen Journals in the course of the administrative proceedings. If the alternative analysis was abandoned in the first appeal, then it is an abuse of process to try to resurrect it now. If, on the other hand, the alternative analysis was not in the first decision at all, which Aberdeen Journals considers to be the case, then it is an abuse of process to raise in subsequent litigation a serious point which should have been taken at an earlier stage: see Johnson v 28 Gore Wood

[2001] 2 WLR 72

at page 90 A-F. Moreover, if the alternative hypothesis was never in the first decision, then the Tribunal cannot have ruled on it in the first judgment. It cannot therefore have formed part of “the matter” which was remitted to the Director. The Director’s submissions

[85]The Director relies on all the facts and matters set out in the decision (see section II above) and emphasises the following points. The Director’s primary case

[86]The Director accepts that he is required to prove his case on the basis of strong and compelling evidence but argues that in relation to a complex question such as market definition, the proper approach is to look at all the evidence in the round. In this case, the Director considers that he has followed the legal principles set out in paragraphs [86] to [104] of the first judgment, and in particular paragraphs [96] to [97] and paragraphs [102] to [104]. Contrary to Aberdeen Journals’ submissions, there is no burden on the Director to prove that recourse to any particular category of evidence was inappropriate in this case, or that he could not have obtained better quality evidence: it is sufficient for him to prove his case on the basis of strong and compelling evidence of whatever nature.

[87]According to the Director, Aberdeen Journals is wrong to dismiss paragraphs 31 to 92 of the decision as containing nothing of any evidential weight. Those paragraphs indicate a careful and detailed approach by the Director and form part of his conclusions. Specifically, the Director’s approach to past Competition Commission reports is consistent with that set out by the Tribunal in the first judgment. Aberdeen Journals has never suggested that the relevant product market should include alternative media , and its suggestion that the Director should have referred to a survey which Aberdeen Journals itself has claimed to prove nothing is contradictory.

[88]In defending his reliance in the decision (at paragraphs 100 to 130) on conduct and documentary evidence, the Director points out that this type of evidence was held in the first judgment (paragraphs [103] and [104]) as being likely to be particularly significant and may, depending on the particular circumstances, be of potentially decisive importance. The Director rejects Aberdeen Journals’ characterisation of this type of evidence as “tertiary”, or as multiple hearsay. It is direct and contemporaneous evidence from people taking commercial decisions in the market place. Aberdeen Journals has produced nothing to show 29 that the ordinary meaning of the documents on which the Director has relied in the contested decision is not the correct meaning.

[89]According to the Director, the evidence of the conduct of the parties and of contemporary documents support the conclusion that the Herald & Post was a “tactical tool” to protect the Evening Express from the Independent (see paragraph 181 of the decision) and that Aberdeen Journals’ conduct was only rational if viewed as a means of protecting the revenues of the Evening Express (paragraphs 107 to 109, and Annex 2, Graph 4, of the decision). For example, according to the Director, prior to the launch of the Independent , the Herald & Post was making a profit of around £20,000 a month. The monthly losses incurred on the Herald & Post in the period November 1998 to March 2000 were more than double that, save in respect of two months. According to the Director, the letters from advertisers (e.g. Cornhill Car Sales of 21 January 1999) also show that Aberdeen Journals and its advertiser customers viewed advertising in the Independent as a viable alternative to advertising in the Evening Express (paragraphs 110 to 114).

[90]Aberdeen Journals is wrong, says the Director, to claim that he has adduced no contemporaneous evidence that Aberdeen Journals’ strategy regarding the Independent lasted until March 2000: such evidence is contained in the memoranda cited at paragraph 181 of the decision, the last one dated January 2000, and in Graphs 1 and 2 of Annex 2 to the decision. Even in March 2000, says the Director, advertising rates of the Herald & Post were cut and pagination increased. The Director relies on the Tribunal’s comment in Napp Pharmaceuticals Limited v Director General of Fair Trading

[2002] CAT 1

[2002] CompAR 13 (“ Napp ”) that “it is relevant to take facts arising before 1 March 2000 into account for the purpose … of throwing light on facts and matters in issue on and after that date” (at paragraph [217]). For example, the letters concerning Aberdeen Journals' practice of giving exclusionary discounts show a consistent strategy from May 1996 to January 1999. There is nothing to suggest that the product market definition changed in January 1999 or at any other time in the relevant period. Aberdeen Journals' termination of such discounts was prompted not by a changed perception of th e market, but because of regulatory pressure from the Director.

[91]The Director rejects Aberdeen Journals’ argument that new entry cannot be evidence of market definition. On the contrary, according to the Director, the evidence in this case is that, in 1996, Mr Barwell’s primary objective was to target the Evening Express . That is supported by the size of the launch costs of the Independent relative to the much smaller 30 revenues being generated by the Herald & Post , and by the market study entitled “The Scottish Opportunity – Aberdeen”. Mr Barwell’s decision to continue to publish the Independent in the face of Aberdeen Journals’ attempts to force it to close further demonstrates a strong belief that the Independent was capable of competing with the Evening Express . The fact that Aberdeen Independent erroneously believed at launch that it could target both the Press & Journal and the Evening Express does not indicate a mistake in the Director’s analysis of the relevant product market, nor does it undermine the Director’s reliance on Mr Barwell’s targeting of the Evening Express as evidence of the relevant product market. The fact that the evidence from Aberdeen Independent does not all point one way does not mean that it should all be disregarded.

[92]The Director rejects Aberdeen Journals’ argument that the Independent was launched on an inefficient basis The Herald & Post's advertising rates were slashed and circulation levels increased, following launch of the Independent (Graphs 1, 2 and 7 in Annex 2 to the decision). The Independent accordingly found itself the subject of predation by the Herald & Post . It was not launched as a fireship: its sustained losses were the result of low prices resulting from predation by Aberdeen Journals. If Aberdeen Journals had genuinely believed that the Independent was not commercially viable, they would not have acted as they did. In fact, the notion of a fireship was not mentioned by Aberdeen Journals until their letter to the Director of 10 February 2000. Now that th e Herald & Post is no longer pricing below average variable costs, the Independent has moved into profit.

[93]Furthermore, the documentary evidence does not show, as Aberdeen Journals contends, that Mr Barwell launched the Independent on the artificial basis that he hoped to force Northcliffe to purchase it: see e.g. Mr Ezzat’s Review of January 2000. On the contrary, the evidence shows that it was Aberdeen Journals which wanted to purchase the Independent to remove it from the market. Moreover, the evidence does not indicate that Mr Barwell misrepresented the position to the Director’s officials at the meeting of 25 April 2002. The Director, by a letter to the Tribunal of 5 February 2003, objects strongly to the admissibility of the confidentiality agreement of 19 April 2002, on the ground that such a document should have been disclosed earlier if it was to be relied on. In any event, that document is not relevant to events between 1996 and 2000, and shows no inconsistency on the part of Mr Barwell.

[94]Aberdeen Journals’ criticisms of the Director’s findings regarding the advertising sales methods of the Independent are unfounded: an advertising manager is unlikely to contact an advertiser if he did not think the advertiser was a potential customer. It is wrong to dismiss 31 the evidence of Mr Farquharson as being irrelevant because he ceased to be an employee of Aberdeen Journals in May 1998, since there is nothing to suggest that market conditions have changed since May 1998. There is nothing, moreover, to indicate that this evidence comprises biased self-serving statements from a disgruntled ex-employee, as Aberdeen Journals suggests.

[95]The Director emphasises that there is substantial contemporaneous evidence that both Aberdeen Journals and Aberdeen Independent believed from March 1996 to March 2000 that the Independent , the Herald & Post and the Evening Express all competed in the same product market. This view both underpinned and explains the consistent commercial conduct of both parties throughout this period, pursued at considerable cost, as shown notably by the letter from Mr Scott of 10 February 2000 (see also paragraph [161] of the first judgment), and by Appendix 2 to the letter of 10 February 2000, Mr Pelosi’s memorandum of 21 May 1999, the note of the meeting of 5 August 1999 and Mr Ezzat’s review of 18 January 2000. The memo of 20 July 1999 is inconclusive : it does not prove the absence of a competitive relationship, as Aberdeen Journals contends.

[96]According to the Director, the fact that the Independent’s circulation was reduced in mid-February 2000 from 125,000 copies a week to 107,000 does not mean that the Evening Express was no longer the main target, or that the Independent no longer posed a threat to the Evening Express . Similarly Mr Ezzat’s figures, properly understood, indicate that the Independent was responsible for a significant proportion of the decline in the advertising revenues of the Evening Express noted in Mr Ezzat’s review of 18 January 2000. In this case, there is nothing to break the chain of causation in events from March 1996 to March 2000. The economic evidence generally

[97]As regards statistical and survey evidence, the Director submits that in a market where the conditions of competition have been distorted by predatory practices over a number of years, any such analysis is unlikely to be reliable for market definition purposes. Aberdeen Journals’ claim that under such circumstances, the observed absence of switching is all the more significant is irrelevant: switching between the Evening Express and the weekly free newspapers did occur (see for example, Mr Ezzat’s memo which notes that the Evening Express’s loss of revenue is in part attributable to the Independent ). 32

[98]As regards the OFT’s internal econometric analysis, the Director maintains that this was an internal document which the Director was not obliged to disclose (see Rule 14(6)(c) of the Director’s Rules). The Director’s letter of 8 August 2002 to Aberdeen Journals provided information about the analysis as a sign of good faith but the analysis is not relied on. According to the Director, Professor Yamey is correct to highlight that the analysis should have been based on prices at the competitive level and not on actual prices: this shows the difficulty in producing any meaningful statistical analysis in circumstances where prices are distorted away from the competitive level. The results of this analysis were not exculpatory, as Aberdeen Journals claims, but counter-intuitive. In particular, they appeared to indicate that an increase in the price of the Independent would lead to a reduction in volumes of the Herald & Post , when it is uncontested that these titles compete directly. The Director’s view that no firm conclusions could be drawn from the econometric analysis undertaken is consistent with the Tribunal’s observation that economic or econometric techniques may be of limited value in particular circumstances (paragraph [102] of the first judgment). Finally, Aberdeen Independent informed the Director that actual invoice data was not available for the period in question.

[99]As regards Aberdeen Journals’ claim that the Director should have compiled a survey of advertisers, the Director considers that given(a) that the OFT’s first limited survey did not provide any useful results and

(b) the strength of the other evidence, it was reasonable not to undertake a survey, especially since the Tribunal has recognised that survey evidence may be inconclusive (see paragraph [102] of the first judgment). Aberdeen Journals’ claim that the Competition Commission undertook a survey in 16 days in Johnston Press/Trinity Mirror cited above underestimates what is involved: the fieldwork may have taken this long but the design and planning of the survey and the analysis and reporting of the results would have taken months, not days. Finally, there is no burden on the Director to show that a survey would not have disproved his conclusions: if Aberdeen Journals contends that this is so, it should have undertaken a survey of advertisers itself. The RBB reports

[100]As regards the RBB reports, the Director submits that(a) RBB’s analysis ignores advertisers who used the Evening Express at some point other than in March 2000, notably advertisers who had already switched away from the Evening Express to the Herald & Post or the Independent because of their low prices. The sample is therefore biased towards those advertisers least likely to switch.

(b) Advertising is a seasonal business and a ten month sample period is therefore inadequate: at least two years’ worth of data should have been 33 used to see changes on a year-on-year basis.

(d) RBB have based their analysis on changes in circulation on the basis that a decrease in circulation represents an increase in the effective price of advertising. However, most advertisers are likely to be less sensitive to changes in circulation than of price: since most advertising in the Evening Express , the Herald & Post and the Independent is focused on Aberdeen, changes in circulation in the outlying areas (which the changes in question were), are less significant to advertisers than a change in price.

(e) RBB fails to take account of the fact that advertising for the Evening Express and the Herald & Post was sometimes bundled together by Aberdeen Journals, as noted in paragraphs 111 to 115 of the decision. Such bundling will distort switching between these titles and the Independent , as will Aberdeen Journals' discounts for exclusivity.

(f) The data relied upon by RBB is incomplete: nearly 20 per cent of the relevant data is missing, all in relation to the Independent .

(g) RBB takes no account of any time lag in switching. Because advertising campaigns are planned and booked in advance, it may take weeks or months for advertisers to react to price changes.

[101]The Director argues that the application of RBB’s methodology would indicate that there is no substitutability between the Herald & Post and the Independent , when there is no dispute that these titles at least share the same product market: this exposes the flaws in the RBB reports’ methodology. Moreover, the database used by RBB appears to be not only incomplete but inaccurate: Table 2 of the second RBB report omits between seven (on Aberdeen Independent's assessment) and eleven (on the Director's assessment) of the advertisers who used both the Evening Express and a free title in March 2000.

[102]Furthermore, the detailed findings in the first and second RBB reports are neither objective nor based on a balanced assessment of the data presented. In particular, says the Director, RBB has ignored results which could be indicative of switching between the Evening Express and the Independent and has overstated the extent of behaviour indicating a lack of substitution. For example, RBB overlooked Virgin Cinema which between December 1999 and March 2000 placed an increase of its advertising in the Independent , following an increase in the distribution of the latter, rather than in the Evening Express which had carried all Virgin’s advertising up to that point.

[103]In marked contrast to the conclusions reached by RBB, the Director’s review of the evidence gathered by RBB (see Annex 2 of the defence) indicated that only one of the 50 examples considered was indicative of a lack of substitution between the Herald & Post and the 34 Evening Express . As regards the 22 advertisers who were subject to a rate change, the Director identified two examples where the evidence appeared to indicate switching between the Evening Express and the free titles (Odeon Cinema, Castlegate Arts Ltd). The remainder were either not inconsistent with advertisers viewing the Evening Express and the free titles as substitutes (7 examples) or were not determinative either way (11 identified by RBB and a further 2 identified by the Director). As regards the 28 advertisers who were subject to distribution changes only, the Director identified two examples where the evidence appeared to indicate switching between the Evening Express and the free titles (Virgin Cinema, Messrs Jamieson and Cradock). The remainder were either not inconsistent with advertisers viewing the Evening Express and the free titles as substitutes (7 examples); or were not determinative either way (11 identified by RBB and a further 7 identified by the Director). Only one case indicated a lack of substitutability. Overall, the Director claims that his review of RBB’s analysis demonstrates that the evidence it presents is not inconsistent with the Evening Express being regarded by advertisers as a substitute for the free titles.

[104]Finally, even RBB’s results were inconclusive in relation to 11.7 per cent of advertisers in the Evening Express by volume in March 2000. Since only a small proportion of advertisers would need to switch to provide evidence of substitutability on the Director’s analysis , this significant proportion of inconclusive results means that the RBB reports cannot be said to demonstrate that the Evening Express , the Herald & Post and the Independent are not in the same product market. The impact of other media

[105]The Director submits that paragraphs 86 to 92 of the decision deal properly with possible constraints from alternative advertising media. Aberdeen Journals does not seriously contend that alternative media form part of the relevant market, only that the Herald & Post and the Independent together form a separate market. Dominance in the Director’s primary market

[106]The Director points out (defence in the first appeal, paragraphs 29 et seq) that his finding of dominance is based on both high market shares and barriers to entry. He was entitled to rely on market shares alone: see Case 85/76 Hoffman-La Roche v Commission [1979] ECR 461 (“ Hoffman-La Roche ”), paragraph 41;

Case T-30/89

Hilti v Commission

[1991] ECR II-1439

at paragraph 92. There are no “exceptional circumstances” here to rebut the presumption following from Aberdeen Journals’ high market share. The loss by the Herald & Post of 35 some customers in or before March 2000 does not negate dominance: see United Brands , cited above, at paragraph 65. In any event, the modification of Aberdeen Journals’ commercial conduct which apparently led to some loss of business did not come about as a result of competitive pressures, but as a result of the Director’s investigation. The alternative product market

[107]The Director rejects Aberdeen Journals’ criticisms of his alternative market definition set out at paragraphs 133 to 139 of the decision. The Director’s conclusion that the Evening Express alone constitutes a product market for paid-for local papers in Aberdeen is based notably on the conclusions set out at paragraphs 69 to 70 and 91 to 92 of the decision, to the effect that neither the Press & Journal nor alternative media exercise a sufficient competitive constraint on prices for advertising space in paid-for and free newspapers because of their significantly and specifically different characteristics (described at paragraphs 69 to 70 and 86 to 90 respectively of the decision). Even if the Press & Journal were to be included, Aberdeen Journals would still have 100 per cent of the market for advertising in paid-for newspapers with a local content circulating in Aberdeen. Aberdeen Journals’ position that the Director has failed to demonstrate a competitive relationship between the Evening Express and the free weekly newspapers is based on a misunderstanding of Tetra Pak II, cited above: it is necessary only to show sufficient “associative links”. Uncontested evidence of strong associative links is set out at paragraph 136 of the decision.

[108]As regards Aberdeen Journals’ argument that the Director failed to link the alternative market definition to his findings on abuse and penalty, the Director found in paragraphs 133 to 139 of the decision (notably paragraph 138) , that his findings of abuse were not sensitive to the alternative market definition. If the Tribunal finds that the alternative market definition is the correct one, then it is for the Tribunal to decide whether the level of penalty requires adjustment. However, since the paid-for and free local newspaper markets were both affected by Aberdeen Journals’ predatory behaviour (see paragraph 138 of the decision), the Director considers that no adjustment to the level of the penalty is required.

[109]The Director contends that he was entitled to include the alternative market definition in the decision. The alternative market definition is part of the question of market definition, and was acknowle dged by the Tribunal to be “potentially relevant” (paragraph [118] of the first judgment) and so falls squarely within the scope of the remitted “matter”. Aberdeen Journals was fully able to (and did) exercise its rights of defence regarding the alternative market definition and has not been unfairly prejudiced. There is no question of an abuse of process: 36 administrative proceedings are not “litigation”; and the present proceedings, following the Tribunal’s remittal, are part of the same litigation as the first appeal. Aberdeen Independent’s submissions

[110]Aberdeen Independent supports the position of the Director and has made only supplemental submissions.

[111]In relation to product market definition, Aberdeen Independent considers that even if (which it denie s) advertising in the two free weeklies is not strictly “substitutable ” for advertising in the Evening Express , the two types of newspaper are nevertheless sufficiently “interchangeable ” to form part of the same relevant market: see Hoffman-La Roche , cited above, at paragraph 28; Tetra Pak II , judgment of the Court of First Instance, cited above, at paragraph 63. Advertisers use both types of newspaper and if the two free weeklies were closed, they would switch to the Evening Express . The three titles therefore form part of the same product market. Aberdeen Independent further points to the evidence that the Herald & Post’s revenue barely covered the monthly running costs of the Independent ; and that the Herald & Post was used as a fighting ship to protect the Evening Express (see the memorandum of Mr Davidson of 12 July 1996). Aberdeen Independent notes that the Competition Commission heard evidence in its report on proposed transfers between Regional Independent Media Limited and Gannet UK Limited/Johnston Press plc/Guardian Media Group plc (November 2000, Cm 4887) to the effect that publishers of local titles fought hard to protect their markets from new entry and some would maintain a loss-making free newspaper to support a paid-for title as part of a layered or segmented market (see paragraph 2.25).

[112]Regarding the product differences between the newspapers in question, Aberdeen Independent considers that the Evening Express’s practice of having a weekly supplement for different types of advertising on different days of the week does not differentiate it from the Independent : the Independent simply publishes all types of advertising on the same day of the week, once a week. Aberdeen Independent also submitted a witness statement dated 9 January 2003 from Jill Hoyle of Media Force (which is an agency selling advertising space on behalf of over 471 regional newspapers to national advertisers) to the effect that free newspapers compete head on, at both a local and national level, with weekly and daily title s in each local marketplace. 37

[113]As regards its reduction in circulation in February 2000, Aberdeen Independent contests that the Independent was no longer targeting the Evening Express . As an established title, it did not need to continue a high level of circulation if the same revenue could be achieved at a lower circulation level. The opinions of Mr Ezzat are no more than his own view of possible strategies for the Independent .

[114]Aberdeen Independent denies Aberdeen Journals’ unsubstantiated contention that the Independent entered the market on an inefficient basis, and states that in the last quarter of 2002, the Independent moved into profit. There is no basis to assume that a free newspaper could not be successful in Aberdeen: indeed, as Mr Barwell’s first witness statement of 7 November 2001 comments, advertising rates in Aberdeen were ten times higher than those in England at the time of launch of the Independent . The background and strategy followed by Mr Barwell are fully explained in his second witness statement dated 17 May 2002, which refers notably to the report entitled “The Scottish Opportunity– Aberdeen”.

[115]As regards Aberdeen Journals’ submissions about the discussions for the sale of the Independent and the signature of a confidentiality agreement, Mr Barwell, in a third witness statement of 29 January 2003 stated that at the beginning of 2002, he had been told by his medical advisers to wind down his business commitments. On 24 January 2002 he wrote to several parties inquiring whether they would be interested in buying the Independent . That situation was forced upon Mr Barwell, and was not part of a long held plan. Mr Barwell denies making any misleading statements at the meeting of 25 April 2002. He had not been approached by Northcliffe with an offer to buy the Independent and had never, prior to 24 January 2002, made any suggestion that he would be interested in selling that newspaper to the Northcliffe Group.

[116]Aberdeen Independent also denies that it was in breach of Newspaper Society rules regarding giving information about changes in the number of copies distributed, and points out that the Advertising Standards Authority had also ruled against the Evening Express on two occasions on similar matters.

[117]Finally, Aberdeen Independent challe nges the status of the first and second RBB reports: the authors do not fulfil the requirements of experts giving expert evidence to a court: see CPR 35 and the Lord Chancellor’s Department’s Code of Guidance on Expert Evidence, in particular paragraphs 14 to 16. Aberdeen Independent further submits that it has not had the opportunity to check the accuracy of the data in the RBB reports about customers of 38 Aberdeen Journals or the Independent , which has not been disclosed to Aberdeen Independent for reasons of confidentiality. Up to 20 per cent of the data about the Independent is apparently missing from RBB’s sample. The basis on which the analysis was undertaken is impossible to work out: for instance, how have they treated smaller classified advertisements which form a large proportion of the whole ? Aberdeen Independent identifies substantive errors in the second RBB report, including seven omissions from a list of advertisers who used the Evening Express but not a weekly free newspaper in March 2000. In any event, says Aberdeen Independent, there are so many variables in an advertiser’s decision to place an advertisement that it is very difficult to draw conclusions from statistical analysis. IV THE TRIBUNAL’S FINDINGS ON RELEVANT MARKET AND DOMINANCE

[118]In this section we deal first with certain general issues. We then consider the case made by the Director in the decision to the effect that the relevant product market is the supply of advertising in local newspapers (paid-for and free) in the Aberdeen area. For the reasons given below, we conclude that the Director has adduced strong and compelling evidence to that effect, despite the criticisms made by Aberdeen Journals. We then turn to consider the case made by Aberdeen Journals to see whether that case raises doubts in our minds as regards the correctness of the Director’s analysis. For the reasons given below we conclude that no such doubt arises. We then deal briefly with alternative media, and with the issue of dominance in the market as defined by the Director. Finally we deal with the alternative product market advanced by the Director, namely ‘paid-for’ local newspapers in Aberdeen. 1. GENERAL A. The relevant law

[119]In the Tribunal’s judgment of 19 March 2002 in the first appeal, at [86] to [94], we summarised the relevant law in the following terms:

“86. … In order to fall within the Chapter II prohibition, it must be established that the undertaking in question has a dominant position. As traditionally defined, a dominant position is: “a position of economic strength enjoyed by an undertaking which enables it to prevent effective competition being maintained on the relevant market by allowing it the power to behave to an appreciable extent independently of its competitors, its customers and ultimately of the consumers.” 39 See Case 85/76 Hoffman-La Roche v Commission [1979] ECR 461 , paragraph 38; Case T-228/97 Irish Sugar v Commission [1999] ECR II-2969, paragraph 70. 87. However, …: “such a [dominant] position does not preclude some competition ... but enables the undertaking which profits by it, if not to determine, at least to have an appreciable influence on the conditions under which that competition will develop, and in any case to act largely in disregard of it so long as such conduct does not operate to its detriment.”

Hoffman-La Roche v Commission , cited above, at paragraph 39. 88. In order to determine whether, in any given case, an undertaking has the necessary degree of economic strength or, to use the more modern term, market power, so as to give rise to dominance, it is self-evidently necessary to define the market in which that market power is said to exist. As the Commission of the European Communities (“the Commission”) has put it in paragraph 2 of its Notice on the definition of relevant market for the purposes of Community competition law (“the Commission’s Notice on Market Definition ”) OJ 1997 C372/5:

“Market definition is a tool to identify and define the boundaries of competition between firms ... The objective of defining a market in both its product and geographic dimension is to identify those actual competitors of the undertakings involved that are capable of constraining those undertakings’ behaviour and of preventing them from behaving independently of effective competit ive pressure.” 89. The Director’s Guideline on Market Definition OFT 403, March 1999, follows the same approach: “The approach described in this guideline is not mechanical, it is a conceptual framework within which evidence can be organised. The Director General will not follow every step described below in every case. Instead, he will look at the areas of evidence which are relevant to the case in question – and will often be constrained by the extent to which evidence is available. Market definition is not an end in itself, but rather a step which helps in the process of determining whether undertakings possess, or will possess, market power.” (paragraph 1.5) … 91. As far as the relevant product market is concerned, the Court of Justice said in Hoffman-La Roche , cited above, “The concept of the relevant market in fact implies that there can be effective competition between the products which form part of it and this presupposes that there is a sufficient degree of interchangeability between all the products forming part of the same market in so far as a specific use of such products is concerned.” (paragraph 28). 92. In its judgment in Tetra Pak II [1994] ECR II-755 , the Court of First Instance held at paragraph 63: 40 “A preliminary point to note is that, according to settled case-law, the definition of the market in the relevant products must take account of the overall economic context, so as to be able to assess the actual economic power of the undertaking in question. In order to assess whether an undertaking is in a position to behave to an appreciable extent independently of its competitors and customers and consumers, it is necessary first to define the products which, although not capable of being substituted for other products, are sufficiently interchangeable with its products, not only in terms of the objective characteristics of those products, by virtue of which they are particularly suitable for satisfying constant needs, but also in terms of the competitive conditions and the structure of supply and demand on the market (see the judgment of the Court of Justice in Case 322/81 Michelin v Commission [1983] ECR 3461 , paragraph 37).” 93. In Case T-9/93 Schöller v Commission [1995] ECR II-1611 the Court of First Instance emphasised that “it is settled law that account must also be taken of the consumer’s point of view” (paragraph 40) 94. The overall effect of the above case law is summarised in the judgment of the Court of First Instance in Case T-504/93 Tiercé Ladbroke v Commission [1997] ECR II-923 at paragraph 81: “According to settled case-law, for the purposes of applying Article [82] of the Treaty, the relevant product or service market includes products or services which are substitutable or sufficiently interchangeable with the product or service in question, not only in terms of their objective characteristics, by virtue of which they are particularly suitable for satisfying the constant needs of consumers, but also in terms of the conditions of competition and/or the structure of supply and demand on the market in question (Case 31/80 L’Oréal [1980] ECR 3775, paragraph 25; Case 322/81 Michelin v Commission [1983] ECR 3461 , paragraph 37; Case C-62/86 AKZO Chemie v Commission [1991] ECR I-3359 , paragraph 51; Case T-30/89 Hilti v Commission [1991] ECR II-1439, paragraph 64, and Case T-83/91 Tetra Pak v Commission [1994] ECR II-755 , paragraph 63).”

[120]At [96] to [97] of the Tribunal’s first judgment, we identified the factors to be taken into account in defining the relevant product market, as follows:

“96. The foregoing cases indicate that the relevant product market is to be defined by reference to the facts in any given case, taking into account the whole economic context, which may include notably (i) the objective characteristics of the products; (ii) the degree of substitutability or interchangeability between the products, having regard to their relative prices and intended use; (iii) the competitive conditions; (iv) the structure of the supply and demand; and (v) the attitudes of consumers and users. 97. However, this check list is neither fixed, nor exhaustive, nor is every element mentioned in the case law necessarily mandatory in every case. Each case will depend on its own facts, and it is necessary to examine the 41 particular circumstances in order to answer what, at the end of the day, are relatively straightforward questions: do the products concerned sufficiently compete with each other to be sensibly regarded as being in the same market? Are there other products which should be regarded as competing in the same market? The key idea is that of a competitive constraint: do the other products alleged to form part of the same market act as a competitive constraint on the conduct of the allegedly dominant firm?”

[121]At [126] of the Tribunal’s first judgment we formulated the question we now have to answer in the following terms:

“[H]as the Director established that in March 2000 the activities of the Independent in the supply of advertising space in the Aberdeen area constituted a sufficient competitive constraint, or brought sufficient competitive pressure to bear, on the advertising business of the Evening Express , for those two newspapers sensibly to be regarded as both competing in the market for advertising space in local newspapers in Aberdeen? If the answer to that question is in the affirmative, then in our view the Director’s approach to the relevant market is correct.”

[122]No criticism having been made of those passages, we propose to follow the same approach in this second appeal. B. The burden and standard of proof

[123]It is common ground that the legal burden of proof rests throughout on the Director to prove the infringements alleged (see Napp , cited above, at [100]), albeit that the Director may properly rely on inferences or presumptions that would, in the absence of any countervailing indications, normally flow from a given set of facts: Napp , at [110] to [111].

[124]As to the standard of proof, the Tribunal in its first judgment (at [105]) followed the approach set out in Napp at [109]:

“In those circumstances the conclusion we reach is that, formally speaking, the standard of proof in proceedings under the Act involving penalties is the civil standard of proof, but that standard is to be applied bearing in mind that infringements of the Act are serious matters attracting severe financial penalties. It is for the Director to satisfy us in each case, on the basis of strong and compelling evidence, taking account of the seriousness of what is alleged, that the infringement is duly proved, the undertaking being entitled to the presumption of innocence, and to any reasonable doubt there may be.”

[125]Again, we propose to follow the same approach. We bear in mind, however, that an issue such as the relevant product market may require a more or less complex assessment of numerous interlocking factors, including economic evidence. Such an exercise intrinsically 42 involves an element of appreciation and the exercise of judgment. On such issues it seems to us that the question whether the Director has “proved” his case involves asking ourselves: Is the Tribunal satisfied that the Director’s analysis of the relevant product market is robust and soundly based? C. The Tribunal’s approach to evidence

[126]One of the principal submissions made by Aberdeen Journals is that the Tribunal should accept a ‘hierarchy’ of evidence, in which objective economic evidence, such as consumer surveys, market studies or statistical analysis, should be accorded greater weight than what Aberdeen Journals describes as “subjective” evidence deriving, for example, from the statements or conduct of the parties. In particular, says Aberdeen Journals, evidence on the demand side from consumers or users is of crucial importance, and likely to be more significant than evidence from the supply side. The relative lack of evidence from consumers or users in this case is, says Aberdeen Journals, unprecedented in EC law (see paragraph 61 above).

[127]In our view, there is no set “hierarchy” of evidence in Community law on issues such as market definition. As the European Commission puts it at paragraph 25 of its Notice on Market Definition OJ (1997) C272/5:

“There is a range of evidence permitting an assessment of the extent to which substitution would take place. In individual cases, certain types of evidence will be determinant, depending very much on the characteristics and specificity of the industry and products or services that are being examined. The same type of evidence may be of no importance in other cases. In most cases, a decision will have to be based on the consideration of a number of criteria and different items of evidence. The Commission follows an open approach to empirical evidence, aimed at making an effective use of all available information which may be relevant in individual cases. The Commission does not follow a rigid hierarchy of different sources of information or types of evidence.”

[128]Similarly, although evidence of the attitudes of consumers or users will often be highly pertinent to an analysis of the relevant product market, there is in our view no rule of law which requires the Director to base his case on consumer surveys or market studies if he considers that his case is sufficiently proved by other evidence. What evidence the Director chooses to rely on to establish a relevant product market is a matter for him. Whether that evidence is sufficient to prove the case, is ultimately a matter for the Tribunal. In deciding whether the evidence is sufficient, the Tribunal will pay attention to evidence about the attitudes of consumers or users, or the absence thereof, but that is only one element of the 43 Tribunal’s assessment of the evidence as a whole. In this case we propose to look at the evidence “in the round” in reaching our conclusion.

[129]In this case much of the evidence is in documentary form, supplemented by witness statements by, notably, Mr Scott, Managing Director of Aberdeen Journals, Mr Barwell, the proprietor of the Independent , and Mr Farhquarson, an ex-employee of Aberdeen Journals. Aberdeen Journals invites us, for various reasons, to place little weight on the documents relied on by the Director emanating from Aberdeen Journals/Northcliffe, and to give no credence to the witness statements of Mr Barwell and Mr Farquharson.

[130]Rule 20(2) of the Tribunal’s Rules 5 provides:

“The tribunal may admit or exclude evidence, whether or not the evidence was available to the respondent when the disputed decision was taken and notwithstanding any enactment or rule of law relating to the admissibility of evidence in proceedings before a court.”

[131]Having regard to Rule 20(2), the question of whether a document is “inadmissible ” in evidence in the traditional sense will rarely arise. However, many factors, including whether the document contains “hearsay”, may affect the weight to be given to the document in question.

[132]The Tribunal’s normal approach to documentary evidence is to give the document in question what appears to be its natural meaning, and accord it such weight as appears appropriate, given the time when, and circumstances in which, it was prepared, the identity of the author, and any particular factors likely to undermine its credibility. If, as is the case here, the applicant wishes to contest the meaning or significance of a document emanating from the applicant and relied on by the Director, it is open to the applicant to produce a witness statement from the author of the document to clarify, place in context, or explain away the document in question. The Director, if he thought fit, would then be entitled to seek to cross-examine the witness under Rule 17(2)(g) of the Tribunal’s Rules. In the absence of any such witness statement, the Tribunal will normally feel entitled to take a document at face value, giving it such weight as is appropriate.

[133]As regards witness statements, any party has the right to seek to cross-examine a person who has provided a witness statement to the Tribunal under Rule 17(2)(g). In this case, no such 5 The Competition Commission Appeal Tribunal Rules 2000 S.I. 2000 no. 261. As from 20 June 2003 the Competition Appeal Tribunal Rules 2003 apply to appeals lodged on or after that date: S.I. 2003 no. 1372. 44 application has been made. Again, our approach is to give witness statements such weight as seems appropriate in the circumstances, bearing in mind whether cross-examination has been sought.

[134]If there are matters relied on by the Director which could have been contested by way of a witness statement, the fact that there is no witness statement about it from the applicant is also a factor which the Tribunal may bear in mind when assessing the evidence as a whole. 2. THE CASE MADE BY THE DIRECTOR

[135]In the decision the Director relies on different kinds of evidence which, in our view, may be grouped under four headings:(a) the characteristics of the products;

(b) the commercial strategy of the Independent ;

(d) further documentary evidence supplied by Aberdeen Journals. In this section we examine each of these headings in turn, in the light of the arguments of the parties, to determine whether that evidence, standing alone, would suffice to prove the Director’s case, despite the criticisms that Aberdeen Journals makes of the evidence in question. A. The characteristics of the products in question

[136]This case concerns the supply and demand for local newspaper advertising in the Aberdeen area. As the decision points out, notably at paragraphs 31 to 40, local advertisers are of many different kinds and may wish to place different types of advertisement, often in more than one newspaper circulating in the area. The principal kinds of advertisements are “display” advertisements, i.e. advertisements of a general nature appearing within the newspaper’s news pages; and “classified” advertisements, i.e. advertisements classified according to content and included within dedicated sections of the newspaper. Classified advertisements fall into categories such as recruitment, property, motors and “other”, which latter includes articles for sale, local notices, and notices of births, marriages and deaths. According to the decision, classified advertisements may account on average for two-thirds of advertising revenue for local newspapers (paragraph 40). The Press & Journal

[137]The Press & Journal is a broadsheet, paid-for daily newspaper circulating throughout the North of Scotland. The Director excludes the Press & Journal from the relevant product market for the reasons given at paragraphs 50 to 54 and 69 to 70 of the decision. 45

[138]That approach has not been contested by Aberdeen Journals. It is true that, in the evidence, there are indications that, at least in the early stages, the Independent saw the Press & Journal as part of its target market (see e.g. “The Scottish Opportunity – Aberdeen” referred to above). Similarly it appears that at various times Aberdeen Journals saw the Independent as having an impact not only on the Evening Express but also on the Press & Journal . This evidence indicates to us that the boundaries of the market for local advertising in Aberdeen may, at the margin, be somewhat blurred. However, the reasons given by the Director, at paragraphs 50 to 54 and 69 to 70 of the decision, for excluding the Press & Journal , seem to us to be soundly based. We also agree with the Director’s statement, at footnote 62 and paragraph 70 of the decision, that if the Press & Journal were to be included in the relevant market, that would only enhance Aberdeen Journals alleged dominance in the Aberdeen area.

[139]Leaving aside the Press & Journal , a potential advertiser in the Aberdeen area in March 2000 would have been presented with a choice between one or more of the Evening Express , the Herald & Post and the Independent . The picture as painted in the decision as at March 2000 may be summarised as follows: The Evening Express

[140]The Evening Express is a paid-for, tabloid, daily, evening newspaper, focussed on Aberdeen and its outlying travel-to-work area. The cover price during the relevant period was 30p. According to the decision, its editorial coverage covered mostly local news and human interest stories, with no regular business coverage but some regular financial market information. Sports coverage was limited to local sports events and there was a local entertainment guide. It seems that advertising content comprised some 38 per cent of the whole.

[141]The Evening Express’s advertising is predominantly focussed on local businesses. About 70 per cent by value of advertising is represented by classified advertising, the remainder being display advertisements. Measured by value, the most important categories of classified advertisement were recruitment (23 per cent), property (11 per cent), motors (28 per cent) and other (37 per cent). Supplements were published for property on Friday, recruitment on Fridays and Saturdays, and motors on Tuesdays and Thursdays. The average yield for the Evening Express i n January to March 2000 was £4.90 per sccm. 46 The Herald & Post

[142]The Herald & Post is a free, tabloid, weekly newspaper, distributed every Wednesday, mainly in the City of Aberdeen. Before the entry of the Independent , the Herald & Post averaged only 20 pages per edition, and contained little or no property or motors advertising (paragraph 61 of the decision, and footnote 55). Following the Independent’s entry, the Herald & Post’s advertising rates were substantially reduced, and the quantity of advertising space increased dramatically, as did the quality of editorial coverage. The distribution was also increased. It appears that in March 2000 the Herald & Post averaged about 72 pages per edition.

[143]According to paragraph 62 of the decision, around 76 per cent of the Herald & Post’s content is advertising. The editorial coverage is said by the Director to be similar in tone to the Evening Express and the Independent , concentrating on local news and sport and human interest stories. The Herald & Post publishes a local entertainment guide but has more limited TV listings than the Independent and the Evening Express .

[144]Advertising in the Herald & Post is heavily localised. About 42 per cent by value were display advertisements, and 58 per cent by value classified advertisements. In March 2000 the Herald & Post published dedicated sections for property, recruitment and motors. Measured by value, the most important categories of classified advertisements were recruitment (27 per cent), property (17 per cent), motors (36 per cent) and other (21 per cent). In January to March 2000, the advertising yield for the Herald & Post averaged £1.22 per sccm. The Independent

[145]The Independent is a free, tabloid, weekly newspaper, distributed every Thursday in the City of Aberdeen and outlying areas. Geographically, the coverage is wider than that of the Herald & Post , and, as we understand it, for much of the period in question approximated to that of the Evening Express . It appears that about 73 per cent of the Independent’s content is advertising. The tone of its news coverage is broadly similar to that of the Evening Express and the Herald & Post , concentrating on local news and sport and human interest stories. A local entertainment guide is published.

[146]About 28 per cent by value of the Independent’s advertising is accounted for by display advertising, and 72 per cent by value is accounted for by classified advertising. There are 47 dedicated sections for recruitment, property and motors. In January to March 2000, the average advertising yield for the Independent was £1.88 per sccm.

[147]At paragraph 69 of the decision the Director concluded that, as far as product characteristics are concerned, the Evening Express sat “somewhere between” the Press & Journal and the Herald & Post and Independent , “sharing with the Press & Journal the fact that it is a daily paid-for paper but sharing it s format, circulation area, and general style more closely with the two weekly free newspapers”. Analysis of the characteristics of the product

[148]It follows from the foregoing that local advertisers in the Aberdeen area wishing to place a display or classif ied advertisement in March 2000 had, in principle, three newspapers to choose from, namely the Evening Express , the Herald & Post and the Independent , all three of which are tabloid newspapers similar in tone and style . The choice available to the advertiser appears to be particularly clear in the sections devoted to property, motors, and recruitment advertisements.

[149]Our own visual inspection of an admittedly limited sample of those newspapers for the week beginning 20 March 2000 (Monday 20 th , Tuesday 21 st , Wednesday 22 nd , Thursday 23 rd , Friday 24 th and Saturday 25 th for the Evening Express ; Wednesday 22 March for the Herald & Post , and Thursday 23 March for the Independent ) seems to confirm that the three newspapers are presented in similar tabloid format and style, save that the Independent and the Herald & Post carry considerably more advertising and have less editorial coverage, than the Evening Express , and vice versa. Visually speaking, it is difficult to detect much material difference in the layout and appearance of many of the classified advertisements appearing in the three newspapers. It is also apparent that many of the business advertisers in question advertise in more than one newspaper, sometimes in identical terms.

[150]It is, however, argued by Aberdeen Journals that the Independent is not in competition with the Evening Express , but is at most a complement to the latter. The three main points made are that(i) the Evening Express is paid for by the reader, whereas the Independent and Herald & Po st are free;

(ii) the Evening Express is a daily newspaper, whereas the Independent and Herald & Post are weekly; and

(iii) there are other significant differences in quality and readership profile between the three newspapers. For these reasons, says Aberdeen Journals, there is no significant competitive overlap between the Evening Express and the Independent 48 and Herald & Post . As far as advertising is concerned they are “complements”, not “substitutes”. — Paid -for versus free

[151]At paragraphs 76 to 82 of the decision, the Director points out that, the fact that a reader has to pay for a newspaper does not mean that paid-for and free newspapers are not substitutable from the point of view of an advertiser . The Director relies on the findings in a number of reports by the Competition Commission and its predecessor the Monopolies and Mergers Commission to the effect that paid-for and free weekly newspapers are in competition as far as advertisers are concerned.

[152]The Competition Commission reports referred to by the Director, namely Johnston Press plc and Home Counties Newspaper Holdings plc , June 1998, at paragraphs 2.9; Portsmouth & Sunderland Newspapers plc and Johnston Press plc/Newsquest (Investments) Limited News Communications and Media plc , June 1999, at paragraphs 2.22 and 4.23; News Communications and Media plc and Newsquest (Investments) Limited/Johnston Press plc/ Trinity Mirror plc , April 2000 at paragraph 4.52; and Johnston Press plc/Trinity Mirror plc , May 2002, at paragraphs 5.11 to 5.13 and 5.126, are all to the effect that weekly paid-for and free newspapers compete directly for readers and advertisers.

[153]Although the reports in question are concerned with competition between weekly titles, we for our part can see no reason for saying that a title which is free to the reader is necessarily in a different market to a paid-for title as far as the advertiser is concerned. As the Director points out at paragraph 76 of the decision, the key factor for the advertiser is whether the readership of a newspaper gives broadly equivalent coverage for an advertisement, at a comparable or lower cost. That will not depend on whether or not the newspaper is ‘free’ to the reader, but on other factors such as the circulation and the target readership which the newspaper is intended to reach.

[154]In the present case in March 2000 the Independent reached most households in urban Aberdeen, i.e. the same households that the Evening Express could have been expected to reach. 49 — Weekly versus daily

[155]On the perhaps more important question of how far the daily Evening Express is in competition with the weekly Independent for the business of advertisers in Aberdeen, the Director states at paragraph 83 of the decision that he can see no compelling reason why weekly newspapers cannot compete for advertising with daily newspapers. Any increased flexibility the daily newspaper may offer is unlikely to be relevant at local level, notably as regards the important classified advertising categories of recruitment, motors and property, which tend to be placed on a weekly basis.

[156]Although a weekly free newspaper may in certain circumstances be a closer substitute for a weekly paid-for newspaper than for a daily paid-for newspaper, we agree with the Director that there is no intrinsic reason why a weekly free newspaper should not be in competition with a daily paid-for newspaper for the business of a significant number of advertisers.

[157]It is true, as Aberdeen Journals point out, that in its report on Johnston Press plc/Trinity Mirror plc May 2002 (“ Johnston Press/Trinity Mirror ”), the Competition Commission said this at paragraph 5.125: “5.125. Looking first at daily (evening) and weekly newspapers, Trinity Mirror suggested to us that evening titles (but not morning newspapers which had a wider circulation area), paid-for weeklies and free weeklies were all in the same advertising market. However, the opportunities for substitution by advertisers between these seem to be relatively limited, for essentially two reasons. First, evening titles offer advertisers more flexibility than weekly titles in that advertisements can be placed or changed at more frequent intervals. Secondly, the two types of newspaper tend to be read and used in different ways, with weekly newspapers more likely to be retained by the household over several days for reference, compared with dailies whose impact is more short-lived. It has been put to us that advertisers therefore view the two forms of advertising as complements rather than substitutes, with advertising in dailies being used for immediate impact on the reader, and that in weeklies being used to reinforce the advertising message by providing more detailed information. On this point, Johnston commented that this assessment failed to take account of the increased use of ‘supplements’ by evening titles, ie the inclusion of a separate section for a particular type of advertising (for example, motoring, recruitment etc) on a particular day. There was no reason to suppose, it added, that readers who were interested in such advertisements would not retain the relevant supplement in the same way that they might retain a weekly newspaper. We note also in this regard that paid-for evening newspapers typically command significantly higher advertising yields (ie 10 per cent or more higher) than do weekly titles operating in the same local area.” 50

[158]Those views are similar to the views expressed by the Competition Commission in Regional Independent Media Limited and Garnett UK Limited/Johnston Press plc/Guardian Media group plc , (“the RIM report”), November 2000, at paragraph 4.34:

“4.34. In terms of the distinction between daily and weekly newspapers, it would again appear that the opportunities for substitution by advertisers between the media are relatively limited, for essentially two reasons. First, daily titles offer advertisers more flexibility than weekly titles in that advertisements can be placed or changed at more frequent intervals. Second, the two types of newspaper tend to be read and used in different ways, with weekly newspapers more likely to be retained by the household over a number of days for reference, compared with dailies whose impact is more short-lived. This suggests that advertisers will tend to view the two forms of advertising as complements rather than substitutes, with advertising in dailies being used for immediate impact on the reader, and advertising in weeklies being used to reinforce the advertising message through more detailed information provision. ”

(emphasis added)

[159]However on the facts of the present case it does not seem to us that the flexibility and other matters to which the Competition Commission refers in the above citation are likely to be significant, particularly in the context of classified advertisements. In particular, in the case of the Evening Express , supplements or sections for particular kinds of advertisements tend, as we understand it, to be published weekly, for example property on Fridays, recruitment on Fridays and Saturdays and motors on Tuesdays and Thursdays. We can see no reason in principle why, from the point of view of advertisers, the property section of the Independent published on a Thursday should not represent an alternative to the property section of the Evening Express published on a Friday. Similarly the motors section of the Independent published on a Thursday appears to us to be competing for business with the motors section of the Evening Express published on the same day. The recruitment section of the Independent published on Thursdays presents an alternative to the recruitment section of the Evening Express published on Fridays and Saturdays. Similarly, for smaller classified advertisements, we see no reason in principle why an advertiser wishing to place an advertisement for the sale of a motor car should not see the Independent on a Thursday as an alternative to the Evening Express on a Thursday. We bear particularly in mind that the Independent is delivered to most households in Aberdeen, so the potential readership of the Evening Express and the Independent overlaps to a very large extent. Any difference in response rates between the two newspapers will simply be reflected in their relative prices.

[160]We also note in passing that in paragraph 5.125 of the Johnston Press/Trinity Mirror report cited above both Trinity Mirror and Johnston Press apparently saw paid-for evening titles, paid-for weeklies and free weeklies as all being in the same advertising market. Although we 51 bear in mind that in that case both those companies had an interest in arguing for a wide market, the points they make are consistent with the evidence before the Tribunal in this case.

[161]The evidence before the Tribunal is also consistent with the general trend in the local newspaper industry noted by the Competition Commission, namely that local free newspapers have been competing increasingly effectively with local paid-for newspapers. It is also consistent with the statement of Jill Hoyle, submitted by Aberdeen Independent and not contested by Aberdeen Journals, who has worked in the regional newspaper business for 20 years, mainly selling advertising space. Ms Hoyle states: “Free newspapers have brought significant competition to many local market-places across the country. Free newspapers compete head on, at both a local and national level, with weekly and daily titles in each local marketplace. In my experience the national advertiser simply views local newspapers as commodities and judges each paper on its merits – be it paid or free. I witness strong competition between both paid and free rival newspapers across the UK.” — Other differences

[162]Mr Scott, in his witness statement of 1 February 2001, points out, notably, that readership per copy for “paid-for” newspapers in Aberdeen is approximately double that of free newspapers, and that readers spend less than half the time reading the free newspapers as they spend reading Aberdeen Journals’ paid-for titles. The Evening Express is targeted towards the 15-44 age group, says Mr Scott, whereas the Herald & Post is targeted towards those that are 44+. There are also obvious differences in the depth of editorial coverage, and in the fact that 27 per cent of the circulation of the Evening Express is outside urban Aberdeen. Aberdeen Journals also relies on the differences in yields as between the two newspapers.

[163]Again, it seems to us that none of those considerations are decisive when it comes to the question whether the Evening Express , the Independent and the Herald & Post are competing in the same market for advertisers . It is evident from the Competition Commission reports already referred to that the kind of differences between the paid-for titles and free titles to which Mr Scott refers are not confined to the Aberdeen area but apply to paid-for and free titles generally. That has not, however, prevented the Competition Commission from concluding that paid-for and free newspapers may be in strong competition with each other for the business of advertisers: see paragraph 152 above. In the present case, that appears to be notably so as regards property, motors, recruitment and other kinds of classified advertisements. 52

[164]We accept that there is a marked difference between the average yields achieved by the Evening Express (£4.90 per sccm) and the Independent (£1.88 per sccm) in January to March 2000. For the reasons given below, it is hard to attach importance to the size of this gap, because it reflects the distorted conditions of the market over the previous few years. The yield for the Evening Express for the last quarter of 1995 (£5.46 per sccm, according to our calculations from Annex 11 to the decision) compared to the Independent ’s initial yield at launch (just under £5 per sccm according to Annex 2, Graph 7 of the decision) appears to us to be a more reliable indicator of what might be expected in normal, rather than distorted, market conditions.

[165]Moreover, in principle a difference in yields per sccm between a paid-for and free newspaper would not seem to us to weaken the Director’s case. From an advertiser’s point of view, he has, broadly speaking, the choice of paying more for advertising in the Evening Express , in return for a higher readership per copy, a longer reading time per copy, and perhaps some difference in the “quality” of the readership, or paying less for advertising in the Independent , knowing that the reading time per copy, for example, is less. In other words, from the advertiser’s point of view, such “product” differences as there may be are compensated for by the price difference. As in many other markets, in our view advertisers are likely to be regularly weighing the price/quality equation and, depending on shifts in relative prices, deciding whether to switch all or part of their business from one newspaper to the other. That is a normal competitive situation, and we see no reason in principle why in normal circumstances the relative price and quality of the Independent should not act in that way as a constraint on the pricing policy of the Evening Express . — “ Substitutes” and “complements”

[166]That takes us on to one of the principal issues raised by Aberdeen Journals, namely whether and to what extent the Independent is a substitute, or potential substitute, for advertisers in the Evening Express , or whether the Independent and the Herald & Post are to be regarded purely as “complements” which are not in competition with the Evening Express .

[167]The Director deals with this issue at paragraphs 24 and 25 of the decision:

“24 As noted at paragraph 33 below, in certain circumstances advertisers may pla ce advertisements in more than one medium, or more than one newspaper, to reach a maximum number of potential customers in the most cost-effective manner. To this extent, an advertiser may use the different media concerned as complements to each other. In other words, whilst advertising in one medium may not be wholly or directly substitutable with 53 advertising in another medium, for example because it conveys a slightly different message aimed at a different target audience, it may be complementary to it, in the sense that it reinforces the advertiser’s overall message and thus increases the effectiveness of each type of advertising. It is in this sense that the term ‘complement’ is generally used by advertisers and providers of newspapers advertising space. 25 It is important to note in the context of this case that the potential for advertising space in a particular newspaper to act as a constraint on prices for space in another newspaper is not necessarily dependent upon whether the former is viewed, on these terms, as a complement or substitute for the latter. If a particular advertiser uses two newspapers at the same time, and is to this extent using them as complements, he is still able to vary his spread of advertising between the two newspapers, depending on relative prices. Such switching of advertising spending at the margins is itself capable of acting as a constraint on prices. In addition, it should be noted that the degree to which an advertiser will view space in a particular newspaper as a substitute or complement for space in another newspaper will depend to a large extent on the relative price of advertising space in the newspapers concerned. Thus, whereas an advertiser may view advertising space in one newspaper, at a certain price, as a complement for space in another newspaper, rather than as a direct substitute, this perception may change if prices of the former are raised to a sufficient degree.”

[168]Aberdeen Journals has not challenged those observations by the Director. Nor has it challenged the Tribunal’s own observations at [144] of our first judgment:

“144. However, depending on the circumstances, the idea that two products are, loosely speaking, “complements” does not necessarily exclude the possibility that they are also substitutes. Thus, a particular advertiser may have an advertising budget that he chooses to divide between different means of communication in the hope of reaching slightly different audiences, so that the different media in question are, in a loose sense, ‘complementary’. On the other hand, depending on the products in question, changes in relative advertising rates may still lead to switching between the different means of communication as advertisers choose to devote a greater proportion of their advertising budget to one product rather than another. Thus the comments of the reporting panels of the Competition Commission, cited above, notably [paragraph 4.34 of] the RIM Report, to the effect that in some circumstances daily and weekly newspaper titles or free and paid-for newspapers might be viewed as ‘complements rather than substitutes’ do not exclude the possibility that advertisers might switch a proportion or even perhaps all their advertising between a daily and weekly title (or between a free and paid-for title) if the changes in the advertising rates made it sufficiently attractive to do so. This decision by advertisers would equally be influenced by such matters as changes in pagination or distribution area, which might make advertising in one kind of newspaper rather than another relatively more attractive.”

[169]In our view those comments apply not only to paragraph 4.34 of the RIM report, but also to paragraph 5.125 of the Johnston Press/Trinity Mirror report, cited above. 54

[170]Applying these principles to the present case, it seems to us likely that there will be some advertisers in the Evening Express who would not regard the Independent or the Herald & Post as a particularly close substitute, and vice versa. Equally there may be advertisers who tend to use both newspapers in order to reach slightly different audiences – for example some motor dealers may tend to advertise new cars in the Evening Express rather than in the Independent , while some recruitment agencies may advertise senior professional vacancie s in the Evening Express rather than the Independent , as Mr Scott suggests in his statement. However, in the ordinary way we would expect that, for a significant number of advertisers, there is unlikely to be a rigid and immutable dividing line between the Evening Express and the Independent , such as to render the relative prices of the two newspapers immaterial to the advertiser’s decision. We would expect that, from the advertiser’s point of view, the choice as between the various newspapers, or as to the relative volumes of advertising to be placed in the different newspapers will, for many advertisers, continue to be governed by a balance between, on the one hand, the price and, on the other hand, the relative commercial advantages of advertising in one newspaper rather than another.

[171]In other words, even if it could be said that, to some extent, the market for advertising in local newspapers in Aberdeen is “segmented” as between advertising in “paid-for” newspapers, and advertising in “free” newspapers, we would not expect the boundary between these two possible “segments”, if they exist, to be clear-cut, fixed or impermeable , so as to give rise to separate markets for present purposes (see also paragraphs 318 and 319 below). We bear in mind, in particular, that it is not suggested in the present case that the readership of the newspapers is quite different, as it may be, for example, as between a national broadsheet and a national tabloid. On the contrary, at least in urban Aberdeen, and for much of the period prior to March 2000 in the outlying areas as well, the potential readership of both the Evening Express and the Independent was largely the same, given that most readers of the Evening Express live in Aberdeen and the Independent was delivered to most households in the Aberdeen area.

[172]For these reasons, we conclude, from our examination of the characteristics of the products concerned, that the possibility that a significant number of advertisers might switch some or all of their advertising from the Evening Express to the Independent , or might threaten to do so, is likely, at first sight, to act as a competitive constraint on the pricing policies of the Evening Express . We point out in that regard that it is not necessary for all, or even most, advertisers in the Evening Express to be willing or able to contemplate a switch, or change in their “mix”. In principle, it is only necessary for a sufficient proportion of the advertisers or 55 potential advertisers in the Evening Express to have a choice as to the relative volumes they are prepared to place with each newspaper for an effective competitive constraint to exist.

[173]Our examination of the characteristics of the products in question thus indicates to us that, at first sight, the Independent is, actually or potentially, in competition not only with the Herald & Post but also with the Evening Express , for the business of local advertisers in the Aberdeen area.

[174]We now turn to a more detailed examination of the evidence to see whether that provisional conclusion is confirmed. B. The commercial strategy of the Independent

[175]At paragraphs 102 to 106 of the decision, the Director relies on the launch strategy of the Independent as providing strong evidence that, from its launch in 1996 and up until at least March 2000 the Independent was competing both with the Evening Express and the Herald & Post . He relies on the witness statement of Mr Barwell; the document prepared prior to the launch of the Independent entitled “The Scottish Opportunity – Aberdeen”; th e investment made by Mr Barwell; and his decision to remain in the market despite Aberdeen Journals’ response. Aberdeen Journals submits, essentially, that the evidence relied on by the Director is insufficient to support his conclusion; that Mr Barwell and Mr Robins, the Managing Director of the Independent , have made inconsistent statements and are, in any event, unreliable; and that the Independent was launched as an unviable “fireship” with the intention of forcing Northcliffe to buy it. The position in 1996

[176]In considering these arguments, we start with the position as it was in 1996. At that time, according to the figures at Annex 11 of the decision, the Evening Express accounted for around 90 per cent by value of local newspaper advertising, with the Herald & Post accounting for around 10 per cent. The Herald & Post had about 20 pages per edition, with little or no motors or property advertising. Its monthly contribution prior to March 1996 appears to have been averaging some £20,000 a month or less, see Annex 2, Graph 4, of the decision. Aberdeen Journals enjoyed a monopoly of local newspaper advertising. 56 “The Scottish Opportunity – Aberdeen”

[177]Prior to the launch of the Independent , Mr Alder, who later became Chief Executive of the Independent , carried out a study for Mr Barwell entitled “The Scottish Opportunity – Aberdeen”. That document begins with a “Summary” that is in these terms:

“Launching into the Aberdeen market is a viable proposition, however it will take an initial investment of between £1.5m to £2m and will take three years to return its first profits. Initial losses of £881,472 in year 1 followed by a loss of £294,603 in year 2 will start to be reversed by a profit of £46,953 in year 3. By year 5, initial losses will have been recouped and by this point the newspaper is capable of producing a turnover of £5m and making £1m profit – such is the opportunity in the marketplace. Aberdeen is a £26m publishing market with the Aberdeen Journals having all of it. The Press and Journal is by far the market leader in both revenue and volume. It has 70% of all revenue £18m) and 55% of advertising volumes. Its average yield is a staggering £10.27 scc. To gain a foothold and acceptance in the market will take more than just breaking the monopoly by launching into the area. The product introduced will have to be of the highest quality and must set out to be cheaper and better than its competitors. With the right staff and everyone pulling as a team in the same direction that can be achieved. By introducing a range of new initiatives into the marketplace, the opposition can be reduced to reaction. By grasping the initiative in the market and being committed to keeping it, Aberdeen Journals will find us a handful. I believe the project will succeed spectacularly but it first needs to dispel the stigma of free newspapers which has been deliberately perpetrated by Aberdeen Journals. I believe the deliberate way in which Thomson’s in the past have manipulated the market to suit their own needs will come home to roost. And past industrial disputes have left a legacy of mistrust and anger in some important local institutions.”

[178]It is then stated under the heading “Local media”:

“Yields are significantly higher than most marketplaces in the UK i.e. Press and Journal average yield is £10.27 single column centimetre (scc) and that is a conservative estimate. Even the Herald & Post, which isn’t held in very high regard in the area, achieves an average yield of £4.21 scc. Advertisers are well aware of the hopelessness of their situation in respect of press advertising. They fully admit they have no choice because of the monopoly held by Aberdeen Journals. 57 They also do not see local radio or TV as a real alternative to press advertising. The advertisers that I interviewed who used TV/radio, only did so for specific purposes i.e. image or new launches.”

[179]Turning to the newspapers in the area, Mr Alder considered that the Press & Journal was “still expensive and could easily be vulnerable to an aggressive competitor”. As regards the Evening Express he said:

“Undoubtedly the Evening Express is vulnerable, especially when the above readership levels are related to rate card/yields for cost effectiveness.”

[180]As regards the Herald & Post , Mr Alder said:

“This title is also part of Aberdeen Journals – I have no need to introduce either the title’s name or the free newspaper concept any further than but to say that this is a very poor example of the branded product. The Herald and Post is published every Wednesday. Paginations are only averaging 20 pages at present and major advertising markets i.e. property and motors have little or no presence. … The Herald & Post is being maintained purely to discourage any potential aggressor from launching into the market – however in its current format, it offers little protection. Serious advertisers think the publication is a joke and many of them said it “goes straight in the waste bin” or “it isn’t being delivered properly”

. Aberdeen Journals are maintaining it at a level where it doesn’t lose them any money. Yield is good averaging £4.21 scc and its leaflet delivery business is going a storm. It regularly carries 3-4 leaflets/brochures every week.”

[181]Mr Alder then set out his perceived strategy for the launch of the Independent . It is of interest he said of ‘motors’, for example:

“It should be relatively easy to take a slice of new and used car advertising from both the Press & Journal and the Evening Express. Coverage in the Herald & Post is restricted to an odd page for Anderson Cars or Reg Vardy. The evening paper is vulnerable highlighted by its own research information with only 31,000 readers for new car advertising and 69,000 for used car advertising much of which relates purely to privately advertised cars. The better coverage and much more competitive rates coupled with their ability to now negotiate in a competitive market should ensure a healthy number of motor dealers come our way.”

[182]The study “ The Scottish Opportunity – Aberdeen ”, read as a whole in our view strongly supports the Director’s case that the Independent was launched as a competitor to, notably, the Evening Express . 58 The launch of the Independent against the Evening Express

[183]Following Mr Alder’s study, the Independent was launched in the Aberdeen area as a “high quality” free newspaper with a pagination significantly above that of the Herald & Post , a distribution area that was wider than that of the Herald & Post , and at advertising rates that were higher than those then being charged by the Herald & Post : see Annex 2, Graph 7 to the decision, but lower than those of the Evening Express . According to Mr Barwell, the Independent carried a considerable amount of advertising for property and motors, up to then principally carried by the Evening Express and to a lesser extent the Press & Journal , as well as a detailed TV Guide which the Herald & Post did not have but the Evening Express did.

[184]Moreover, as Mr Barwell points out, credibly in our view, his initial investment of between £1.5 million and £2 million would not have been made if he was simply aiming to compete with the modest revenues of the Herald & Post rather than with the Evening Express , and, to some extent, with the Press & Journal .

[185]Against that background, we accept Mr Barwell’s evidence, which is corroborated by the matters set out above, that the Independent was not launched primarily as a competitor of the Herald & Post but was pitched with a view, notably, to obtaining business from the Evening Express . Indeed, the evidence before the Tribunal is consistently to the effect that the Independent saw the Evening Express as a principal competitor at the time of launch. That evidence in our view supports the case made by the Director. The Independent’s activities after launch

[186]After the launch of the Independent , it is common ground that Aberdeen Journals reacted by drastically cutting the advertising rates of the Herald & Post , often to below average variable cost, while at the same time increasing the pagination and distribution of that newspaper, thus incurring heavy losses. The Independent was constrained to reduce its rates in response, although not as low as those of the Herald & Post (see Annex 2, Graph 7). Thereafter, the Independent remained in the market, although incurring heavy losses, throughout the period from 1996 to 2000. In our view, from those facts the Director was entitled to draw the inference he does at paragraph 105 of the decision, that: “The continued presence of the Independent on the Aberdeen market from 1996 to 2000, despite Aberdeen Journals’ vigorous defence of its position, suggests that Mr Barwell continued to view a share of the advertising revenues generated by Aberdeen Journals’ newspapers, and the Evening Express in particular, as a valuable and attainable prize.” 59

[187]Further support for the Director’s conclusion is found in the note of a meeting between Mr Barwell, Mr Scott and Mr Blair of Aberdeen Journals of 5 August 1999, cited at paragraph 125 of the decision. That note states:

“[Mr. Barwell] now believes that particula rly with the H&P and the Independent both having a distribution of 125,000 each that the response from the Frees is better than the Evening Express. He saw the P&J as being unassailable but the Evening Express was vulnerable and the Herald & Post was irrelevant.”

(emphasis added by the Director).

[188]That note in our view confirms that throughout the period up to at least August 1999 Mr Barwell saw the Evening Express as one of his main targets. We see no reason to think that that had changed in March 2000, a matter to which we revert below. The fact that the note of 5 August 1999 records Mr Barwell’s view that the Press & Journal was “unassailable”, whereas in the document “The Scottish Opportunity – Aberdeen” the Press & Journal was seen as a potential target by Mr Alder, does not in our view weaken the Director’s case. Mr Barwell’s conclusion, reached after three years’ experience of the market, that the Press & Journal was not, after all, vulnerable, does not alter the fact that the Evening Express was still seen as vulnerable to the Independent in the autumn of 1999, as it had been in late 1995 and early 1996.

[189]The call logs of the Independent from March 2000 referred to in paragraph 117 of the decision constitute further evidence of direct competitive activity between the Independent and the Evening Express during that month. We think it unlikely that the sales staff of the Independent would be contacting advertisers whose business they did not think they had any hope of securing. Mr Robins’ letter of 8 March 2000

[190]Aberdeen Journals seeks to undermine the foregoing evidence by citing from a letter to the OFT from Mr Robins, Managing Director of Aberdeen Independent on 8 March 2000:

“The relevant product market that the Herald & Post competes in is the local newspaper market, and more specifically the local free newspaper market. We are their only direct competitor within both these markets. … Both the other newspapers within Aberdeen are daily, paid-for titles owned by Aberdeen Journals.”

[191]Although not perhaps crystal clear, Mr Robins there refers to “the local newspaper market” and “the local free newspaper market” and states that the Independent is the only competitor of Aberdeen Journals in “ both these markets” (emphasis added). Thus, although Mr Robins 60 sees “a free newspaper market” in which the Independent is competing, he also identifies a “local newspaper market” in which Mr Robins apparently includes both Aberdeen Journals’ paid-for titles. It seems to us that in this note Mr Robins was intending to include the Evening Express as a competitor of the Independent . The OFT meeting of 25 April 2002

[192]Aberdeen Journals also refers to an OFT meeting note of 25 April 2002 where it is reported:

“KB [Mr Barwell] questioned if the OFT had addressed the complements/ substitutes issue. He noted that from Aberdeen Journals’ perspective the Herald & Post and the Evening Express were complements and not competitors, due to the low quality of the Herald & Post . AD pointed out that this contradicted his earlier statements [?].”

[193]We for our part see no contradiction here. Mr Barwell is not here dealing with the question as to whether the Independent is targeting the Evening Express , but stating how he thinks Aberdeen Journals sees the relationship between the Evening Express and the Herald & Post which they maintain in Aberdeen under common ownership. As we illustrate below, the evidence is that at least until after March 2000 the Herald & Post was not seen by Aberdeen Journals as a free-standing product expected to make a normal return on investment in its own right, but primarily as a defensive tool to be used to prevent or hinder competitive entry into the Aberdeen marketplace. In 1996, the low-quality 20-page Herald & Post , run on a break-even basis, would not have been seen by Aberdeen Journals as a competitor to the higher quality and profitable Evening Express . In the following few years the Herald & Post was run artificially at a loss, not as a competitor to the Evening Express , but in order to protect the Evenin g Express from the Independent . Aberdeen Journals thus positioned its two titles in the market so as to suit its overall strategy. In our view those facts do not undermine the evidence before the Tribunal that both Aberdeen Independent and Aberdeen Journals took the view that the Independent’s competitive target was primarily the Evening Express rather than the Herald & Post . Was the Independent launched as a “fireship”, primarily in order to force a sale to Aberdeen Journals?

[194]Aberdeen Journals argues that the Independent was an “inefficient market entrant” and a “fireship” intended to wreak havoc rather than making a normal profit, with the primary aim of forcing Aberdeen Journals to buy out Mr Barwell. The possible relevance of this argument is to lay the foundation for the contention that the inferences from the documents relied on by the Director cannot be drawn, because the Independent had in some way created an 61 “artificial” situation in the Aberdeen marketplace, with the ulterior motive of engineering a sale to Aberdeen Journals.

[195]Aberdeen Journals relies, for this assertion, on various comments in Mr Scott’s letter of 10 February 2000 to the Director , and i n Appendix 2 to that letter, which allege(i) that Mr Barwell had a history of starting free newspapers and then selling the titles profitably;

(ii) that the Independent was targeting the Evening Express ; and

(iii) that the Independent was not viable at the distribution levels it was achieving with its particular cost base (see generally Appendix 2 to the letter of 10 February 2000). According to Aberdeen Journals, misleading statements were made by the Independent about the circulation figures of the Evening Express and other matters. Aberdeen Journals further relies on a memo of 21 December 1998, and a witness statement of Mr Pelosi dated 28 January 2003, which allegedly indicate Mr Barwell’s interest in selling the Independent to the Northcliffe Group. Lastly, Aberdeen Journals challenges the reliability of Mr Barwell’s evidence, on the grounds that he made a confidentiality agreement with Northcliffe on 19 April 2002 with a view to selling the Independent , and misled the OFT about this at a meeting on 25 April 2002.

[196]A number of these allegations are only distantly related to the issue which we are presently considering, namely the definition of the relevant product market for the purposes of the Chapter II prohibition. In our view, however, Aberdeen Journals’ assertions are unfounded.

[197]As to the various points made in the letter of 10 February 2000 and Appendix 2 to that letter, it is true that Mr Barwell is a successful newspaper proprietor who has been active in establishing free newspapers, many of which he sold to the Thomson organisation in 1988 (see paragraph 7 above). However, we have no evidence that Mr Barwell’s earlier activities were conducted on anything other than a normal commercial basis.

[198]Aberdeen Journals’ perception that Mr Barwell’s strategy was to undermine the Evening Express (paragraph 2(a) of Appendix 2 to the letter of 10 February 2000) does not show that the Independent was unviable, but strongly corroborates the Director’s view that the Independent was primarily competing with the Evening Express , not just with the Herald & Post .

[199]Indeed, the allegation that the Independent was unviable appears to rest almost entirely on Aberdeen Journals’ unsupported view, expressed in paragraph 2(b) of Appendix 2 to the letter of 10 February 2000, that a distribution figure of 120,000 copies represented an 62 unsustainably high cost base, and that the Independent ’s cost base in terms of staff was high compared with that of the Herald & Post .

[200]As to the distribution figures, we note that the Independent reduced its circulation from around 125,000 copies to 107,000 6 copies a week on 14 February 2000. The Independent thereby reverted to a distribution level only slightly less than the level prior to October 1999, which was 111,000 copies a week (Mr Scott’s witness statement of 1 February 2001). Given that the distribution of a newspaper such as the Independent can be varied at short notice, up or down, we do not think that a high distribution figure in a particular period is an indicator that the Independent was not a viable proposition over the longer term.

[201]In any event, Aberdeen Journals has advanced little more than unparticularised assertions as to the “unviability” of the Independent , with no underlying calculations, even hypothetical ones, to support its case. As against that, we have the solid fact that, seven years after its launch, the Independent remains in the market, and is, we are told, now trading profitably. We think it highly unlikely that Mr Barwell would have been prepared to sustain the Independent ’s losses after launch had he not believed that the newspaper was, in normal competitive conditions, a viable business proposition. Had the Independent not been a viable business proposition, it seems to us that the Independent would not have survived in the way that it has.

[202]Moreover, claims made by Aberdeen Journals as to the viability of the Independent in the early part of March 2000 have to be seen in the light of the fact that, by that date, the Independent ’s advertising rates had long been artificially forced down by the policy of pricing below cost followed by the Herald & Post over the previous four years. In the present case, it is undisputed that the Independent entered the market at rates above those of the Herald & Post , and that it was the Herald & Post that then reduced its prices to below cost, forcing the Independent to follow. We have no reason to suppose that the Independent would not have attained profitability in the period 1996 to 2000 had the Herald & Post not acted as it did. As Mr Davidson puts it in the letter of 10 February 2000, at paragraph 2.4, Aberdeen Journals believed that the Independent ’s actions in targeting the Evening Express , “ coupled with the advertising rates charged by the Independent ” (emphasis added) was inherently unprofitable . Thus, according to Mr Davidson, it was the combination of low rate s and targeting that made the Independent inherently unprofitable, in his view. That approach, 6 According to the RBB reports, this figure was 101,000. The figure of 107,000 is used in Aberdeen Journals’ application. In our view the difference is not material. 63 however, entirely overlooks the fact that the rates charged by the Independent were low precisely because of the actions of the Herald & Post .

[203]In our view, the fact that, in the event, the Independent made losses for the first few years of its existence is not due to the fact that the Independent was an “inefficient market entrant” or “fireship”, but rather to the reaction of Aberdeen Journals in drastically reducing the advertising rates of the Herald & Post to levels far below cost.

[204]Indeed, if further proof were necessary, the fact that the Independent was launched as a viable business proposition is solidly supported by the contemporaneous study “The Scottish Opportunity – Aberdeen”, already cited above. That study appears to us to be a sensible business appraisal which shows, with supporting calculations, that the Independent was launched with the intention of making a healthy profit.

[205]That is further corroborated by Mr Ezzat’s description of the background in his review of January 2000:

“Keith Barwell launched the Aberdeen Independent in March 1996. It is a free weekly newspaper distributed in the Greater Aberdeen area. Barwell believed that a good quality, editorially led, free newspaper in Aberdeen with a circulation of 90,000 could compete with the Evening Express and within 5 years would have turnover of some £5m and operating profit of £1m.”

[206]As to Aberdeen Journals’ suggestion that Mr Barwell’s real motive was all along to sell out to Northcliffe, for the reasons set out above we find on the evidence before us that the Independent was launched on a commercial basis. In those circumstances, in our view, whether or not Mr Barwell in fact contemplated selling the Independent at one or another point in time is not relevant to the issue of relevant product market.

[207]In any event, the suggestions in various Northcliffe documents (e.g. the note of the meeting of 5 August 1999, memos of 6 December 1999 and 5 January 2000, and Mr Ezzat’s review of January 2000) that one tactic Northcliffe could follow would be to buy out Mr Barwell, appear to come from Northcliffe, and not from Mr Barwell.

[208]A memo of 21 December 1998 from Mr Scott of Aberdeen Journals to Mr Davidson of Northcliffe indicates that Mr Scott had asked an intermediary to speak to Mr Barwell in September 1998. According to that note the intermediary said :

“KB felt it was a window of opportunity once Northcliffe had purchased Aberdeen Journals to open a freesheet, cause disruption, and sell it to Northcliffe. 64 KB also indicated that he was willing to sell come March/April when he had built a marketplace, mainly concentrating on Property and Motors.”

The memo of 21 December 1998 also suggests that there was a later contact in December 1998 from which it appears that one intermediary informed another intermediary that “KB was interested in selling in mid-January …”.

[209]It is difficult to place much weight on this document which consists of a report from one person to another about what a third party alleges Mr Barwell to have said. In our view, the important points to note are, first, that the initial approach came from Aberdeen Journals and, secondly, that nothing transpired.

[210]Mr Pelosi’s witness statement of 28 January 2003 is to the effect that in 1999 a Mr Harkness contacted Mr Barwell as an intermediary from Northcliffe on the subject of selling the Independent and that Mr Barwell supplied some information. Again, however, nothing came of this.

[211]This sporadic and indirect evidence, even if taken at face value, does not in our view support the conclusion that Mr Barwell’s strategy was to engineer a sale to Northcliffe. All the approaches are initiated by intermediaries apparently acting for Northcliffe, and nothing is pursued. Indeed, the meeting report of 5 August 1999 states:

“He [Mr Barwell] stressed he was in it for the long term and if it took 10 to 15 years he would do that. He did say that he had been in this position many times before having owned 33 free newspapers at various times and saw no point in trying to ‘bullshit the opposition’. He stated that he had two offers to buy in the past … but had refused them both. He again repeated what he had previously said to AB that he would be very unlike ly to sell to Northcliffe or to sell at all in the next 5 to 10 years.”

That evidence is quite inconsistent with Aberdeen Journals’ contentions.

[212]It thus seems to us impossible to infer from all the evidence before the Tribunal that the commercial strategy of the Independent was in some way distorted by a desire on Mr Barwell’s part to force a sale. In any event, such a strategy would seem to us to be highly improbable. Few businessmen would be likely to risk investing substantial sums in an unviable product simply in the hope that a competitor might one day buy them out: if the product is unviable, why should the competitor do so? 65

[213]It was suggested, at the hearing on 29 January 2003 that we should not rely on Mr Barwell’s evidence, because it was reportedly said at a meeting with the OFT on 25 April 2002: “AD [the OFT case officer] asked KB if he was approached by Aberdeen Journals offering to buy the Independent. KB responded that he had not been directly approached,” whereas on 19 April 2002 Mr Barwell ha d entered into a confidentiality agreement with Northcliffe with a view to entering into discussions about a possible sale of the Independent to Northcliffe.

[214]It is regrettable that this allegation was launched at the hearing on 29 January 2003, apparently without prior warning. Northcliffe, although not necessarily its legal team, had been in possession of the confidentiality agreement relied on for several months. Normally the Tribunal would not permit new matters to be raised at such a late stage.

[215]Mr Barwell tells us in his third witness statement of 29 January 2003 that, in January 2002 he was told by his doctors to cut down his commitments. In those circumstances, on 24 January 2002, he initiated contacts with several parties with a view to exploring a sale of the Independent . It was in those circumstances that the confidentiality agreement of 19 April 2002 came to be made.

[216]We have no reason to doubt Mr Barwell’s explanation. In any event, it does not appear to us that Mr Barwell’s decision , i n early 2002, to explore the possibility of selling the Independent to Northcliffe, has any bearing on the issue with which we are concerned, namely that of the relevant product market during the period 1996 to 2000 and, in particular, March 2000.

[217]As regards the note of the meeting of 25 April 2002, that note, even if accurate, does not seem to us to be relevant to the issue of relevant product market. In any event, the evidence suggests that Northcliffe never did make a direct approach to Mr Barwell seeking to buy the Independent , which is in accordance with what is stated in the note of 25 April 2002. We regard the points taken by Aberdeen Journals on the basis of the note of the meeting of 25 April 2002 as unwarranted.

[218]Finally, as to the suggestion that the position of the Independent could only be sustained on the basis of misleading claims as to circulation and other matters, it may be that on occasion Aberdeen Independent was found to have made incorrect or exaggerated claims. Aberdeen Independent tells us that the Evening Express has also been criticised in this regard by the relevant authorities. We do not find it necessary to make findings about these points. In our 66 view, it is not credible to suppose that the advertising business obtained by the Independent over four years is due to any material extent to exaggerated claims about circulation, rather than to the intrinsic merits of the Independent ’s service to advertisers. On the contrary, the fact that, as the material before the Tribunal shows, the Independent regularly promoted itself to advertisers as a rival to the Evening Express , seems to us to be further corroboration that the two newspapers were in competition.

[219]We conclude that the matters relied on by the Director at paragraphs 102 to 106 of th e Decision provide strong support for the Director’s conclusion as to relevant product market. C. The commercial strategy of Aberdeen Journals The Herald & Post as a “fighting title”

[220]At paragraphs 107 to 109 of the decision, the Director relies on the commercial strategy of Aberdeen Journals in response to the launch of the Independent as further evidence that the Independent was at all material times an actual or potential competitor of the Evening Express . The Director states:

“107. The strategy of Aberdeen Journals appears to have been to maintain the Herald & Post as a low quality, low volume publication, which posed no threat to the revenues of its more profitable titles, at least until it was faced by an aggressive competitor for those revenues. This interpretation is supported by the statements made by Aberdeen Journals and Northcliffe management, cited in the table following paragraph 181 below, that the Herald & Post was viewed as ‘ a tactical tool in the company’s armoury ’, apparently retained by the company as a low cost, ‘defensive free’ newspaper, ready to be activated as a competitive newspaper at short notice to defend the revenues of Aberdeen Journals’ more lucrative paid-for titles. According to those statements, such a strategy, if successful in expelling the new entrant, would have been followed by a ‘deactivation’ of the title and its reversion to a lower, less competitive status. 108. As noted above, Aberdeen Journals’ decision to cut advertising rates and increase pagination in the Herald & Post immediately after the launch of the Independent , and to continue that strategy for an extended period, as demonstrated by Graphs 1 and 2 of Annex 2, can only be viewed as rational as a means of protecting the Evening Express ’s considerable revenues, rather than the Herald & Post ’s much lower revenues. 109. The clear and consistent strategy of Aberdeen Journals and Northcliffe, namely to respond to the Independent ’s entry to the Aberdeen market by funding the Herald & Post to increase its quality, distribution and pagination to a level far above that which could have been justified by the newspaper’s own revenues, can therefore only be understood as an attempt to use the Herald & Post as a ‘fighting title’ to defend the lucrative business of the Evening Express from erosion by the 67 Independent . As a result, this conduct itself provides further compelling evidence supporting the Director’s conclusion that, from 1996 until at least March 2000, all three newspapers were present on the same product market.”

[221]The Director relies in particular on the memoranda set out in paragraph 181 of the decision and in particular the memorandum of Mr Davidson, Managing Director of Northcliffe, to Mr Scott, of 12 July 1996:

“You view the Herald & Post as a tactical tool in the company’s armoury. Barwell’s move to Aberdeen has caused you to increase your efforts on this and if and when he goes away you will leave a three month gap between that happy event and running it down again’ … ‘Next year’s [budget for the Herald & Post ] would include the £500,000 investment we are making against Barwell.”

[222]The Director also relies on the memorandum of Mr Pelosi, Deputy Managing Director of Northcliffe, dated 29 July 1998 to Mr Scott which states:

“The closure of the Aberdeen Independent would allow you to reduce gradually investment in the Herald & Post .”

[223]It is apparent from the table at paragraph 171 and Annex 2, Graph 4 of the decision that prior to the launch of the Independent , the Herald & Post was making a contribution (that is to say covering the costs allocated to it in its management accounts) of between £11,000 and £27,000 per month. As a result of the combined effect of a reduction in its advertising rates, an increase in pagination, and some increase in distribution, the Herald & Post then moved into “loss” (i.e. its contribution was negative), as follows: June 1996 to September 1998 — “losses” between £20,000 and £46,000 per month October 1998 to September 1999 — “losses” between £54,000 and £76,000 per month October 1999 to 29 March 2000 — “losses” between £33,000 and £52,000 per month

[224]On the above basis, by our calculations the accumulated negative contribution of the Herald & Post in the period from June 1996 to March 2000 is of the order of £1.9 million. If costs not attributed to the Herald & Post in the management accounts are taken into account the accumulated negative contribution of the Herald & Post would have been much higher.

[225]We have no hesitation in accepting the inference drawn by the Director at paragraphs 107 and 109 of the decision that the willingness of Aberdeen Journals to accept losses on this scale 68 over a period of four years as regards the Herald & Post cannot be explicable by a desire simply to protect the modest contribution Aberdeen Journals’ overheads and profits made by that newspaper prior to June 1996. The negative contribution incurred on the Herald & Post month after month in our view is only explicable by the desire of Aberdeen Journals to protect the revenues of its paid-for titles, and notably the Evening Express , by seeking to expel the Independent from the Aberdeen market place.

[226]We observe, finally, that although Aberdeen Journals has taken many points in this appeal, including many small points of detail, there is no witness statement from anyone from Northcliffe or Aberdeen Journals which seeks to rebut the allegation, which is central to the Director’s case, that the Herald & Post was all along being used as a “fighting title” or “defensive free” to protect the Evening Express from the Independent . The evidence that that was, in fact, the case, is in our view overwhelming. The evidence about bundling

[227]The Director further relies, in support of that conclusion, on the documents referred to at paragraphs 110 to 114 of the decision, submitted to the Director in the context of an earlier investigation. Those documents (relating to Cornhill Car Sales, 21 July 1999, Bruce & Partners, 25 November 1998, and Reg Vardy plc , 1 May 1996) show Aberdeen Journals offering higher discounts in the Evening Express , or better terms in the Press & Journal , Evening Express and/or Herald & Post , to customers prepared to cease advertising in the Independent . Again, in our view those documents are further evidence that the Evening Express was in competition with the Independent .

[228]Aberdeen Journals’ contention that such “bundling” and exclusionary practices had ceased by March 2000 (paragraph 115), presumably after the undertakings to that effect that Aberdeen Journals was required to give to the Director in December 1999, does not in our view alter the inference that Aberdeen Journals saw the Independent as a competitor of the Evening Express . Whether the situation was materially different in March 2000 is a matter we revert to specifically below. Mr Farquharson’s evidence

[229]Lastly on this part of the case the Director relies on the witness statement dated 1 May 2002 of Mr Farquharson, who was from 1996 to 1998 the display advertising sales manager of the 69 Evening Express , which is cited at paragraph 118 of the decision. Mr Farquharson now works for the Independent .

[230]At paragraphs 3 to 5 of his witness statement he says this:

“3. After the inauguration of the “ Aberdeen & District Independent” (“the Independent ”) I was involved in weekly meetings with Alan Scott, th e Managing Director of Aberdeen Journals Limited (“Journals”), together with the Sales Director, Classified Advertisements Manager and the National Sales Manager. We all met together to discuss the “ Independent ”

. The meeting had one purpose only which was to discuss any advertising which appeared in the “ Independent ” and the reasons why those advertisements were still appearing there and what the staff at Journals were doing to prevent it happening. 4. Whilst I was not directly involved as a Manager of staff at the “ Herald & Post ”, I was the Display Ads Manager for the “ Evening Express ” and “ Press & Journal ” and as such was able to arrange deals in relation to these two newspapers for the customers of the “ Independent ”. One of the sales team from the “ Evening Express ”, for example, would make contact with a customer of the “ Independent ” and offer them a deal in relation to a rate for their business if they transferred across to the “ Evening Express ”. Basically, the staff were free to do whatever deal they wanted to, in relation to advertising in the “ Herald & Post ”. The idea was to protect the “ Evening Express ” from the “ Independent ”. Generally, in terms of doing a deal with those customers who wished to transfer, it was the profit of the “ Herald & Post ” which was sacrificed and if possible the rates were kept up in the “ Evening Express ”. 5. I can say that I hated the meetings which we had with Alan Scott which were often ill tempered affairs. Although I was responsible for display ads, this is a less critical area in terms of advertising revenue than the classified advertisements, which were dealt with by Donna Henderson. As part of the unofficial group charged with removing the “ Independent ” from the Aberdeen market, I understood what the strategy was.”

[231]Although Aberdeen Journals described Mr Farquharson as “the classic disgruntled employee” there was no application by Aberdeen Journals to cross-examine him. Mr Farquharson’s evidence in our view further corroborates the Director’s conclusion that the Evening Express and the Independent were in competition for customers in the Aberdeen area.

[232]Again, in our view the evidence relied on by the Director at paragraphs 107 to 119 of the decision is strong support for the Director’s conclusion on relevant product market. That evidence is consistent with Aberdeen Journals’ defending, or seeking to re-establish, its monopoly on local newspaper advertising in Aberdeen. It would not, in our view, have reacted in that way had the Independent not been a credible competitor, actually or potentially, of the Evening Express . 70 D. The further documentary evidence supplied by Aberdeen Journals

[233]The Director further relies at paragraphs 120 to 130 of the decision on material contained in Aberdeen Journals’ letter of 10 February 2000, Appendix 2 to that letter, a memorandum of 21 May 1999, the meeting note of 5 August 1999, and Mr Ezzat’s review of January 2000. Aberdeen Journals’ letter to the Director of 10 February 2000

[234]We cite the passages from this letter relied on in the decision:

“[2.4] … first the Independent was launched as a title specifically targeted at our paid-for Evening Express title and its distribution area was focused squarely on the principal circulation area of our evening paper... second, we needed to respond to the launch of the Independent by lowering the advertising rates of the Herald & Post to what advertisers told us was necessary to enable us to retain their business. We had and have no wish to exit this layer of the advertising market and to do so would have made the Evening Express vulnerable [emphasis added]. [2.6] The Aberdeen market has now seen, for a period of four years, an effective war of attrition between our titles [emphasis added] and the Independent . [3.1] We currently face a serious dilemma the Independent poses a real threat to the advertising revenues of both our evening title [i.e. the Evening Express ] and the Herald & Post ”

[emphasis added]. If matching this competition is indeed to be characterised as predatory the only obvious solution would be for us forthwith to increase the rates for advertising in the Herald & Post (and possibly to cut back on our distribution area even further) so as to enable us to cover our costs. If we do so, Mr Barwell’s demonstrable ability to continue to fund his loss-making title in the long term poses a real commercial threat to the future both of our free and evening titles.”

[235]This letter is written, apparently on advice, as a formal response to an allegation of predatory pricing made to the Director by Aberdeen Independent. As we have already stated in our first judgment (at [103] and [104]), documents emanating from the undertaking concerned showing how that undertaking saw the market, and the commercial strategy it had adopted in that market, may be decisive evidence of what the market is. (See also, as to the importance of contemporary documents in lieu of abstract economic analysis, the judgment in Napp , cited above, at [251] and [252]).

[236]Mr Scott’s letter of 10 February 2000 seems on its face to show that, in the view of Aberdeen Journals, the Evening Express was vulnerable to competition from the Independent . The letter itself refers to the fact that the Independent was targeted specifically against the Evening Express and that the distribution area of the Independent “was focused squarely on the principal circulation area of our evening paper”. In at least three places the letter of 71 10 February 2000 seems to describe the contemporary situation in February 2000 e.g. “The Aberdeen market has now seen, for a period of four years, an effective war of attrition between our titles (plural) and the Independent”; “The Independent poses a real threat to the advertising revenues of both our evening title and the Herald & Post ”; “Mr Barwell’s [strategy] poses a real commercial threat to the future both of our free and evening titles”.

[237]As we said in our first judgment (at [161]) Aberdeen Journals’ letter of 10 February 2000 appears to us on a fair reading to support the relevant market contended for by the Director. Appendix 2 to the letter of 10 February 2000

[238]The Director relies on the following statements from Appendix 2 to the letter of 10 February 2000:

“On several occasions Barwell has stated his aim to close the Evening Express . He sees it as a vulnerable target overshadowed by the Press & Journal . Support for the validity of this statement can be seen throughout the Independent ’s business activities. Within 12 months of launching the Independent ’s distribution area matched that of the Evening Express ’s core area (extending way beyond the Herald & Post ’s city centre area into the commuter belt). Throughout the Independent ’s existence in Aberdeen it has sought to undermine the Evening Express in its promotional literature within the newspaper and within its external promotional materials… For example the marketing of the launch of his title included several comparisons with the Evening Express … Similarly he tried to compare the circulation performance of his Free newspaper with that of our Evening Express (a paid for publication) … The Independent ’s strategy has been to target itself at the Evening Express and its customer base. In order for that to be credible with advertisers it has sought to replicate the EE ’s core area. It therefore has expanded its distribution area until it mirrors the EE ’s circulation area i.e. expanding beyond Aberdeen City to include outlying commuter towns. This has resulted in a distribution in excess of 120,000 and a cost base well in excess of £1.5 million p.a. It is very difficult to understand how the Independent can build a viable long-term business with this cost base. It does however become a substantial thorn in the EE’s business in the shorter term which is Barwell’s primary objective [emphasis added] … The first time AJL was able to get some form of verification of the rates being charged by the Independent was on the publication of their first year’s financial results. AJL estimated that, on the basis that it had a reasonably good estimate of the Independent ’s costs, their average yield was no more than £1.50/sccm. This confirmed that AJL had to remain competitively priced with the Independent even though it was having a significantly detrimental effect on AJL’s advertising volumes with the Evening Express [emphasis added]. Competition intensified even further when the Independent gained some substantial new Property business from some of the major solicitors in Aberdeen. 72 Previously, they had not wished to use Free newspapers to advertise Property for sale. In order to defend its business that was with the P&J and EE , AJL offered its major Property clients a package whereby they could appear in the Herald & Post if they continued to advertise in the P&J and/or the EE ”

[emphasis added]. Other documents — The memorandum of 21 May 1999

[239]In this memorandum Mr Pelosi, deputy managing director of Northcliffe, wrote to Mr Davidson, Northcliffe’s Managing Director, as follows: “The approach taken to the Independent has had a serious adverse impact on yields for property and motors. We have submitted to advertisers’ threats of switching to the Independent by granting higher discounts. … We must accept, however reluctantly, that the Independent competes with us in the market … We will not drive this business out of the market because its wealthy proprietor can support its trading losses. … Therefore, we must stop the damaging process of ever increasing discounts. Instead we should return to selling the benefits of advertising in the Press & Journal and the Express .” — The meeting note of 5 August 1999

[240]The meeting note of 5 August 1999 states:

“[Mr Barwell] believes now that particularly with the H&P and the Independent both having a distribution of 125,000 each that the response from the Frees is better than the Evening Express . He saw the P&J as being unassailable but the Evening Express was vulnerable and the Herald & Post was irrelevant.”

(emphasis added). — Mr Ezzat’s Review of January 2000

[241]In a document entitled ‘ Review of Aberdeen Independent ’ prepared by Taymour Ezzat (the London Financial Controller of Northcliffe) in January 2000 and also supplied to the Office with Aberdeen Journals’ submission of 10 February 2000, Mr Ezzat stated: “Keith Barwell launched the Aberdeen Independent in March 1996. It is a free weekly newspaper distributed in the Greater Aberdeen area. Barwell believed that a good quality, editorially led, free newspaper in Aberdeen with a circulation of 90,000 could compete with the Evening Express [emphasis added] and within 5 years would have turnover of some £5 m and operating profit of £1 m. … Interestingly, the Independent has not reduced its circulation levels following the lead from the Herald & Post . One argument for them not following suit is that Barwell is targeting the Evening Express and requires these high coverage levels to achieve the required advertising response. … In terms of the Herald and Post increasing its distribution levels there are two schools of thought. The first school believes that Barwell will maintain his level at 125,000 regardless of the Herald and Post levels as he is targeting the Express 73 and therefore Aberdeen Journals can run at lower distribution levels to contain their losses. … Although Aberdeen Journals have been successful in restricting the battle to the two frees by relaunching their free title the Evening Express has suffered . I have summarised below the revenues, volumes yields and % changes period on period for the Evening Express for the last two years: [TABLE OF FIGURES OMITTED] This table cle arly shows the effect of the battle between the Independent and Herald and Post on the Evening Express . Average weekly revenues have declined from £95k per week to £75k (a decline of 21%), volumes from 17k cms per week to 15k cms (a decline 12%) and a decline in yields from £5.60 per cm to £5.10 per cm (a decline of 9%). Half of the decline can be attributed to the fall in Sits Vac revenues and is not related to the Independent . However a considerable portion of the remainder is due to reduced retail/property revenues which can be partly attributed to the Independent (emphasis added). … The Independent ’s ability to break even/make a profit is very dependent on the Aberdeen Journals tactics with the Herald and Post and Evening Express ” [ emphasis added].

[242]As regards Mr Ezzat’s review, Aberdeen Journals argues, on the basis of later work apparently carried out in 2002 by RBB Economics, that in fact almost the whole of the decline in revenues of the Evening Express between October 1997 and December 1999 was accounted for by a decline in “sits vac” revenues which Mr Ezzat claimed was “not related to the Independent ”. The Director argues that that still leaves an appreciable proportion of the decline as partly attributable to the Independent . Aberdeen Journals says that what is left is de minimis .

[243]We think it somewhat improbable that Mr Ezzat’s contemporary view that the Independent had caused the Evening Express a loss of business was in fact mistaken, confirmed as it is by Mr Pelosi’s memorandum of 21 May 1999 (“The approach taken to the Independent has had a serious adverse impact on yields for property and motors”) and Appendix 2 to the letter of 10 February 2000 (“… [the Independent ] was having a significantly detrimental effect on AJL’s advertising volumes with the Evening Express ”). However, in our view the important point is that Mr Pelosi, Mr Scott and Mr Ezzat all perceived the Independent to be damaging the Evening Express and acted accordingly. It does appear, moreover, that by 1999, yields in the Evening Express were falling, despite Aberdeen Journals’ efforts to protect them: see paragraph 281 below.

[244]Even if the precise magnitude of the Evening Express ’s loss of business is debateable Aberdeen Journals has not contested the remainder of Mr Ezzat’s memorandum, the general 74 tenor of which plainly supports the proposition that the Independent is competing with the Evening Express . We note in particular that Mr Ezzat also says:

“Many local advertisers have a vested interest in keeping the Independent as a competitor to the Aberdeen Journals titles and will not transfer or reduce their spend in the Independent . We will not be able to “close” the title but can maintain their loss position at some £500k p.a.”

[245]The clear inference from this statement is that advertisers in Aberdeen regarded the Independent as a competitor to, notably, the Evening Express . The statement also illustrates very clearly the tactic of Aberdeen Journals in forcing the Independent to sell at a loss. — Letter of 1 August 2000

[246]Finally, in Appendix 4 to the letter dated 1 August 2000 sent by Aberdeen Journals to the OFT, the commentary in relation to Maitlands of Fraserburgh states:

“The majority of the customer’s spend is with the two daily titles, principally the P&J . Terms have not changed with the customer in either of these titles. The customer has elected to move more of its spending onto these titles as a result of the H&P price increases ”

[emphasis added].

[247]In our view, read as a whole, the documents cited at paragraphs 120 to 129 of the decision provide further strong support for the conclusion reached by the Director. Taking the evidence as a whole, we agree with the Director’s conclusion at paragraph 130 of the decision that:

“(i) Aberdeen Journals and Northcliffe considered that the Herald & Post , Evening Express and Independent were direct competitors during the relevant period. Aberdeen Journals’ commercial strategy for meeting the challenge posed by the Independent was prompted by, and based on, this analysis of the relevant market; and (ii) Mr Barwell, the owner of the Independent , shared the same view of the market.”

E. Did the position change in March 2000?

[248]Aberdeen Journals argues that the conclusions to be drawn from the foregoing cannot be carried over to the period of infringement, namely March 2000. Many of the documents are purely historical. In any event, matters changed in March 2000. The Herald & Post was raising its rates, and changing its pagination and distribution. The Independent itself reduced its distribution from around 125,000 to 107,000 copies a week in mid-February 2000, which, according to Aberdeen Journals, shows that it was no longer targeting the Evening Express . 75

[249]In our view, as the Tribunal said in Napp , cited above, at [217], “it is relevant to take facts arising before 1 March 2000 into account for the purpose … of throwing light on facts and matters in issue on and after that date”. In our view the whole tenor of the facts and matters set out at length above is that the Independent was in competition with the Evening Express from the date of its launch in 1996 onwards. Indeed, certain of the documents such as the letter of 10 February 2000, Appendix 2 to that letter, and Mr Ezzat’s review of January 2000 are closely contemporaneous to March 2000. Those documents refer to competition between the Independent and the Evening Express in the present tense.

[250]We have therefore asked ourselves whether there is anything to suggest that the position regarding market definition had changed by March 2000.

[251]It is true that the Herald & Post reduced its distribution from about 123,000 copies a week to around 107,000 copies a week in October 1999. Distribution was later further reduced to 84,000 copies a week, but only with effect from the edition of 29 March 2000. Similarly pagination was reduced from around 300 pages a month to around 200 pages a month, but again only with effect from the edition of 29 March 2000. From October 1999 the Herald & Post had sought to raise advertising rates, but it does not appear that its average advertising rates were materially higher in March 2000 than in January or February 2000 (table following paragraph 50 of the Director’s defence in the first appeal and Annex 2, Graph 7 of the decision). Average revenue for the month of March 2000 was still below average variable costs.

[252]It does not appear to us that in March 2000 the Herald & Post had materially changed its character.

[253]In any event, the relevant question here is not so much whether the Herald & Post had changed its character, but whether in March 2000, the Independent was still an effective competitive constraint as regards the pricing policy followed by the Evening Express . In that regard, according to Mr Scott’s evidence, it appears that the only material change was that, in mid-February 2000, the Independent reduced its distribution to 107,000 copies from 120,000 copies, although that change was apparently not announced until 13 April 2000. The distribution of the Independent thereby apparently reverted to a level not far below its distribution level prior to October 1999 of 111,000 copies a week (Mr Scott’s witness statement of 1 February 2001, at p.7). 76

[254]We do not think that the change in distribution from 120,000 copies to 107,000 copies per week in mid-February 2000 shows that, in March 2000, the Independent was no longer a competitor of the Evening Express . As we have already pointed out, the level of distribution of a free newspaper may be changed relatively easily, up or down, from time to time. Aberdeen Journals has not contested Mr Barwell’s evidence that in March 2000 the Independent covered about 78 per cent of the area in which the Evening Express was distributed in March 2000. For advertisers wishing to reach customers within that area the Independent was in our view a viable alternative in March 2000.

[255]In our view there is no reason to suppose that the relevant product market in March 2000 was any different from that in January or February 2000, or indeed in any earlier period. F. Conclusion on the Director’s case

[256]At this stage of the analysis, we conclude that the case made by the Director in the decision is based on strong and compelling evidence to the effect that in March 2000 the relevant product market for the purposes of this case was the market for advertising in local newspapers in Aberdeen (paid-for and free). In the absence of any countervailing con-siderations that evidence would, in our view, suffice to establish the relevant market for which the Director contends. We now examine whether Aberdeen Journals’ further submissions are such as to raise a reasonable doubt as to the correctness of that conclusion. 3. THE CASE MADE BY ABERDEEN JOURNALS A. Criticism of the Director’s approach to economic evidence

[257]Aberdeen Journals criticises the Director for seeking to prove his case without economic evidence. In particular, says Aberdeen Journals,(i) the Director failed to carry out a survey of advertisers, as he could have done;

(ii) the inconclusive survey that the Director did carry out, if anything, supports Aberdeen Journals;

(iii) the Director’s statistical analysis was not properly carried out and, if anything, supports the case made by Aberdeen Journals (see paragraphs 59 to 61, and 70 to 71 above);

(iv) evidence about conduct is not a relia ble indicator of market definition when prices are distorted; and

(v) the absence of any evidence about switching, even when the disparity between prices in the Evening Express and the free newspapers were at their maximum extent, is a strong indication that those products are in different markets. 77

[258]We have already found, at paragraphs 126 to 127 above, that there is no hierarchy of evidence under the 1998 Act on such issues as market definition. It is for the Director to decide what evidence he considers is sufficient for his decision, and for the Tribunal to decide whether that evidence is sufficient or not.

[259]In this case the Director concedes, at paragraph 98 of the decision, that “the economic evidence available does not, in itself, provide sufficiently strong and compelling evidence” to prove that the Independent and the Evening Express were active on the same advertising market in March 2000. For that reason, says the Director he is entitled “to attach particular importance to other factors, namely the conduct and statements of the undertakings concerned, to establish the relevant product market”.

[260]At paragraphs 93 to 95 of the decision the Director explains why, in his view, economic evidence has its limitations in the circumstances of this case. He points out, in particular, that the so-called “hypothetical monopolist” or “SSNIP” 7 test (which measures whether a dominant supplier would be able to maintain prices above competitive levels without customers switching to substitute products) must be applied with caution where competition is already distorted. According to the Director, “[g] iven the conduct of the Herald & Post in maintaining prices of its advertising space below the level required to cover its average variable costs up to the end of March 2000, the scope for applying such a substitution analysis in this case is limited” (paragraph 94). The Director further points out that advertising prices are not transparent, that there is extensive discounting, that prices may be bundled across different newspapers, and that a customer may react to price rises by altering the balance of their advertising spending between different newspapers rather than switching outright (paragraphs 35 and 95). The Director further considers that the failure of the statistical analysis which he carried out to yield “reliable statistically significant econometric results” was largely due to these factors, and to a lack of sufficiently detailed data for advertising in the Independent for the period covered by that analysis (paragraph 96). However, the Director nonetheless provided to Aberdeen Journals, in a letter of 8 August 2002, an outline of the data used and the results of his statistical analysis.

[261]Aberdeen Journals and Professor Yamey do not contest in principle the Director’s explanation of the limitations on economic evidence in this case as set out in paragraphs 35 and 94 to 95 of the decision. Indeed Professor Yamey confirms that it is dangerous to draw conclusions about substitutability unless both the products concerned are being priced at a This test measures the perceived reactions of consumers to a Small but Significant Non-transitory Increase in Price. 78 competitive level and not at some distorted level (paragraphs 2.20 and 2.25 of Professor Yamey’s evidence).

[262]In our view paragraphs 35, 94, 95 and 96 of the decision provide a convincing explanation for the Director’s view, expressed at paragraph 98, that there are good reasons in the present case for not relying on economic evidence as, in itself, sufficient to establish the relevant market. In particular, the fact that market conditions were already distorted, means that extreme caution must be exercised when dealing with the presence or absence of switching patterns. Such evidence is not a reliable guide to what would occur in normal competitive conditions. We therefore accept the Director’s approach and do not consider that the absence of economic evidence in support of the Director’s definition of the relevant product market detracts from the force of the evidence that we have already analysed above.

[263]As to Aberdeen Journals’ submission that the Director should have carrie d out a survey of the kind carried out by the Competition Commission in Johnston Press/Trinity Mirror , cited above, the Director did, prior to the first Rule 14 notice, carry out a survey of advertisers in the Aberdeen area and found the results to be inconclusive. In our view, the Director was under no obligation to carry out a further survey if he considered he could prove his case by other means.

[264]As far as we are aware, the survey conducted by the Competition Commission in Johnston Press/Trinity Mirror was not addressing the particular situation that exists in this case, namely that of a monopoly proprietor of a paid-for daily newspaper (the Evening Express ) maintaining a “defensive free weekly” (the Herald & Post ) which is then activated in order to expel a new free weekly (the Independent ) perceived by that proprietor as a threat to the paid- for daily. In these specific circumstances we do not think that the Director needed specific survey evidence to prove that the Evening Express and the Independent were, in fact, in competition for the business of advertisers. Nor do we think that the results of any such further survey would necessarily have been reliable, because of the distorted market conditions prevailing in the Aberdeen area.

[265]As regards the survey which the Director did conduct, the Director obtained only 10 replies to some 40 enquiries sent out to advertisers in the Aberdeen area in March 2000. Those replies are at pages 832 to 844 of the bundle. Despite the submissions made to us at the hearing of the first appeal, we do not think anything in these generally extremely brief replies detracts from the weight of the evidence we have already considered above. Two of the replies tend in our view to support the Director, namely those from Willia m Wilson (“If Aberdeen 79 Journals raised their rates I’d complain bitterly and I would be inclined to look for alternatives. In Aberdeen City I’d look at the Aberdeen Independent .”) and Alan Grant (Grampian) Ltd (“If the Independent left the market it would certainly be a problem for us as their rates provide us with good exposure at realistic prices”). However, in our view the overall result of the survey of March 2000 is inconclusive, partly because of the small number and brevity of the replies received and partly because, as we have said, it is difficult to rely on survey evidence in distorted market conditions. Since the Director does not rely on this evidence, neither do we. On the other hand, having examined the replies received, we find nothing in that evidence which undermines the case which the Director makes in the decision.

[266]As regards the Director’s statistical exercise, referred to in paragraphs 96 and 97 of the decision, Professor Yamey suggests that the Director should have used invoice level data. The Director replies that the Independent informed him that invoice level data was not available. We have no reason to doubt that explanation. Moreover, an analysis based on invoice data would itself have been subject to conceptual difficulties (see paragraph 295 below). In any event, a criticism of the way the Director carried out a statistical exercise which the Director was not obliged to undertake and upon which he does not rely does not seem to us to be a relevant criticism for present purposes.

[267]In our view, the same applies to Professor Yamey’s suggestion, cautiously put forward, that the OFT may have carried out its analysis with a view to seeking confirmation for a preconceived opinion that paid-for and free newspapers were in the same market, and thus have been too ready to dismiss results showing the opposite conclusion as counter-intuitive, not credible or anomalous. We have no evidence to that effect. Since the Director does not rely on his statistical analysis to support his case, we do not find this criticism to be relevant in the present context. Given the wholly atypical situation prevailing in Aberdeen over the period for which data is available , in our view it is unsurprising that the Director apparently obtained somewhat bizarre statistical results.

[268]In our view, the only relevance of the Director’s statistical analysis is whether paragraphs 96 and 97 of the decision, and the letter of 8 August 2002, contain anything capable of undermining the Director’s conclusion on the issue of relevant market. In our view, that is not the case.

[269]First, we are not persuaded that the OFT’s results as explained in the letter of 8 August 2002 can reliably be taken to support Aberdeen Journals’ submissions. As Professor Yamey points 80 out, neither he nor the Tribunal is in a position actually to verify that the OFT’s methodology is soundly based. Aberdeen Journals has taken no procedural point on that. Nevertheless, the OFT’s letter of 8 August 2002 seems to us to contain a sufficiently full account of the procedures followed and results obtained to demonstrate that the analysis was conducted in a professional manner and that the OFT’s decision to discount the detailed results was warranted.

[270]Moreover, as the decision points out, and Professor Yamey confirms at paragraphs 2.20 and 2.25 of his commentary, the scope for a substitution analysis of this kind is severely limited where the prices of the products concerned are not being priced at normal levels. In addition, in the present case there are the problems of lack of transparency in prices, the bundling of Aberdeen Journals’ products, the changes in the advertising mix, and the gradual adaptation of the market to the entry of the Aberdeen Independent and the change in the character of the Herald & Post . All these factors lead us to doubt whether the Director’s attempted statistical analysis was in fact a useful exercise. In those circumstances, the evidence about the Director’s statistical analysis does not raise a doubt in our mind as to whether the Director’s relevant product market is correct.

[271]Professor Yamey nonetheless argues (at paragraph 2.30 to 2.35 of his commentary) that it is unsafe for the Director to rely only on the conduct of Aberdeen Journals to prove his case, without supporting economic analysis, since that conduct could itself have been driven by distorted market factors, notably if the Independent was priced at launch at well below the competitive level. However, this theoretical difficulty does not arise in the present case, since we have already found (at paragraphs 194 to 218 above) that the Independent was launched on a normal competitive basis with the intention of establishing itself as a viable competitor to the Evening Express . The fact that the Independent was forced after launch to reduce its prices was due to the reaction of Aberdeen Journals, not to any intention on the part of Aberdeen Independent to launch a “fireship”. In those circumstances, we see no reason why we should not give appropriate weight to the conduct of the parties in determining market definition in this case.

[272]Lastly on this part of the case, Aberdeen Journals argues generally that there is little or no evidence of switching by advertisers between the paid-for daily Evening Express and the free weekly Herald & Post and Independent . Reliance is placed in particular on Mr Scott’s evidence that when the Herald & Post raised its advertising rates after March 2000 there was little effect on the volumes of advertising in the paid-for titles. He also states that, even when rates in the Herald & Post and Independent were low, he “did not detect advertisers 81 switching volumes of advertising from the paid-for titles to the free titles as a result. If that had been the case, I would have been forced by the market to reduce advertising rates in the paid-for titles.” Aberdeen Journals suggests that if there was no switching when the gap between rates was at its highest, and if yields in the Evening Express did not fall, that is a strong indication that the products are in different markets.

[273]We consider that such arguments are fallacious, for the following reasons.

[274]First, it is common ground that throughout the period under consideration the competitive conditions in the market were distorted by the fact that the Herald & Post was being priced below average variable cost, with the Independent being thereby compelled to sell at loss if it wished to remain in the market. In our view, as Professor Yamey points out at paragraph 2.25 of his report, evidence about what occurs at atypical competitive prices cannot be relied on to show what products would be substitutable for each other in normal competitive conditions.

[275]To take, for example, Mr Scott’s argument that when the Herald & Post raised its rates after March 2000 there was little observed switching to the paid-for titles, we note that at that stage rates in the Herald & Post and the Independent were well below competitive levels. In the market circumstances then prevailing, it would be natural for customers faced with an increase in the Herald & Post to turn to the Independent , the nearest price rival, rather than to the Evening Express . But that is a result of the distorted conditions in the market. In our view in determining the relevant market, the correct question is not whether customers of the Herald & Post would be more likely to switch to the Independent than to the Evening Express , at the artificial price levels then prevailing, but whether the Independent would, actually or potentially, be a competitive constraint on the Evening Express in normal competitive conditions. The evidence already set out in this judgment points overwhelmingly to an affirmative answer to that question.

[276]To put the matter in more formal terms, in our view, when determining whether an undertaking has a dominant position “in a market” for the purposes of section 18(1) of the 1998 Act, the “market” to be taken into consideration is the market that would exist in normal competitive conditions, disregarding any distortive effects that the conduct of the dominant firm has itself created. If it were otherwise, there would always be a risk that the dominant firm could escape the Chapter II prohibition by artificially manipulating its prices so as to create the appearance of a “separate” market when in normal circumstances no such separation would exist. Indeed, that seems to us to be largely what has occurred in the 82 present case. Aberdeen Journals has distorted the market conditions in Aberdeen. In our view, it is unsafe to rely on what takes place in those distorted market conditions as evidence of the “separate market” for which Aberdeen Journals contends.

[277]There is, moreover, a second distorting factor in this case. For much of the period in question, at least up to December 1999, Aberdeen Journals seems to have been prepared to sell advertising on a “package” basis across its three titles or, at least, across the Evening Express and the Herald & Post . Thus, to take property as an example, it appears to be the case that to avoid property advertisers in the Evening Express switching to the Independent , Aberdeen Journals’ tactic was to offer those advertisers very low rates (or, according to Aberdeen Independent , no charge) for advertising in the Herald & Post . In other words, the policy pursued was not to reduce rates significantly for the Evening Express , but instead to offer potential customers of the Independent below-cost prices in order to induce them to advertise in the Herald & Post instead. In these circumstances, in our view one cannot reliably draw any conclusions about the absence of switching.

[278]As we see it, the situation in Aberdeen for most of the period in question was as follows. For those customers for whom a free title would not have been a close substitute for the Evening Express , Aberdeen Journals would not have been under pressure to reduce its advertising rates significantly. For those customers for whom a free title would have been a reasonable substitute, Aberdeen Journals’ response was to offer very low rates in the Herald & Post , rather than reducing the rates for the Evening Express . By this means, so it seems to us, Aberdeen Journals could seek to protect the revenues of the Evening Express while at the same time attacking the Independent by using the Herald & Post as a “fighting title”, thus having the best of both worlds.

[279]For these reasons, in our view the Aberdeen marketplace was not, as suggested by Aberdeen Journals, an “ideal laboratory” in which to study switching patterns in order to establish whether the Evening Express was in the same relevant market as the Independent and Herald & Post . Aberdeen Journals’ argument that, at the point of maximum difference between the respective prices of the Evening Express and the Independent , one would expect to see switching between the two, is in our view fallacious. Any lack of such switching is explicable because(i) for many of those customers for whom a free newspaper was a reasonable substitute for the Evening Express , Aberdeen Journals gave them an artificial incentive to switch to the Herald & Post , rather than the Independent ;

(ii) at the artificially low prices then prevailing, rate changes as between the Independent and the Herald & Post would have been more likely to induce switching between those two newspapers, rather than 83 between either of them and the Evening Express , and

(iii) the prices charged by the free newspapers for advertising may already have been so low that even substantial changes in the relative price charged for advertising in the Evening Express would not induce switching between that paper and the free papers. For those reasons, any lack of switching does not in our view negate the proposition that the Independent was actually or potentially in competition with the Evening Express .

[280]In any event, despite Aberdeen Journals’ tactics, there is evidence that the Independent did in fact act as a competitive constraint on yields in the Evening Express , as we have already pointed out. For example: Appendix 2 to the letter of 10 February 2000 states:

“[the Independent ] was having a significantly detrimental effect on AJL’s advertising volumes with the Evening Express … In order to defend its business that was with the P&J and EE , AJL offered its major Property clients a package whereby they could appear in the Herald & Post if they continued to advertise in the P&J and/or the EE .”

Mr Pelosi’s memorandum of 21 May 1999 states:

“The approach taken to the Independent has had a serious adverse impact on yields for property and motors. We have submitted to advertisers’ threats of switching to the Independent by granting higher discounts.”

Mr Ezzat’s memorandum of 18 January 2000 states:

“This table [of average weekly revenues, volumes and yields for the Evening Express ] clearly shows the effect of the battle between the Independent and the Herald & Post on the Evening Express .”

[281]Despite the dispute between the parties over Mr Ezzat’s attempted quantification of the impact of the Independent on the Evening Express , it seems to us unlikely that Aberdeen Journals’ and Northcliffe’s perception was mistaken. That is further supported by the figures at Annex 11 to the decision, which appear to show revenue, volumes and yields for the Evening Express for the 12 months to September 1999, and the following nine months, October 1999 to June 2000, as lower than for the 12 months to September 1997 and September 1998 respectively. A similar picture of a downward trend in yields is shown in Figure 1 of the first RBB report, making allowance for seasonal “spikes” around Christmas and the New Year.

[282]For those reasons we do not regard either the evidence about yields in the Evening Express , or any observed lack of switching between the Evening Express and the free titles, as support for Aberdeen Journals in the particular circumstances of this case. 84

[283]There is, finally, the self-evident point that, despite the approach favoured by Aberdeen Journals, the Independent in fact attracted very many advertisers in the Aberdeen area. In certain sections, such as property and motors, the Independent provided an alternative to the Evening Express where previously there was none. As already mentioned, and further illustrated below, many advertisers use both the Evening Express and a free newspaper, and some use all three titles. Whatever the arguments regarding “complements” and “substitutes”, common sense suggests that, for at least a significant number of advertisers there is likely to be a relative price at which advertising in the Independent is a viable alternative to advertising in the Evening Express . If the Independent was not there, Aberdeen Journals would have a monopoly. That fact is likely in principle to constrain the Evening Express from raising its prices. In the course of the lengthy arguments addressed to us in this appeal we have seen nothing to detract from that conclusion, and strong and compelling evidence to support it. B. The RBB reports

[284]The methodology of the RBB reports is to study the behaviour of those advertisers who advertised in the Evening Express in March 2000. In the first RBB report, RBB found that 61.2 per cent by volume of the advertisers advertising in the Evening Express in March 2000 did not advertise in a free newspaper during that month. Table 5 of the first RBB report sets out 29 advertisers who placed more than half a page in aggregate in more than one newspaper in March 2000. This table is as follows: 85 Table 5 Advertisers with Evening Express volume of over half a page who also use the free titles, Size and Distribution of Volume, March 2000 Share of Volume in Advertiser Evening Express Volume Evening Express Herald & Post Independent 1 1 1 1 Advertisers in all three papers Reg Vardy [ ■■■ ] 64% 15% 21% Barratt Homes [ ■■■ ] 42% 28% 30% Bruce & Partners [ ■■■ ] 22% 49% 29% Aberdeen City Council [ ■■■ ] 29% 53% 17% Aberdein Considine [ ■■■ ] 20% 36% 44% Advertisers in the Evening Express and the Herald & Post Arnold Clark Ltd [ ■■■ ] 81% 19% - Argos [ ■■■ ] 43% 57% - Murison Brothers [ ■■■ ] 86% 14% - Dee Carpets (Flooring) Ltd [ ■■■ ] 47% 53% - Kwik Fit [ ■■■ ] 50% 50% - Culter Car Centre [ ■■■ ] 38% 62% - Charles Phillips & Sons [ ■■■ ] 70% 30% - Codonas [ ■■■ ] 50% 50% - Sterling [ ■■■ ] 50% 50% - Aberdeen Joinery [ ■■■ ] 54% 46% - Pittodrie Car Stadium [ ■■■ ] 40% 60% - Aberdeen Exhibition [ ■■■ ] 89% 11% - Advertisers in the Evening Express and the Independent Anderson Cars Group [ ■■■ ] 39% - 61% John Clark Motor Group [ ■■■ ] 45% - 55% Scotia Properties [ ■■■ ] 61% - 39% Atlantic Telecom [ ■■■ ] 76% - 24% Terracotta [ ■■■ ] 49% - 51% Esslemont & Macintosh [ ■■■ ] 56% - 44% Town & Country Service Station [ ■■■ ] 77% - 23% Aberdeen Football Club Plc [ ■■■ ] 73% - 27% Alba Homes Ltd [ ■■■ ] 50% - 50% Calders [ ■■■ ] 85% - 15% Easy Jet [ ■■■ ] 52% - 48% Asda [ ... ] 44% - 56% Source : RBB analysis based on Aberdeen Journals’ data.

[285]RBB then studied the behaviour of the top 10 individual advertisers advertising in more than one newspaper in March 2000 over the period September 1999 to June 2000 to see whether the relative volumes of advertising placed by those advertisers in the different newspapers varied in response to price changes or changes in distribution (according to RBB, equivalent 86 to a price change) over that period. According to RBB, 6 out of the 10 advertisers representing 23.4 per cent of total advertising volumes in the Evening Express in March 2000, did not switch advertising in response to the price or distribution changes studied. RBB concludes that those advertisers view advertising in the free newspapers as complementary to, rather than substitutable for, advertising in the Evening Express . The position as regards the remaining four advertisers was inconclusive, according to RBB. From this RBB concludes that 84.6 per cent (61.2 per cent + 23.4 per cent) of advertisers advertising in the Evening Express in March 2000 did not regard the free newspapers as effective substitutes for the Evening Express .

[286]In the second RBB report, RBB studied whether those advertisers who had advertised only in the Evening Express in March 2000 had in fact advertised in another newspaper in the period between September 1999 and June 2000. According to RBB some 3,468 advertisers advertising only in the Evening Express i n March 2000 did not use a free title in the period September 1999 to June 2000 either. Those advertisers represent 46.4 per cent by volume of advertising in the Evening Express in March 2000. RBB then studied 50 advertisers who had used the Evening Express only in March 2000, but had used a free title (even if for only 1 cm of space) in the period September 1999 to June 2000. RBB concluded that in 28 cases (representing 9.3 per cent of volume) the advertisers’ response to changes in price or circulation was inconsistent with those advertisers regarding a free weekly title as an effective substitute for the Evening Express . In 22 cases (representing 2.7 per cent of volume) the result was inconclusive. RBB similarly studied in more detail a further 19 advertisers who had used the Evening Express and a free weekly title in March 2000, in addition to the 10 advertisers in this category studied in the first RBB report, making 29 advertisers in all in this category. According to RBB, 9 of these advertisers (representing 24.5 per cent of volume) behaved in a way that was not consistent with them viewing the free weeklies as a substitute for the Evening Express . The remaining 20 advertisers in this category (representing 8.7 per cent of volume) yielded results that were not determinative either way. Finally, RBB carried out the same exercise on a further group who had been omitted from the first report, and concluded that advertisers representing a further 0.9 per cent of Evening Express volume did not regard th e free newspapers as substitutes. From these analyses RBB concludes that advertisers representing some 81.1 per cent (46.4 per cent + 9.3 per cent + 24.5 per cent + 0.9 per cent) of advertising volume in the Evening Express in March 2000 did not regard the free weeklies as effective substitutes. 87

[287]The Director and Aberdeen Independent having pointed out that RBB had missed a number of advertisers who advertised in both the Evening Express and the Independent in March 2000, RBB stated in a revised summary of their analysis handed up at the hearing on 29 January 2003 that 42 per cent by volume of advertisers advertising in the Evening Express in March 2000 did not advertise in a free title between September 1999 and June 2000. C. The Tribunal’s Analysis

[288]We begin by pointing out that the RBB reports do not contain any acknowledgement by RBB that it is its duty to help the Tribunal, and that such duty overrides its obligation to any person from whom RBB has received instructions or by whom RBB is paid, as required by Part 35 of the Civil Procedure Rules and paragraph 10.5 of the Tribunal’s “Guide to Appeals under the Competition Act 1998”. Although, contrary to Aberdeen Independent’s submission, we do not regard RBB’s evidence as inadmissible, the absence of such an acknowledgement does detract from the weight we feel able to place on it.

[289]More fundamentally, we do not regard the RBB analysis as undermining the Director’s conclusion for the following reasons.

[290]First, the data relied on by RBB covers a period when the Herald & Post and the Independent were priced far below competitive levels. That is certainly the case for the period up to the end of March 2000 and in our view remained the case until at least June 2000, as discussed later in this judgment. As we ha ve already pointed out, we do not think that data from a period when market conditions are distorted can be safely relied on to show what the degree of substitutability would be in normal competitive conditions, which in our view is the relevant test: see paragraphs 274 to 276 above. Thus, in our view, evidence that, following rate increases in the Herald & Post , or changes in distribution in the Independent , there was no switching to the Evening Express is wholly inconclusive. At the artificially low pr ice levels then prevailing one would expect the advertiser, if he switched at all, either to switch to the rival free newspaper, the nearest price rival or, conceivably , to drop out of the market altogether, rather than switch to the Evening Express .

[291]Secondly, as regards at least the first four months of the ten-month period covered (from September 1999 to December 1999) the picture is likely to be further distorted by the possibility of Aberdeen Journals offering various “deals” across its three newspapers. We have no means of verifying whether or not RBB’s analysis could be affected by that factor. 88

[292]Thirdly, the RBB reports are very largely reliant on distribution changes as a proxy for price changes. It seems to us difficult safely to rely on a distribution change as equivalent to a price change without knowing whether the particular customer was sensitive to, or even aware of, the distribution change in question, and what volume of advertising he might have placed had the distribution change not taken place. For this reason, we find it hard to place reliance on RBB’s work.

[293]Fourthly, the RBB analysis is based on those customers of the Evening Express who advertised in the Evening Express in March 2000. It does not, therefore, pick up potential or actual customers who did not advertise in the Evening Express in March 2000. There is no reason to suppose that these advertisers did not regard free newspapers as alternatives to the Evening Express , either in March 2000 or over the period September 1999 to June 2000. Thus in our view the RBB reports leave out an important part of the competitive picture.

[294]Fifthly , we are not satisfied that we can safely rely on the RBB survey as regards customers of the Independent . First, as already stated, we are not persuaded that changes in circulation levels, largely affecting outlying areas outside urban Aberdeen, are as reliable as price changes for determining issues such as substitutability. Secondly, RBB – though admittedly through no fault of its own – did not have access to invoice price data for the Independent . In our view it is very difficult to make a reliable study of substitutability between three products if the prices of one of the three products are unavailable.

[295]Sixthly , because in this market transaction prices are negotiated prices, a study of the RBB kind is unlikely to pick up the full dynamics of the competitive forces at work. For example, suppose Aberdeen Journals seeks to raise the price by 5 per cent, but the customer indicates that that is unacceptable, and threatens to switch to the Independent . Aberdeen Journals’ response is, say, to waive or postpone the price increase. In this example, the volume placed with the Evening Express remains the same, and there is no price change, but competit ive forces have been operating. The same holds good, for example, in a case where Aberdeen Journals seeks a price increase of, say, 5 per cent, but the customer negotiates the increase down to 2½ per cent by threatening to switch some advertising to the Independent . Again, there is a price change (and no switching) but competition has nonetheless been at work. Similarly, Aberdeen Journals might quote a new price to an existing advertiser, who then decides to switch, rather than pay the new price. The result is that no invoice is ever rendered at the new price. If one is using invoice data, there is nothing to show a quoted price change, 89 but again competitive forces have been at work. Thus the conceptual basis underlying the RBB study is in our view fla wed.

[296]Seventhly, we have doubts about the underlying data. As regards the volume of advertising in the Independent , RBB was apparently working from a database compiled by Aberdeen Journals’ employees. It is uncertain whether that database is accurate, and data for the Independent for some periods is lacking. Both the Director and Aberdeen Journals pointed out that certain advertisers advertising in the Independent had been omitted, a fact that led RBB to correct its figure for advertisers advertising in the Evening Express in March 2000 who did not advertise in another newspaper in the period September 1999 to June 2000 from 46.4 per cent by volume to 42 per cent by volume.

[297]It seems to us that even the figure of 42 per cent is likely to be overstated, since in the case of smaller classified advertisements Aberdeen Journals would not necessarily be able to identify the names of the advertisers advertising in the Independent , e.g. because the advertisement carried only a box number or phone number, not a name. It seems to us quite likely that among the advertisers advertising in the Evening Express in March 2000 identified as not advertising in another newspaper between September 1999 and June 2000, there would in fact have been some advertising in the Independent in that period, albeit not identifiable by name.

[298]Eighthly, it seems to us that the criteria used by RBB in individual cases to judge whether there was switchability are unduly restrictive. Annex 2 to the Director’s defence lists many examples whic h are open to more than one interpretation and which, contrary to RBB’s assertion, do not permit the possibility of competition between the Evening Express and the free titles to be wholly excluded.

[299]Lastly, and reverting to common sense, even the RBB figures, showing that some 40 per cent by volume of advertisers advertising in the Evening Express in March 2000 did not advertise in a free title in the period September 1999 to June 2000, lead to the conclusion that at least 60 per cent of such advertisers did advertise in a free title during that same period. Moreover Table 5 to the first RBB report, set out above, shows that the leading advertisers placed substantial volumes in both the paid-for titles and the free titles. It is in our view inconceiv-able that all those advertisers regarded the paid-for and free titles simply as “complements”. It is much more probable, in our view, that a significant number of those advertisers would have regarded the paid-for and free titles as substitutes, depending on relative price levels, at 90 least for a significant part of the advertising in question, thus exercis ing a competitive constraint. We also note that in the period studied there appear to be few price increases for the Evening Express and at least nine instances when customers received a price decrease for advertising in the Evening Express . We have difficulty in seeing where the underlying price constraint was coming from, if it was not from the Independent.

[300]We conclude that RBB’s work does not raise a reasonable doubt in our mind as to the correctness of the Director’s analysis of the relevant product market contained in the decision. None of Aberdeen Journals’ other arguments do so either, in our view. D. Conclusion

[301]For the foregoing reasons we conclude that the Director has established that the Evening Express , Independent and Herald & Post all formed part of the market for advertising in local newspapers in the Aberdeen area in March 2000. 4. ALTERNATIVE MEDIA

[302]At paragraphs 86 to 92 of the decision the Director sets out his reasons for concluding that various other media – namely national media, the internet, local radio, recruitment agencies, the solicitors’ Property Register, property centres, direct mail, leaflets, niche/advertising-only publications, directories, outdoor advertising, billboards and cinema advertising – tend, at present, not to exercise a sufficient constraint on prices for advertising space in local newspapers to form part of the same market (paragraph 91). Whilst each type of advertising in local newspapers may be subject to a differing degree of constraint from other media, none is sufficient to alter the Director’s conclusion on product market definition. Even if it did, this would not fundamentally alter Aberdeen Journals’ market position (paragraph 92).

[303]While putting in issue the Director’s conclusion on this point, Aberdeen Journals has limited itself to alleging a contradiction between paragraphs 91 and 92 of the decision; criticising the Director for failing to take into account certain responses to the Director’s own consumer survey in March 2000; and relying on Professor Yamey’s evidence, which is to the effect that the Director should have considered whether alternative media, taken as a whole, exercised a competitive constraint on the Evening Express .

[304]The Director’s analysis at paragraphs 86 to 92 of the decision seems to us to be robust. As far as we are aware, none of the reports of the Competition Commission on newspaper 91 mergers referred to it under the Fair Trading Act 1973 has yet found other media to exercise a significant constraint on prices for advertising space in local newspapers.

[305]As to Aberdeen Journals’ arguments, we see no contradiction between paragraphs 91 and 92 of the decision. At paragraph 92 the Director is simply saying that even if, in certain sectors, alternative media might exert some price restraint (e.g. advertising-only publications for motors) that would not fundamentally affect the position of Aberdeen Journals in the local newspaper advertisin g market in Aberdeen. We see no reason to disagree with that view.

[306]As regards the Director’s own survey in March 2000, it is true that some of the ten local advertisers who responded to that survey indicated that they used alternative media such as the internet (mainly their own website), leaflets, local radio and other forms of media. However, we see nothing in that to undermine the Director’s conclusion that other media do not exercise a sufficient competitive constraint so as to form part of the same market as advertising space in local newspapers in Aberdeen.

[307]As to Professor Yamey’s point about looking at alternative media as a whole , in our view this has been recognised by the Director, who says (at paragraph 86 of the decision) , “[t]he key question is whether sufficient marginal local newspaper advertisers would consider switching all or a sufficient portion of their advertising expenditure to those alternative media, if faced with an appreciable price rise from a competitive price level, to render that price rise unprofitable.” The Director then goes on to consider (in paragraphs 87 to 90 of the decision) a range of alternative media and the extent to which they might constrain the price charged for advertising in local newspapers. In so doing, he correctly considered each alternative medium in turn, before reaching his overall conclusion that “for most advertisers, the alternative media listed above tend, at present, not to exercise a sufficient constraint on prices for advertising space in local newspapers to form part of the same market” (paragraph 91). Aberdeen Journals has not offered any concrete evidence to challenge this conclusion.

[308]It follows in our view that the Director was correct to exclude alternative media from the relevant market in the present case. 5. DOMINANCE IN THE MARKET FOR ADVERTISING IN LOCAL NEWSPAPERS IN ABERDEEN

[309]On the view we have formed of the relevant product market – advertising space in local newspapers in Aberdeen (paid-for and free) – Aberdeen Journals had a market share of 78 per 92 cent by value (67 per cent by volume) in the period January to March 2000. If account is taken of the fact that the Evening Express is partly distributed outside the geographic market of urban Aberdeen, Aberdeen Journals’ market share in that period was 73 per cent by value and 63 per cent by volume. On the basis of those market shares and given the existence of significant barriers to entry, the Director finds that Aberdeen Journals enjoys a dominant position in that market (paragraphs 144 to 148 of the decision).

[310]In our view the Director is correct to conclude that market shares of this order suffice to establish that Aberdeen Journals was dominant unless exceptional circumstances are shown: see e.g. Hoffman-La Roche , cited above, at paragraph 41; AKZO , cited above, at paragraph 60;

Case T-30/89

Hilti v Commission

[1991] ECR II-1439

, paragraphs 91-92; Tetra Pak II , cited above,

[1994] ECR II-755

, at paragraph 109. See also the Tribunal’s judgment in Napp , cited above, at [156] to [160]. In our view, that dominance is reinforced in this case by the fact that Aberdeen Journals had only one competitor, the Independent , which had been loss- making for four years. We also accept that Aberdeen Journals’ reaction to the launch of the Independent would have been likely to deter others seeking to enter the Aberdeen marketplace.

[311]Contrary to Aberdeen Journals’ argument, we see no “exceptional circumstances” in this case. It may be true that the Herald & Post lost some customers as a result of rate increases before or during the first quarter of 2000, but those customers had previously been paying prices well below cost, or even nominal prices in the case of property advertisers. In our view the fact that the Herald & Post lost some customers who had, up to that point, been paying prices well below cost is not an indicator of competitive forces operating in normal competitive conditions.

[312]In any event, what is relevant here is not what happened to the artificially priced Herald & Post , but the combined market share of the Evening Express and the Herald & Post in March 2000. Aberdeen Journals has not contested the market share figures of the Director as far as the period January to March 2000 is concerned.

[313]We therefore conclude that Aberdeen Journals had a dominant position in the market for local newspaper advertising (paid-for and free) in Aberdeen in March 2000, as found by the Director in the decision. 93 6. THE DIRECTOR’S ALTERNATIVE PRODUCT MARKET

[314]At paragraphs 135 to 139 of the decision, the Director argues, in the alternative, that Aberdeen Journals was dominant in the supply of advertising in paid-for local newspapers in Aberdeen (i.e. the Evening Express ) and that its conduct constituted an abuse of that position. Aberdeen Journals argues(i) it is not open to the Director to advance such an alternative market, which was abandoned during the first appeal; and

(ii) the Director’s alternative market is not, in any event, substantiated.

[315]In our view the Director was not debarred from relying in the decision on the alternative market analysis set out at paragraphs 135 to 139. The background is that such an analysis was included in the first decision, but had not formed part of the first Rule 14 notice. In his defence in the first appeal the Director indicated that, in consequence, he would not rely on that analysis in order to uphold the first decision. The Tribunal in its first judgment at [117] to [120] indicated that the alternative market analysis was potentially relevant, and stated that it was unsatisfactory that the Tribunal was not in a position to rule on this issue. However, the effect of the Tribunal’s first judgment was to set aside the first decision in its entirety. For the reasons set out at [190] to [194] of the judgment, the Tribunal remitted to the Director for further consideration “the issue of the definition of the relevant product market, in particular, which newspapers constitute the relevant product market”. The proceedings before the Director then resumed. The second Rule 14 notice included the alternative product market, and Aberdeen Journals responded to the Director on that point during the resumed administrative procedure.

[316]In our view, once the first decision had been set aside, the Director was at liberty to recommence hi s analysis of the relevant product market, which was “the matter” remitted to him. The fact that, for procedural reasons, he had not relied on the alternative product market during the first appeal did not in our view debar the Director from putting that issue to Aberdeen Journals in the second Rule 14 notice. That was done with due expedition, in accordance with the Tribunal’s order, and Aberdeen Journals was able to exercise its rights of defence on that issue during the administrative procedure. In those circumstances we find that the Director was entitled to rely on the alternative product market in the decision.

[317]However, on the view we have formed of the relevant product market, set out above, we have no need to deal with the Director’s alternative market analysis. We do so only for completeness. 94

[318]This case presents a somewhat atypical set of circumstances. If one takes, first, the situation in Aberdeen in 1996, prior to the launch of the Independent , many advertisers might well have said that the daily Evening Express and the low-quality, 20-page, weekly Herald & Post were not in normal circumstances substitutes, but were serving different kinds of demand. That specific situation arose from the way in which Aberdeen Journals chose to position its two newspapers in the Aberdeen market at that time. As we see it, the launch of the Independent in 1996 as a high-quality, 80-page, free weekly, carrying advertising for property and motors and other features which the Herald & Post did not, and with a wider distribution area eventually matching that of the Evening Express , gave many advertisers in the Evening Express a viable alternative for the first time. However, to counteract this, Aberdeen Journals dramatically changed the Herald & Post by transforming it into a high-quality, 80-page newspaper with a wider circulation, at the same time reducing the advertising rates of the Herald & Post to below cost.

[319]As we see it, the reaction of Aberdeen Journals does not alter the fact that the Independent was, actually or potentially, a competitor of the Evening Express in March 2000. If and to the extent that the market for local newspaper advertising in Aberdeen had, by March 2000, become segmented as between the “paid-for” Evening Express , on the one hand, and the free weekly newspapers, on the other hand, any such segmentation would in our view have been artificial, as a result of Aberdeen Journals’ tactics. For this reason, among many others, we adhere to our principal conclusion, already set out at length above, that all three newspapers formed part of the same relevant market in March 2000.

[320]However, even if, contrary to our view, it could be said that by March 2000 advertising in the Evening Express was in some sense in a separate market from advertising in the free titles, in our view the Director was correct to conclude that Aberdeen Journals was nonetheless dominant in the “market” for advertising in paid-for local newspapers in Aberdeen. The Evening Express is, in effect, the only supplier in that market.

[321]The fact that the Director did not analyse in this regard the precise borderline between the Evening Express and the Press & Journal as regards advertising in “paid-for” newspapers is not in our view material. First, paragraph 70 of the decision set out why the Director considers the Press & Journal to form part of a separate market, and Aberdeen Journals has not challenged that conclusion (see also paragraphs 137 to 138 above). Secondly, Aberdeen Journals owns the Press & Journal as well. Similarly, there is nothing to suggest that any further analysis on the Director’s part of alternative media or other newspapers would alter 95 the Director’s conclusion. On the assumption that there is a separate market for advertising space in “paid-for” local newspapers in Aberdeen, served by the Evening Express , we find that Aberdeen Journals is dominant in that market with a share of 100 per cent.

[322]As the Tribunal pointed out at [118] of its first judgment, it is well established that an undertaking with a dominant position in one market may abuse that position by engaging in predatory conduct on a neighbouring or associated market. The classic case where that occurs is where an undertaking with a dominant position in one market takes steps on another market which are aimed at reinforcing or protecting its position in the market on which it is dominant. Thus in AKZO , cited above, an undertaking dominant in the organic peroxides market engaged in predatory pricing in the flour additives sector, where the undertaking was not dominant, in order to deter a supplier of flour additives from entering the plastics sector of the organic peroxides market. That activity in the flour additives sector was held to be an abuse of the undertaking’s dominant position in the organic peroxides market, which it was designed to protect (see paragraphs 35 to 45 of that judgment). In

Case T-65/89

BPB Industries and British Gypsum v Commission

[1993] ECR II-389

, there was a shortage of plaster. An undertaking dominant in the separate market for the supply of plasterboard gave priority in deliveries of plaster (a product where no dominance had been found) to customers who were not stockists of imported plasterboard. It was held that this activity in the plaster market constituted an abuse of dominance in the plasterboard market, since the practice in question had the effect of disadvantaging or excluding potential competition from imported plasterboard: see paragraphs 92 to 97 of that judgment.

[323]In the present case, we have already found that the purpose of the policy followed by Aberdeen Journals as regards the free weekly Herald & Post was, notably, to protect the revenues of the paid-for Evening Express : see paragraphs 220 to 247 above. We thus agree with the Director’s assessment at paragraph 138 of the decision:

“… the loss making strategy of the Herald & Post can only be rationalised either as an attempt to prevent the Independent from attacking the Evening Express ’s revenues directly (on the basis that both titles were on the same market – as the Director argues) or as an attempt to eject the Independent from the free newspapers market before it could become a threat to Aberdeen Journals’ position on the separate but associated paid-for market, on which it enjoyed a monopoly.”

[324]As the Director points out, Aberdeen Journals has a significant presence in both the “paid-for” and “free” newspaper sectors, and is able to control the pricing policy of both the Evening Express and the Herald & Post ; the customer base overlaps; there is a common 96 readership; and Aberdeen Journals uses the same production facilities for both titles. In those circumstances, even on the – in our view unfounded – assumption that advertising in “paid-for” local newspapers in Aberdeen is a separate “market” from advertising in “free” local newspapers in Aberdeen, the links between Aberdeen Journals’ position on both markets are extremely close. Similarly, Aberdeen Journals’ conduct in relation to its “free” title is clearly linked to the protection of its “paid-for” title. In those circumstances, in our view the use by Aberdeen Journals of its position in relation to advertising in free newspapers in Aberdeen in order to protect its position in relation to advertising in paid-for newspapers in Aberdeen may properly be characterised as an abuse of its dominant position in advertising in “paid-for” local newspapers in Aberdeen.

[325]We add for completeness that in Tetra Pak II , cited above, the Court of Justice upheld a finding of abuse of a dominant position in circumstances where the conduct complained of on the non-dominant market (non-aseptic containers and machinery) was not shown to have been directed at strengthening or reinforcing the dominant position relied on (in aseptic containers and machinery) see

[1996] ECR I-5951

, at paragraphs 21 to 31. The present case, however, is a case of an undertaking with a dominant position in advertising in “paid-for” local newspapers using its position as regards advertising in “free” local newspapers to protect that dominant position, thus clearly linking the conduct with the dominant position being protected. We do not, therefore, need to consider whether there are special circum-stances of the kind that arose in Tetra Pak II . Nor do we see anything in the judgment of Laurence Collins J in Claritas (UK) Limited v The Post Office [2001] ECC 12, as casting doubt on the conclusion we have reached on the facts of the present case.

[326]Finally, as regards the assessment of penalty on the basis of the Director’s alternative market, we accept the Director’s submission that this is ultimately for the Tribunal to assess. We revert to that aspect below. V ARGUMENTS OF THE PARTIES ON THE ISSUE OF ABUSE

[327]At paragraphs 150 to 212 of the decision the Director found that Aberdeen Journals had abused it s dominant position by supplying advertising space in the Herald & Post at below average variable cost in the period 1 to 29 March 2000. That conduct affected trade within the United Kingdom. Hence, the Chapter II prohibition is infringed. 97 A. Aberdeen Journals’ submissions

[328]Aberdeen Journals argues, first, that the Director was wrong to take into account its pre- March 2000 conduct in assessing its pricing conduct, and in particular its intent after that date (see paragraph 193 of the decision). According to Aberdeen Journals, the pre-1 March 2000 behaviour is irrelevant, since it was lawful at the time, and was never subjected to investigation and scrutiny under either the Competition Act 1980 or the Fair Trading Act 1973, let alone to an adverse finding by the Monopolies and Mergers Commission under that legislation. Moreover, the Director acknowledges at paragraph 192 of the decision that Aberdeen Journals’ conduct, prior to 1 March 2000 was “lawful (absent regulatory intervention)”.

[329]Aberdeen Journals submits, secondly, that in focusing on whether Aberdeen Journals covered its average variable costs in March 2000, the Director has failed to address the fundamental question of whether Aberdeen Journals’ conduct during this period was in fact abusive at all.

[330]According to Aberdeen Journals, an essential element of abuse is the use of practices different from those normally adopted in the market which have the effect of restricting competition or risk the elimination of competitors. See OFT 402, The Chapter II prohibition, at paragraph 2.3, which correctly reflects Hoffmann-La Roche, cited above, at paragraph 91. In AKZO, cited above, at paragraphs 69 to 72 , the Court of Justice was doing no more than applying that rule to below-cost pricing. Similarly in Tetra Pak II , cited above,

[1996] ECR I-5951

at paragraph 41, the Court was concerned with a situation where competitors might be eliminated. Accordingly there is no abuse unless the Director has established, with credible evidence, the existence of a propensity or potential to eliminate competitors. Furthermore, says Aberdeen Journals, since predation is a medium- to long-term strategy, the Director must show predation measured over a period of time which might reasonably be expected to be sufficient for a dominant undertaking to exclude a rival (see Deutsche Post 2001 OJ L215/27, at paragraph 36; AKZO , cited above, paragraph 140; OFT 414 Assessment of Individual Agreements and Conduct , paragraphs 4.1 and 4.8).

[331]According to Aberdeen Journals, none of the evidence available to the Director, such as the limited duration of the conduct, the lack of impact on the Independent , and the stated intentions of Aberdeen Journals, suggests that the potential to eliminate competition in March 2000 could be established in the present case. The shortfall below average variable costs of 98 £17,670 8 for the Herald & Post in March 2000 is de minimis in terms of amount, duration, and the five editions involved. It is impossible to contend that the behaviour of the Herald & Post in this period could have eliminated the Independent or had any material effect on competition.

[332]Moreover, during the relevant period Aberdeen Journals was making every effort to make rapid adjustments to its costs and revenues, and was losing market share in the process. For instance, the Herald & Post reduced its distribution from 123,000 to 107,000 copies in October 1999, and to 83,974 copies in March 2000; reduced its pagination by 33 per cent in March 2000; reduced its non-revenue producing pagination; and moved the Herald & Post ’s administrative offices into those of Aberdeen Journals. At the same time, in October 1999, it increased its revenues in the short term by increasing rates for property advertisers: this resulted in the loss of virtually all of its property advertising revenue. The Herald & Post started increasing other advertising rates in January 2000. According to Aberdeen Journals, the steps taken resulted in the Herald & Post reverting to profitability during the Director’s investigation period.

[333]In any event, says Aberdeen Journals, the final March 2000 edition of the Herald & Post (published on 29 March) in fact covered its average variable costs by a small amount. (See Appendix B to the first notice of appeal.)

[334]All these factors, according to Aberdeen Journals, clearly demonstrate a break in the chain of conduct before and after 1 March 2000. In particular, the Director has not shown that the commercial behaviour of the Herald & Post had, or was likely to have, the aim or effect of eliminating the Independent from the market during March 2000. The memoranda dated from 12 July 1996 to January 2000 on which the Director relies (at paragraph 181) to determine a predatory intent are all irrelevant to the present case, because they were written before the entry into force of the 1998 Act.

[335]In any event, those memoranda cannot be relied upon as evidencing any predatory intention for the period 1 to 29 March 2000 since: 8 We use the corrected figures from the second decision, although the arguments as originally presented to us were based on the figures in the first decision. 99 ¾ The fact that Aberdeen Journals demonstrably changed its conduct before the entry into force of the 1998 Act, and has now returned to profitability, is more compelling than memoranda written before 1 March 2000. ¾ Mr Ezzat’s review of 18 January 2000 was written before any meeting with the OFT and accordingly must be read in its proper context. ¾ Mr Davidson’s letter of 10 March 2000 shows that it was the intention to progress to break-even on the Herald & Post .

[336]Aberdeen Journals further submits that, since in March 2000 it was taking all possible steps to revert the Herald & Post to profitability and did not introduce or maintain any obstacles to free competition, its conduct was both proportionate to the legitimate aim it was seeking to achieve, and objectively justified. The Herald & Post ’s ability to increase advertising rates depended to a significant extent on the competitive stance adopted by the Independent . Nonetheless it sought to return to profitability and covered average total costs in January 2001.

[337]Moreover, says Aberdeen Journals, the fact that the Independent had been unprofitable since its launch and continued to make losses during the period under investigation is due to its inefficient business model. Aberdeen Journals relies on OFT 414, cited above, at paragraph 4.8. which states that “inefficient entry” can lead to the incumbent firm being forced to price below average variable costs. That is not a predatory response.

[338]In addition, Aberdeen Journals submits that at paragraph 206 of the decision the Director wrongly failed to assess whether Aberdeen Journals could recoup losses incurred as a result of predation. The Director’s approach, says Aberdeen Journals, is contrary to OFT 414, cited above, at paragraph 4.19, which provides that “at a minimum, such excessive prices must recoup the losses incurred from the predatory strategy”. It is also contrary to the opinion of Advocate General Fennelly in Cases C-395/96 P and C-396/96 P Compagnie Maritime Belge v Commission

[2000] ECR I-1365

, at paragraph 136. According to Aberdeen Journals, the presumption in OFT 414 is that recoupment will take place where a company is dominant but is rebuttable where subsequent events demonstrate that recoupment will not be possible. That is the case here, since recoupment was plainly not possible in March 2000.

[339]At the oral hearing on the first appeal, Aberdeen Journals emphasised in particular that the short duration of the infringement, absence of intent, lack of competitive effect and inability to recoup losses in the month of March 2000 all demonstrate that there was no abuse of 100 dominance during that period. Aberdeen Journals referred the Tribunal to certain passages in a large bundle of American and Australian cases, without however taking us in detail through any specific authority relied on (see Herbert Smith’s letter of 19 December 2001). Aberdeen Journals also emphasised that the first Rule 14 notice alleged a period of intended predation lasting until September 2001. According to Aberdeen Journals, that was based on a misinterpretation of Aberdeen Journals’ business plan for the period July 2000 to September 2001, which the Director wrongly supposed showed a predatory intent throughout that period. Having abandoned his position in the first Rule 14 notice, Aberdeen Journals submits that the Director was wrong to fall back on the artificial period of one month, March 2000.

[340]Finally, in the second appeal, Aberdeen Journals withdrew its concession that its prices had been below average variable costs at any time during March 2000. Aberdeen Journals argues that the “variable” costs taken into account should only be those costs which vary according to advertising output. Costs such as distribution, editorial costs, and newsprint costs associated with editorial, as distinct from advertising, content, are not to be treated as variable costs since, according to Aberdeen Journals, it would be commercially logical to publish the Herald & Post for a month without advertising, in order to avoid exiting the market. Looked at from this perspective, says Aberdeen Journals, its revenues exceeded average variable costs in March 2000. This approach is supported by RBB Economics. Professor Yamey points out, in addition, that the Director has not investigated whether the Independent was an efficient market entrant, and has not investigated whether maintaining the Herald & Post at prices below average variable cost in March 2000 would have been a commercially rational decision, having regard to the consequences of a temporary closure during that month. B. The Director’s submissions

[341]The Director rejects Aberdeen Journals’ argument that pricing below average variable costs does not give rise to a rebuttable presumption of abuse unless the Director has established the existence of a material threat to competition: see AKZO , paragraph 71 and Tetra Pak II , paragraph 41, cited above. In any event, the potential economic effect of Aberdeen Journals’ pricing strategy after 1 March 2000 must be assessed in light of the situation prevailing in the market at that time, namely that it had been selling at prices below average variable costs for the past four years. Hence, as stated at paragraph 208 of the decision, Aberdeen Journals’ conduct in March 2000 had the potential to expel its only rival from the relevant market, the further month being the “straw that breaks the camel’s back”. Moreover, Aberdeen Journals has not suggested that there was an independent objective business reason for its conduct since 1996, other than to eliminate the Independent . The fact that prices were below average 101 variable costs in March 2000 was the result of that eliminatory policy. In any event, if Aberdeen Journals had not priced below average variable costs in March 2000, it would have attracted fewer customers and the Independent would have attracted more.

[342]Aberdeen Journals’ claim that it was taking “all possible steps” to revert the Herald & Post to profitability is not sustainable (see paragraphs 195 to 200 of the decision). Aberdeen Journals was notified of the case against it on 27 July 1999. The steps actually taken by Aberdeen Journals prior to the end of March 2000 were in fact very limited in nature: ¾ Although distribution was cut in October 1999 from 123,182 copies to below 107,591 and rates for property increased, the Herald & Post was still pricing below average variable costs. ¾ The February 2000 restructuring of the Herald & Post ’s organisation and relocation of the premises would only impact on fixed costs and have no effect in March 2000. ¾ The implementation of a new higher rate card from January 2000 onwards had little immediate impact on profitability. The higher rates did not bite until February 2000 at the earliest and for many advertisers had to be phased in over a longer period. In any event discounts of up to 50 per cent were permitted for new business and lower rates could be offered in specific circumstances. As Graph 1 of Annex 2 to the decision demonstrates, average rates increased noticeably only in April 2000.

[343]In any event, according to the Director, the steps taken by Aberdeen Journals were no t adopted as a normal commercial response to competitive forces, but rather in response to an investigation by the OFT. This does not amount to an “objectively justified measure” under Community law.

[344]Nor does the Director accept Aberdeen Journals’ claims that its pricing policy in relation to the Herald & Post in March 2000 was constrained by the Independent . Since it was possible for the Herald & Post to be priced above average variable costs in April 2000 it would have been possible to do so in March 2000. As stated in paragraph 199 of the decision, save for a reduction in distribution in October 1999, Aberdeen Journals took no decisive action to reduce its variable costs until the end of March 2000.

[345]In relation to recoupment, the Director submits that this does not form part of the applicable test for predation under EC or UK competition law. Furthermore, although Aberdeen Journals did not succeed in driving the Independent out of the market, and hence recoup its 102 losses, if its conduct had continued it would have had the potential to drive the Independent out of the market and to enhance its reputation for aggressive, exclusionary behaviour.

[346]Aberdeen Journals’ argument that the Director should have considered the Herald & Post ’s pricing policy on a weekly , rather than a monthly, basis does not affect the validity of the decision. Firstly, taken on a monthly basis, the Herald & Post was priced below average variable costs during the period 1 to 29 March 2000. Secondly, even if prices were analysed on a weekly basis, the Herald & Post was still pricing below average variable costs for most of March 2000.

[347]As regards Aberdeen Journals’ new arguments on average variable costs, the Director objects to the withdrawal by Aberdeen Journals of its previous concession: see CPR, 14.1.8. The Director in any event rejects the notion that, as a free newspaper, the Herald & Post could credibly be published without advertising. The Director considers that he has correctly identified the period over which the predatory pricing in question should be assessed. C. Aberdeen Independent’s submissions

[348]Aberdeen Independent supports the Director’s arguments. VI THE TRIBUNAL’S FINDINGS ON ABUSE A. The relevant law

[349]In Napp v Director General of Fair Trading

[2002] CAT 1

, [2002] CompAR 13 the Tribunal summarised the relevant law in the following terms:

“207. In Case 85/76 Hoffman-La Roche v Commission [1979] ECR 461 , which concerned a system of loyalty rebates operated by the dominant firm which made it difficult for competitors to enter the market, the Court of Justice stated at paragraph 91: “The concept of abuse is an objective concept relating to the behaviour of an undertaking in a dominant position which is such as to influence the structure of a market where, as a result of the very presence of the undertaking in question, the degree of competition is weakened and which, through recourse to methods different from those which condition normal competition in products or services on the basis of the transactions of commercial operators, has the effect of hindering the maintenance of the degree of competition still existing in the market or the growth of that competition.” 103 208. In Case 322/81 Michelin v Commission [1983] ECR 3451, which also involved a rebate system that etnded to tie dealers to the dominant company, the Court said at paragraph 57: “A finding that an undertaking has a dominant position is not in itself a recrimination but simply means that, irrespective of the reasons for which it has such a dominant position, the undertaking concerned has a special responsibility not to allow its conduct to impair genuine undistorted competition on the common market.” 209. In AKZO (Case C-62/86 AKZO Chemie v Commission [1991] ECR I-3359 ), where the dominant firm offered prices discounted below cost in order to force a competitor out of business, the Court held: “[70] Article 82 prohibits a dominant undertaking from eliminating a competitor and thereby strengthening its position by using methods other than those which come within the scope of competition on the basis of quality. From that point of view, however, not all competition by means of price can be regarded as legitimate. [71] Prices below average variable costs (that is to say, those which vary depending on the quantities produced) by means of which a dominant undertaking seeks to eliminate a competitor must be regarded as abusive. A dominant undertaking has no interest in applying such prices except that of eliminating competitors so as to enable it subsequently to raise its prices by taking advantage of its monopolistic position, since each sale generates a loss, namely the total amount of the fixed costs (that is to say, those which remain constant regardless of the quantities produced) and, at least, part of the variable costs relating to the unit produced. [72] Moreover, prices below average total costs, that is to say, fixed costs plus variable costs, but above average variable costs, must be regarded as abusive if they are determined as part of a plan for eliminating a competitor. Such prices can drive from the market undertakings which are perhaps as efficient as the dominant undertaking but which, because of their smaller financial resources, are incapable of withstanding the competition waged against them.” 210. AKZO was followed in Case T-83/91 Tetra Pak v Commission [1994] ECR II-755 ), on appeal, Case 333/94P Tetra Pak v Commission [1996] ECR I-5951 (“ Tetra Pak II ”). The Court of First Instance, applying the criteria set out in AKZO , found that certain of Tetra Pak’s prices were below variable direct costs, and in one case below average variable cost (paragraph 151), and had no other economic rationale other than ousting Tetra Pak’s principal competitor (paragraphs 147 to 151, and 188 to 192 of its judgment). On the subsequent appeal the Court of Justice held at paragraphs 41 to 44: “41. In AKZO this Court did indeed sanction the existence of two different methods of analysis for determining whether an undertaking has practised predatory 104 pricing. First, prices below average variable costs must always be considered abusive. In such a case, there is no conceivable economic purpose other than the elimination of a competitor, since each item produced and sold entails a loss for the undertaking. Secondly, pr ices below average total costs but above average variable costs are only to be considered abusive if an intention to eliminate can be shown. 42. At paragraph 150 of the judgment under appeal, the Court of First Instance carried out the same examination as did this Court in AKZO . For sales of non-aseptic cartons in Italy between 1976 and 1981, it found that prices were considerably lower than average variable costs. Proof of intention to eliminate competitors was therefore not necessary. In 1982, prices fo r those cartons lay between average variable costs and average total costs. For that reason, in paragraph 151 of its judgment, the Court of First Instance was at pains to establish – and the appellant has not criticised it in that regard – that Tetra Pak intended to eliminate a competitor. ... 44. Furthermore, it would not be appropriate, in the circumstances of the present case, to require in addition proof that Tetra Pak had a realistic chance of recouping its losses. It must be possible to penalise predatory pricing whenever there is a risk that competitors will be eliminated. The Court of First Instance found, at paragraphs 151 and 191 of its judgment, that there was such a risk in this case. The aim pursued, which is to maintain undistorted competition, rules out waiting until such a strategy leads to the actual elimination of competitors.” 211. In Cases T-24-26 and 28/93 Compagnie Maritime Belge v Commission [1996] ECR II-1201, on appeal Cases C-395 and 396/96P Compagnie Maritime Belge v Commission [2000] ECR I-1365 (“ Compagnie Maritime Belge ”), a liner conference, Cewal, was found to have abused a dominant position on certain shipping routes between Europe and West Africa, by selectively lowering its freight rates to match the rates charged by its main independent competitor for ships sailing on the same or similar dates, a practice known as ‘fighting ships’. It was not shown that the members of Cewal had incurred losses, only a reduction in profits. The Court of First Instance held at paragraph 146: “[146] As has already been pointed out, it has been consistently held that whilst the fact that an undertaking is in a dominant position cannot deprive it of entitlement to protect its own commercial interests if they are attacked; and whilst such an undertaking must be allowed the right to take such reasonable steps as it deems appropriate to protect those interests, such behaviour cannot be allowed if its real purpose is to strengthen this dominant position and thereby abuse it 105 (in particular, BPB Industries and British Gypsum v Commission ).”

The Court of First Instance held that the purpose of the practice was to eliminate the conference’s only competitor, and that, in any event, the response by Cewal to the situation which it faced was not reasonable and proportionate (paragraphs 147 and 148). 212. In his opinion (

[2000] ECR I-1365

) on Cewal’s appeal to the Court of Justice, Advocate General Fennelly referred to paragraphs 71 and 72 of AKZO , and said at paragraph 127:

“127. Apparently, therefore, sale s below average variable (or short-run marginal: AKZO , paragraph 70) costs are in effect presumed to be abusive. While it is usually rational to sell above average variable costs, because that permits some return on capital, where the market will not bear a higher price, it is not usually rational to sell below average variable costs. Marginal costs need not be incurred and business has no interest in incurring them so as to make a loss. A dominant firm would be permitted, however, to rebut this presumption by showing that such pricing was not part of a plan to eliminate its competitor.” 213. After considering that even prices above average variable costs, yet still below average total or long-run marginal costs, (see AKZO , paragraph 72), must be considered abusive where it is established that they are part of a plan to eliminate a competitor, Mr Fennelly went on to consider the case where a dominant undertaking prices above average total costs. He concluded at paragraph 132: “132. I would, on the other hand, accept that, normally, non-discriminatory price cuts by a dominant undertaking which do not entail below-cost sales should not be regarded as being anti-competitive. In the first place, even if they are only short lived, they benefit consumers and, secondly, if the dominant undertaking’s competitors are equally or more efficient, they should be able to compete on the same terms. Community competition law should thus not offer less efficient undertakings a safe haven against vigorous competition even from dominant undertakings. Different considerations may, however, apply where an undertaking which enjoys a position of dominance approaching a monopoly, particularly on a market where price cuts can be implemented with relative autonomy from costs, imple ments a policy of selective price cutting with the demonstrable aim of eliminating all competition. In those circumstance, to accept that all selling above cost was automatically acceptable could enable the undertaking in question to eliminate all competition by pursuing a selective pricing policy which in the long run would permit it to increase prices and deter potential future entrants for fear of receiving the same targeted treatment.” 106 214. Dealing with the specific facts of Compagnie Maritime Belge , Mr Fennelly commented at paragraph 137: “137. In all these circumstances, the Court of First Instance committed no error of law in finding that the response of Cewal members to the entrance of G&C was not ‘reasonable and proportionate’. To my mind, Article 86 cannot be interpreted as permitting monopolists or quasi-monopolists to exploit the very significant market power which their superdominance confers so as to preclude the emergence either of a new or additional competitor. Where an undertaking, or group of undertakings whose conduct must be assessed collectively, enjoys a position of such overwhelming dominance verging on monopoly, comparable to that which existed in the present case at the moment when G&C entered the relevant market, it would not be consonant with the particularly onerous special obligation affecting such a dominant undertaking not to impair further the structure of the feeble existing competition for them to react, even to aggressive price competition from a new entrant, with a policy of targeted, selective price cuts designed to eliminate that competitor. Contrary to the assertion of the appellants, the mere fact that such prices are not pitched at a level that is actually (or can be shown to be) below total average (or long-run marginal) costs does not, to my mind, render legitimate the application of such a pricing policy.” 215. In its judgment in Compagnie Maritime Belge the Court of Justice held at paragraphs 112 to 120: “112. It is settled case-law that the list of abusive practices contained in Article 86 of the Treaty is not an exhaustive enumeration of the abuses of a dominant position prohibited by the Treaty (Case 6/72 Europemballage and Continental Can v Commission [1973] ECR 215 , paragraph 26). 113. It is, moreover, established that, in certain circumstances, abuse may occur if an undertaking in a dominant position strengthens that position in such a way that the degree of dominance reached substantially fetters competition ( Europemballage and Continental Can , paragraph 26). 114. Furthermore, the actual scope of the special responsibility imposed on a dominant undertaking must be considered in the light of the specific circumstances of each case which show that competition has been weakened (Case C-333/94 P Tetra Pak v Commission [1996] ECR I-5951 , paragraph 24).”

After referring to the specific circumstances of the maritime transport sector, the Court continued: 107 “117 It follows that, where a liner conference in a dominant position selectively cuts its prices in order deliberately to match those of a competitor, it derives a dual benefit. First, it eliminates the principal, and possibly the only, means of competition open to the competing undertaking. Second, it can continue to require its users to pay higher prices for the services which are not threatened by that competition. … 119. It is sufficient to recall that the conduct at issue here is that of a conference having a share of over 90% of the market in question and only one competitor. The appellants have, moreover, never seriously disputed, and indeed admitted at the hearing, that the purpose of the conduct complained of was to eliminate G&C from the market. 120. The Court of First Instance did not, therefore, err in law, in holding that the Commission’s objections to the effect that the practice known as ‘fighting ships’, as applied against G&C constituted an abuse of a dominant position were justified. …” 216. Finally in

Case T-228/97

Irish Sugar v Commission

[1999] ECR II-2969

(“ Irish Sugar ”), which concerned notably the legality of certain border rebates, the Court of First Instance held (at paragraph 114) that in determining whether a pricing policy is abusive under Article 82 of the Treaty: “it is necessary to consider all the circumstances, particularly the criteria and rules governing the grant of the discount, and to investigate whether, in providing an advantage not based on any economic service justifying it, the discount tends to remove or restrict the buyer’s freedom to choose his sources of supply, to bar competitors from access to the market, to apply dissimilar conditions to equivalent transactions with other trading parties or to strengthen the dominant position by distorting competition ( Hoffman-La Roche , paragraph 90; Michelin , para-graph 73). The distortion of competition arises from the fact that the financial advantage granted by the undertaking in a dominant position is not based on any economic consideration justifying it, but tends to prevent the customers of that dominant undertaking from obtaining their supplie s from competitors ( Michelin , paragraph 71). One of the circumstances may therefore consist in the fact that the practice in question takes place in the context of a plan by the dominant undertaking aimed at eliminating a com-petitor ( AKZO , paragraph 72; Compagnie Maritime Belge Transports , paragraphs 147 and 148).” ” 108

[350]The cases cited above demonstrate, in our view, that the question whether a certain pricing practice by a dominant undertaking is to be regarded as abusive for the purposes of the Chapter II prohibition is a matter to be looked at in the round, taking particularly into account(i) whether the dominant undertaking has had “recourse to methods different from those which condition normal competition in products or services on the basis of the transactions of commercial operators” ( Hoffman-La Roche , cited above, at paragraph 91); and

(ii) whether such conduct has the effect of weakening or distorting competition in the relevant market, having regard to the special responsibility of a dominant firm not to impair genuine undistorted competition. In our view, these principles apply particularly to the case of a dominant firm facing new entry, where retaliatory measures going beyond what is reasonable and proportionate are likely to require close scrutiny under the Chapter II prohibition.

[351]Within that framework, the cases of AKZO , Tetra Pak II and Compagnie Maritime Belge , cited above, give further guidance as to when prices below costs are likely to be regarded as abusive. As the Director points out at paragraphs 151 to 152 of the decision, AKZO (at paragraph 71) and Tetra Pak II (at paragraph 41) show that pricing below average variable costs by a dominant firm is normally to be regarded as an abuse. “Variable costs” are those which vary with the unit of output produced as distinct from “fixed costs” which do not vary with the output produced. An example of a “fixed cost” might be the monthly rental of a company’s premises. Examples of “variable costs” in the present case include, but are not necessarily limited to, newsprint (the paper on which the newspapers are printed), distribution (the cost of distributing the copies produced), and other costs such as ink, plate and film charges, electricity, fuel for transport, overtime and “pre-press productio n” costs (see paragraph 162 of the decision). Thus, for example, to sell the Herald & Post below average variable cost, as so defined, is to sell each copy of the newspaper for less than the average cost of producing that copy. As Mr Fennelly points out at paragraph 127 of his opinion in Compagnie Maritime Belge , cited above, it is not normally rational for an undertaking to act in this way (see also paragraph 4.7 of OFT 414, Assessment of Individual Agreements and Conduct ). When undertaken by a dominant firm, such conduct will normally constitute “recourse to methods different from those which condition normal competition” within the meaning of Hoffman-La Roche .

[352]Similarly, AKZO (at paragraph 71) and Tetra Pak II (at paragraph 41) show that it may be an abuse by a dominant undertaking to price between average variable cost and average total cost, if the intention is to eliminate a competitor. Total costs are variable costs plus fixed costs including, in our view, where appropriate, a share of general overheads. In our view, 109 pricing between average variable cost and average total cost is likely to be abusive when undertaken in anticipation of competitive entry or in order to undercut a new entrant.

[353]One particular aspect of the distinction between average variable costs and average total costs, not yet explored in the case law, but relevant in the present case, is the period of time over which costs are to be assessed as “fixed” rather than “variable”. The longer the period that is taken, the more likely it is that cost will be classified as variable since, for example, over a longer timescale, employees can be dismissed or plant closed in response to changes in output. Indeed, in the long run, almost all costs are “variable”.

[354]The Director suggests, at paragraph 4.6 of OFT 414, referred to in paragraph 175 of the decision, that the period to be used for determining which costs are to be treated as “variable” and which “fixed”, is “the time period over which the alleged predatory price or set of prices prevailed or could reasonably be expected to prevail”. Despite some possible circularity in this approach, paragraph 4.6 of OFT 414 seems to us to be a useful starting point. What the Director should do, in the first instance, is to identify provisionally the period over which pricing below cost is suspected. He should then take that period and examine whether costs are variable over that period. We do not exclude the possibility of the Director taking other periods, for example a year or a period of months , or even less, as a cross check, if to do so would be reasonable from a business perspective. Whether the period taken is a reasonable period will be a matter of fact and degree, to be judged in the circumstances of each particular case.

[355]The consequence is that the longer a dominant undertaking prices at some level below total costs, the more likely it is that costs which might be treated as “fixed” in the short run should be treated as “variable” for the purpose of applying the AKZO test. In applying the Chapter II prohibition, that would not seem to us an unreasonable approach. In order to survive in the market a competitor needs to cover total costs, including overheads. The longer a dominant undertaking prices below total costs, the more likely it is that an equally efficient competitor will be forced to exit the market. That risk is not averted simply because the dominant firm may be covering its average variable costs as measured on a short-run basis.

[356]Similarly, it seems to us, the longer the prices of a dominant undertaking remain below total costs the easier it is likely to be to infer an intent to eliminate competition, in accordance with the AKZO test, absent special circumstances such as recessionary conditions. Such an 110 intention may be inferred, of course, from other circumstances, such as selective price cutting.

[357]The Director accepts, at paragraphs 205 and 209 of the decision, that in some circumstances, a dominant firm may show that its pricing below cost is “objectively justified”. Again that approach is in accordance with the opinion of Advocate General Fennelly in Compagnie Maritime Belge , cited above, at paragraph 127. Some examples of what, in the Director’s view, may constitute “objective justification” are set out in OFT 414, at paragraph 4.8. Even where prices are below average variable costs, and despite the apparently peremptory wording of the judgments of the Court of Justice in AKZO , at paragraph 71 and Tetra Pak II , at paragraph 41, we do not exclude the possibility that, exceptionally, a dominant firm may be able to rebut the presumption of abuse.

[358]However, in our view the presumption of abuse will rarely, if ever, be rebutted if the pricing policy under scrutiny originates as an aggressive response to market entry by a competitor , or is directed towards eliminating a competitor. An objective justification will normally be particularly difficult to establish if there is evidence of selective price cutting by a dominant undertaking that is targeted specifically towards the customers or potential customers of a competitor. Indeed, Compagnie Maritime Belge and Irish Sugar show that price discrimination of that kind by a dominant undertaking may be an independent head of abuse, even if the targeted price cuts in question do not fall below average total cost.

[359]Bearing those comments in mind, we now examine more closely the methodology used by the Director in the decision. B. Some comments on the Director’s methodology in the decision

[360]In the decision, the Director has founded his analysis essentially on the criteria set out in AKZO and Tetra Pak II , cited above (see paragraphs 151 and 152 of the decision). For the reasons he gives at paragraphs 155 to 209 of the decision, the Director considers that Aberdeen Journals sold the Herald & Post below average variable cost in the months of March, May and June 2000, which the Director presumes to be abusive conduct in accordance with the AKZO test (see paragraphs 175 to 180 of the decision). However, the Director accepts that Aberdeen Journals had an “objective justification” for pric ing below average variable cost in May and June 2000 because of the costs incurred in printing the Herald & Post in Leicester during those months as a result of industrial action in Aberdeen (see paragraph 205 of the decision). In consequence, the Director finds that Aberdeen 111 Journals abused its dominant position by pricing below average variable cost during March 2000 (paragraphs 207 to 209).

[361]In our view four aspects of the methodology which the Director has used in the decision call for comment, namely,(i) the negative contribution being made by the Herald & Post ;

(ii) the treatment of printing costs;

(iii) the time period for assessing predation in this case; and (iv ) the costs treated by the Director as varia ble . The negative contribution of the Herald & Post

[362]At paragraphs 158 to 163 of the decision, the Director explains that his primary source of information is the management accounts of the Herald & Post . Those give the “directly attributable revenue” of the Herald & Post (essentially advertising revenue, and to a lesser extent leaflet distribution income), and what are described as “the directly attributable costs”. The categories of cost shown in the management accounts are:(i) newsprint (i.e. the cost of the paper);

(ii) editorial;

(iii) advertising (assumed to be the cost of the staff selling advertising space); and

(iv) circulation (i.e. the cost of distribution). The difference between the directly attributable revenue and the directly attributable costs is the “contribution” (positive or negative) that the Herald & Post makes (or does not make) towards other costs which are not charged to the Herald & Post in the management accounts.

[363]In the five months prior to the launch of the Independent in March 1996, the Herald & Post was making a positive contribution averaging a little under £20,000 a month, representing a “gross margin” of revenue over costs of about 40 per cent: see Annex 2, Graph 5 of the decision.

[364]From 1996 onwards there followed a prolonged period of severely reduced prices, increased pagination and increased distribution. Notwithstanding adjustments made, notably after March 2000, the Herald & Post apparently failed to cover its directly attributable costs in every month up to and inc luding January 2001 (see the information submitted by Aberdeen Journals in response to the first Rule 14 notice). Following further forecasted reductions in circulation and pagination, there appears to have been a small forecasted surplus in February 2001 and for the remaining months in that financial year. Up to that point, a positive contribution had not been made since June 1996.

[365]Paragraphs 171 to 174 of the decision give details of what are described as “the losses”, by which is meant the negative contribution, of the Herald & Post in the period from 1996 to 112 July 2000. The use of the word “losses” in this context is not, in our view, technically correct. In paragraph 171 the figures in question relate only to the Herald & Post ’s negative contribution to Aberdeen Journals on the basis of the management accounts, and while those in paragraph 174 take account also of certain other costs attributed to the Herald & Post , neither set of figures make any allowance for other operations and overhead costs whic h are incurred by Aberdeen Journals but not charged to the Herald & Post in the management accounts. “Operations” not included in the management accounts include pre-press and printing staff costs. Overheads could include, for example, the cost of central management and administration, premises, maintenance and depreciation of machinery, financial controls and audit, information technology, finance charges and so on.

[366]The information before the Tribunal does not enable us to determine whether, or at what point, the Herald & Post ever reached a position in which it could be said that the revenues of the Herald & Post exceeded its average total costs, including an allocated share of operations costs and general overheads. More generally, as a matter of terminology, it seems to us incorrect for Aberdeen Journals to contend that the Herald & Post returned to “profitability” in about February 2001. We would not ourselves regard a product as “in profit” in a normal commercial sense if it is barely breaking even on its directly attributable costs, let alone making an adequate contribution to other operations costs and general overheads.

[367]However, paragraph 174 of the decision shows that even on the basis of the Herald & Post ’s management accounts, the Herald & Post was failing by a long way to make a positive contribution to overheads in the months of March to July 2000. The figures are as follows: Herald & Post negative contribution March to July 2000 on the basis of the management accounts March 2000 April 2000 May 2000 June 2000 £ £ £ £ Income 82,397 68,582 65,225 69,421 Total costs in management accounts 1 122,648 84,380 87,169 88,161 Contribution shown by management accounts (40,251) (15,798) (21,944) (18,740) July 2000 £ 56,314 64,980 (8,666) 1 Including Leicester printing costs for May and June.

[368]It is also plain from paragraphs 161 to 163 of the decision that the management accounts in fact understate the negative contribution of the Herald & Post in the period March to July 2000, since certain items of directly attributable costs are in fact omitted from these 113 documents. In the first decision the Director identified these costs as ink, plate and film charges. Following questions from the Tribunal during the first appeal, further heads of directly attributable costs were identified by Aberdeen Journals, namely electricity, transport-fuel, overtime and production pre-press: see paragraph 162 of the decision. The Director has not apparently verif ied the data supplied by Aberdeen Journals as to the amount of these costs. We are not in a position to do so, although we note Aberdeen Independent’s submission that these costs are even now unrealistically low. For example, we note that Aberdeen Journals’ estimate of the variable cost ele ment for pre-press operations was less than 10 per cent of the costs stated to have been incurred by the Independent for its corresponding pre-press activity. Taking Aberdeen Journals’ figures at face value, the resulting figures for the period March to July 2000 are as follows, on the basis of the table to paragraph 174 of the decision. HERALD & POST CONTRIBUTION SINCE FEBRUARY 2000 March 2000 £ April 2000 £ May 2000 £ June 2000 £ July 2000 £ Income 82,397 68,582 65,225 69,421 56,314 Total costs in management accounts 1 122,648 84,380 87,169 88,161 64,980 Contribution shown by management accounts (40,251) (15,798) (21,944) (18,740) (8,666) Additional costs incurred in producing the Herald & Post 7,787 4,780 1,591 3,994 3,852 Negative contribution of the Herald & Post (48,038) (20,578) (23,535) (22,734) (12,518) 1 Including Leicester printing costs for May and June.

[369]It is clear from the foregoing that the revenues of the Herald & Post were well below total cost after March 2000, even without taking account of operations costs and overheads. We have no reason to suppose that the position changed before early 2001.

[370]Although the Director has approached this case on the basis of pricing below average variable costs, we have already pointed out (at paragraph 355 above) that, in order to survive in the market, a competitor to a dominant firm must normally cover its total costs (including overheads) and earn a return on its investment. Moreover, in our view, in normal commercial business, each product line is expected not merely to cover its variable costs, but to make an 114 appropriate contribution to general overheads. If a dominant firm prices below average total costs, including a proportionate share of general overheads, for a prolonged period, sooner or later an equally efficient competitor will be forced out of the market.

[371]Thus, where prices of a dominant undertaking are above average variable costs but below average total costs, and there is evidence that those prices result from, or originate in, an aggressive response by the dominant firm to competition, we would expect such conduct to be closely scrutinised from the point of view of the Chapter II prohibition. Such scrutiny should, in our view, include the following questions: what proportion of overheads and operations costs should be included in “average total cost”; what costs, including operations costs, should be regarded as being “variable” having regard to the duration of the alleged predation, or such other period as is reasonable from a business perspective; and whether an intention to eliminate a competitor can be inferred. We do not ourselves regard pricing by a dominant firm that is below total cost but on some measure above average variable costs as necessarily a “safe haven” as far as the Chapter II prohibition is concerned, depending of course on the timescale and circumstances involved.

[372]Since, however, in this case the Director has proceeded on the basis of prices below average variable costs, as distinct from prices between average variable costs and average total costs, we do not need to take this issue further. The issue of printing costs

[373]It emerged during the first appeal that, for whatever reason, but somewhat surprisingly, no information had been sought by the Director about printing costs incurred by Aberdeen Journals in printing the Herald & Post other than ink, plate and film costs. After questions from the Tribunal and correspondence with the Registrar, Aberdeen Journals contended that the only printing costs properly attributable to the Herald & Post , apart from ink, plate and film costs, were relatively small amounts for overtime and pre-press production costs, which are now taken into account by the Director at paragraphs 162, 174 and 179 of the decision. The Director seems to have accepted this contention since, apart from these costs, the decision does not attribute any costs to the Herald & Post in respect of the Aberdeen Journals’ workforce engaged in printing operations, the cost of which is a substantial item.

[374]Aberdeen Journals submits that this approach is justified because, it says, almost the entirety of Aberdeen Journals’ printing press is devoted to its daily titles, whereas the Herald & Post is treated as a “residual” publication: all the fixed costs involved in operating Aberdeen 115 Journals’ printing press would be incurred, whether or not the Herald & Post was printed. Aberdeen Journals explains that its practice “has always been to cost the Herald & Post on a marginal basis, using spare capacity where it can” (see Herbert Smith’s letter to the Registrar of 18 December 2001). Aberdeen Journals has, however, included in the figures supplied to the Director the full cost of printing the Herald & Post in the months of May and June 2000, when the Herald & Post was printed at Leicester by another company in the Northcliffe group, as a result of industrial action in Aberdeen.

[375]Although this issue has not been fully argued, we think the following three comments are pertinent.

[376]First, there has been no investigation by the Director as to whether any part of the operations costs, notably those of the Aberdeen Journals’ printing operations, should be allocated to the Herald & Post , or indeed as to whether all those costs are “fixed”, or are partly fixed and partly variable . For example, if the Herald & Post was not published, is it the case that there would be no adjustment to shift patterns or manning levels (other than a reduction in overtime) and no savings in maintenance?

[377]Secondly, it seems to us that, even if the cost figures submitted by Aberdeen Journals are correct (see paragraphs 388 to 402 below), an asymmetry has crept in to the decision, in that the Director has treated the Leicester printing costs in May and June 2000 as variable costs albeit, according to the Director, objectively justified, while he has not taken into account in his calculations printing costs at Aberdeen in March, April and July 2000. However, a competitor such as the Independent has to cover its printing costs, and has to compete with the Herald & Post irrespective of whether the latter is printed at Aberdeen or Leicester. Since in principle the economic cost to Northcliffe of printing the Herald & Post should be much the same whether it is printed in Aberdeen or Leicester, it seems to us at first sight somewhat curious that the application of the 1998 Act should depend in this case to some extent on whether the Herald & Post is printed at one Northcliffe site rather than another.

[378]Thirdly, and more fundamentally, the argument presented by Aberdeen Journals is , in effect, that a dominant firm, using its spare capacity, is entitled to price against a competitor on a marginal basis and that, so long as its marginal revenues exceed its marginal costs (for which average variable costs are a proxy) it is acting lawfully, and does not need to allocate any costs, other than marginal costs, to the particular activity in question. 116

[379]This argument has not, so far as we know, yet been considered by the Court of Justice, or the Court of First Instance. The Tribunal has not heard submissions on this issue, and it is not necessary for the Tribunal to rule on it for the purpose of deciding the present case. Nonetheless, it does seem to us that if a dominant undertaking is able to exclude from its computations significant elements of cost which have to be borne one way or another, and which any equally efficient competitor would have to bear, there is a risk that the dominant firm will always be able , sooner or later, to undercut an equally efficient competitor and drive it from the market. That, in our view, is a particular risk where the marginal cost of a particular strategy, such as use of a “fighting title” (as in this case) or a “fighting ship” (as in Compagnie Maritime Belge , cited above) may well be very low.

[380]In our view, the cost-based rules set out in AKZO and Tetra Pak II , while providing guidance, are not an end in themselves and should not be applied mechanistically. The ultimate aim of the 1998 Act is to secure conditions of undistorted and effective competition. With that primary aim in view, a principal role of the Chapter II prohibition is to prevent dominant firms from defending or strengthening their dominant position in ways that are unreasonable and disproportionate, particularly by using methods different from those found under normal competitive conditions. In our view, the decision of the Court of Justice in Compagnie Maritime Belge itself shows that the guidance available in AKZO and Tetra Pak II is open to further development.

[381]In those circumstances, it will be for the Director to consider in the first instance, in future cases, whether arguments of the kind advanced by Aberdeen Journals in relation to its printing costs are acceptable under the 1998 Act or not. The time period for assessing predation in this case

[382]As we have already said, the Director considers that fixed costs are “costs which do not vary with an undertaking’s output” while variable costs are “costs which vary with an undertaking’s output” : see also the glossary in OFT 414, pages 27 and 28. However, as already indicated, what is a “fixed” cost and what is a “variable” cost may depend on the time scale being considered. Costs that may be “fixed” over the short run may be “variable” if a longer period is taken. At paragraph 175 of the decision the Director refers to paragraph 4.6 of OFT 414 which states: “the relevant timescale for the analysis of costs in assessing allegations of predation is the time period over which the alleged predatory price or set of prices prevailed or could reasonably be expected to prevail.” 117

[383]However, the Director goes on at paragraphs 175 and 176 of the decision:

“Predation prevailed since 1996. However, variable costs have been assessed on the basis of periods limited to a single month, as Aberdeen Journals produces management accounts monthly and it was a period over which short term planning for the Herald & Post might be determined. Over one month, fewer costs are variable than would be the case if a period of several months was used. Accordingly, such a short reference period errs against a finding of predation. Newsprint and circulation costs as proxy for variable costs Over one month, there is a fixed element to the costs of the editorial staff and the advertising team for the Herald & Post , but the costs of newsprint and circulation are variable. Assessing the period from October 1995, the Director has used only newsprint plus circulation costs as a proxy for variable costs.”

[384]It appears to us that in this case the Director has not, in fact, followed paragraph 4.6 of OFT 414. He has sought to take as “variable”, costs which are variable over as short a period as one month. Had he taken “the time period over which the alleged predatory pricing prevailed or could reasonably have been expected to prevail”, in accordance with paragraph 4.6 of OFT 414, he would have been entitled to ask himself which costs were “variable” over the period from mid-1996 to January 2001, i.e. the whole period when the Herald & Post was making a negative contribution. On such an approach, a very high proportion, if not all, of the costs of the Herald & Post could probably have been categorised as “variable”.

[385]Indeed, in the first Rule 14 notice, the Director suggested that Aberdeen Journals was guilty of an abuse since it had been pricing the Herald & Post below its “average avoidable costs” over the period of the alleged predation, in other words its revenues had been less than the costs that would have been avoided if the Herald & Post had not been published over the period mid-1996 to 2001. 9 The Director considered that the same result would follow if the “avoidable costs” were judged over the period of a year. This approach would, it seems to us, have been very similar, in its practical result, to identifying what costs were “variable” over the period of the alleged predation, or over an intermediate period of, say, one year, rather than over the short period of one month. Had the decision followed an approach similar to that in the first Rule 14 notice, it seems to us that a higher proportion of the Herald & Post ’s costs would have been classified as variable.

[386]In our view, the approach the Director has followed in the first and second decisions is favourable to Aberdeen Journals. However we do not explore this issue further, since we do 9 We have not examined in this judgment whether the concept of “average avoidable costs”, not mentioned in AKZO and Tetra Pak II , would materially illuminate the analysis in this case. 118 not think it would be fair to Aberdeen Journals to proceed, at this stage, on a basis different from that adopted by the Director. We are however satisfied that, in assessing variable costs over the period of one month, the Director has almost certainly understated the level of the amount of costs properly to be classified as “variable”. That approach, in our view, tends to favour Aberdeen Journals since the threshold at which Aberdeen Journals’ revenues could rise above variable costs is correspondingly low. For this reason, contrary to Aberdeen Journals’ submission, we see no unfairness to Aberdeen Journals in the approach the Director has taken. Indeed, a criterion that “errs against a finding of predation” (paragraph 175 of the decision) may not strike an appropriate balance between the interests of the dominant firm, the maintenance of an effective competitive structure, and the interests of the undertaking predated against.

[387]Similarly we do not think that Aberdeen Journals can rely on the fact that the Director has not followed paragraph 4.6 of OFT 414 in this case since, had he done so, the result could only have been to bring into account more “variable” costs, to the detriment of Aberdeen Journals. The costs treated by the Director as variable

[388]Turning now more specifically to the costs that the Director has treated as “variable” costs, the Director has approached the matter at two levels.

[389]At the first level, the Director has taken only newsprint plus circulation costs as a proxy for average variable costs (paragraph 176 of the decision). Annex 2, Graph 6, of the decision shows that the monthly revenues of the Herald & Post exceeded newsprint and circulation costs on only a few occasions between July 1996 and April 1998, and remained well below newsprint and circulation costs for the whole period from April 1998 up to and including March 2000, although revenues exceeded newsprint and circulation costs in each month from April to July 2000 (paragraph 177 of the decision). Aberdeen Journals has not challenged the Director’s finding that in March 2000 its revenues were below newsprint and circulation costs.

[390]As already indicated there are, however, other costs which may properly be regarded as “variable” costs. By fax of 17 July 2000 the Director served a notice on Aberdeen Journals under section 26 of the 1998 Act requesting Aberdeen Journals , amongst other things, to allocate costs incurred under the headings editorial, advertising, newsprint and circula tion to either “fixed” or “variable”. Aberdeen Journals was asked to classify as “short run fixed 119 costs” costs which would still be incurred if the Herald & Post were not produced for a month. According to the fax of 17 July 2000:

“These are the short run fixed costs. All other costs are variable in the short run.”

[391]In response to that notice, Aberdeen Journals supplied certain figures on 1 August 2000, which figures were updated on 23 August 2000. Costs were allocated between “short run” fixed costs and variable costs with the following explanation:

“ Editorial Variable Costs include freelance photographers and reporters. Advertising Variable costs include staff incentives/bonuses, stationery and other office consumables, vehicle costs Fixed costs are salary and associated costs (employer’s NIC office equipment rental charges) of permanent employees. Newsprint Assumed 100% variable May and June’s figures include amounts related to printing at another location (Leicester). These are exceptional costs and have been deducted in the total variable cost calculation. Circulation Assumed all variable with the exception of the salary and associated costs of the permanent distribution employees. Other Production Charges Ink charges are based on the H&P’s newsprint consumption as a proportion of total tonnage consumed and this is also applied to the overall volumes of ink consumed. Plate charges are based on the number of mono, colour and blank plates used.”

[392]Apart from, apparently, rejecting Aberdeen Journals’ classification of part of the circulation costs as “fixed”, the Director appears to have accepted Aberdeen Journals’ figures without further verification.

[393]This part of the Director’s analysis seems to us to give rise to a difficulty, namely that the figures supplied by Aberdeen Journals appear to relate to costs that are variable within a month, rather than what the Director appeared to be seeking, namely an estimate of costs that are variable from month to month . Thus, on the figures supplied, editorial costs such as freelance photographers are treated as variable, while all other editorial costs are treated as fixed. Similarly, for advertising, the only elements that are treated as variable costs are staff bonuses, stationery, office consumables etc, rather than the salaries of the staff. 120

[394]However, if one examines the underlying figures it does appear that a different approach could have been taken. For example, on the Director’s approach, a large proportion of advertising costs (mainly salaries of sales staff) are classified as “fixed costs”. But if the volume of advertising carried by the Herald & Post is reduced from month to month, it is likely that a corresponding reduction will be made in the numbers employed in the advertising department, or that employees working on the Herald & Post will be switched to other titles. Depending on such matters as the terms of the relevant employment contracts, it seems to us that staff salaries in the advertising department could be regarded as “variable” over a relatively short timescale. That conclusion could be borne out by the figures supplied by Aberdeen Journals to the Director on 23 August 2000, relating to the period March to July 2000 (pages 653 to 657 of the bundle ). Those figures show that “fixed” advertis ing costs progressively fell, from £14,080 in March 2000 to £8,003 in July 2000, a decline of 43 per cent. In the same period, average weekly advertising volumes fell from 10,440 sccm to 5,620 sccm, a decline of 46 per cent. The apparently close correlation between these changes in volume and the changes in costs, suggests that almost all the advertising costs classified as “fixed” costs could be reclassified as a “variable” cost (see also, on this point, AKZO , cited above, at paragraphs 92 to 95, where the opposite situation prevailed). Similar considerations apply to editorial and circulation costs.

[395]If, as a matter of fact, costs such as advertising and editorial costs do vary with output, the whole amount of such costs are to be treated as “variable ” for the purposes of the AKZO test.

[396]Again, the Director’s apparently contrary approach appears to us to favour Aberdeen Journals in that, even on a “month by month” basis, a higher proportion of costs could have been classified as variable. However, since this issue has not been the subject of detailed argument, we make no definitive findings about it.

[397]We are also somewhat surprised that the Director accepted, apparently without question, the figure for the charge apparently made to Aberdeen Journals by the Leicester Mercury , another Northcliffe title, for printing the Herald & Post at Leicester in May and June 2000 (see paragraph 1.3 of Herbert Smith’s letter of 14 June 2000), referred to in paragraph 179 of the decision and the table thereto. Despite Mr Scott’s contention in his statement of 1 February 2001 that Aberdeen Journals paid “an appropriate commercial rate” we have seen no invoices or other data to support what appears to have been an intra-group transaction, and are therefore not in a position to know what exactly these charges covered. The OFT’s letter to the Registrar of 23 November 2001, at the bottom of p.2, suggests to us that the Director 121 was not wholly clear on this point either. The table to paragraph 179 of the decision suggests that some costs (e.g. overtime and pre-press) which would not have been incurred in May and June 2000 have in fact been allowed, whereas no adjustment has been made to the Leicester costs in respect of ink, plate and film charges that would otherwise have been incurred.

[398]Finally although further questioning from the Tribunal elicited certain additional heads of “variable cost”, (see paragraph 368 above), the question whether or what account should be taken of the labour costs of printing the Herald & Post (other than overtime and a small element of pre-press production costs) and whether such labour costs are at least partly variable, remains unresolved, since that issue was never investigated by the Director (see paragraphs 373 et seq above).

[399]These various considerations point to the need, in future cases, for more rigorous verification and analysis of costs figures submitted to the Director.

[400]Nonetheless, taking the figures used by the Director in the decision, the conclusion he reached as to the relationship between the Herald & Post ’s revenues and its variable costs in the period March to July 2000 can be summarised as follows, on the basis of the table at paragraph 179 of the decision: Herald & Post ’s variable costs versus income, 2000 (£) March April May June July Income 82,397 68,582 65,225 69,421 56,314 Variable costs 1 100,067 65,121 69,006 2 73,068 2 49,950 Surplus/deficit above/below variable costs (17,670) 3,461 (3,781) (3,647) 6,364 Surplus/deficit after deducting Leicester printing costs (17,670) 3,461 8,219 3,979 6,364 1 Treating all newsprint and circulation as variable, but allocating a large proportion of editorial and advertising to fixed costs in accordance with Appendix 5 to Herbert Smith’s letter of 1 August 2000; including Aberdeen Journals’ figures for printing costs at Leicester in May and June 2000 as “variable”, but including nothing for printing labour at Aberdeen in March, April and July 2000 other than overtime and production pre-press; and including other additional variable costs not included in the management accounts: see paragraph 179 of the decision. 2 Includes Leicester printing costs.

[401]The Director, in our view correctly, treated the claimed additional costs of printing at Leicester as variable costs (paragraph 179 of the decision). However, he accepted (at paragraph 205) that the threat of industrial action constituted “an objective justification” for 122 pricing below average variable cost in May and June 2000. In the result, the Director found an infringement of the Chapter II prohibition only in March 2000.

[402]As is apparent from the above, we consider that the Director’s approach on costs has been favourable to Aberdeen Journals. Nonetheless, for reasons of fairness we take the case as the Director has presented it in the decision, and deal with it on that basis. C. Analysis of Aberdeen Journals’ arguments

[403]Aberdeen Journals argues, essentially,(i) it did not price below average variable costs in March 2000;

(ii) alternatively, its prices were above average variable costs in the last week of March 2000;

(iii) there is no link between its conduct before 1 March 2000 and its conduct after that date, when Aberdeen Journals was seeking to restore the Herald & Post to profitability;

(iv) there is no relevant evidence of an intention to eliminate competition after 1 March 2000;

(v) there was no prospect of Aberdeen Journals distorting competition or eliminating the Independent after 1 March 2000;

(vi) there was no possibility of the Herald & Post recouping its losses after 1 March 2000;

(vii) the duration of the alleged conduct (one month) is too short to be regarded as abusive; and

(viii) the Independent was an inefficient entrant. In support of these submissions Aberdeen Journals also sent us a large bundle of US and Australian decisions. We deal with Aberdeen Journals’ arguments in turn. Was the Herald & Post sold at below average variable cost during March 2000?

[404]Aberdeen Journals did not challenge at any stage the Director’s finding that it had sold below average variable costs until this second appeal to the Tribunal. In a report now annexed to Aberdeen Journals’ application, RBB Economics argue that the Director should have taken a different approach to variable costs, on the basis that it would have been commercially logical to continue to publish the Herald & Post for a month, even without advertising content. Thus editorial costs, and newsprint and distribution costs associated with editorial content, should be treated as “fixed”.

[405]We reject this approach. In our view, the product being produced here is a free newspaper, with both editorial and advertising content. It is neither a free newspaper with no advertising content nor an advertising-only publication. In those circumstances, in our view the costs to be measured are not just those of the column inches of advertising, but those of the newspaper as a whole. 123

[406]Nor do we agree that in the circumstances of this case, it would be right to treat as “fixed” some part of the costs that would normally be regarded as variable on the grounds that it would be commercially logical to continue to publish the Herald & Post without advertising for a month. In our view, such a proposition is divorced from practical reality. In taking pricing decisions, it does not seem to us that the management of the Herald & Post would ask themselves what costs they could continue to incur if they published the Herald & Post without advertising for a month. For management purposes, most businesses seek to base their pricing decisions on estimates of total costs and variable (or marginal) costs as conventionally defined, rather than the hypothetical assumptions suggested by RBB Economics.

[407]We add that, in this case, as we understand it, the Director intended to take as “short run” variable costs those costs that varied with output from month to month. In order to do that, he could have simply asked for a breakdown of detailed headings of costs and identified which costs remained constant regardless of output, and which costs in fact varied with output. Instead of doing that, the Director in his letter of 17 July 2000 asked Aberdeen Journals to identify which costs would not be incurred if the Herald & Post was not produced for a month. This approach gave rise to the confusion as to whether the Director was asking about costs that were variable within the month, as distinct from costs that were variable from month to month , which we have already discussed (paragraphs 390 et seq above).

[408]However, the Director’s approach in the letter of 17 July 2000 was in our view merely a shorthand technique for obtaining the information he was seeking, namely which costs of producing the Herald & Post vary with output. The Director’s use of this shorthand technique did not in our view imply, and was not intended to imply, that it would be commercially realistic to close the Herald & Post for a month. As we have said, the argument of RBB Economics to the effect that such a scenario should be considered does not in our view reflect how most businesses would seek to identify variable costs.

[409]Similarly, we are unpersuaded by Professor Yamey’s argument that the Director should have investigated whether it would have been commercially rational to publish the Herald & Post at prices below average variable cost in March 2000, having regard to the consequences of a temporary closure. We can see that, in some circumstances it may be commercially sensible to maintain capacity even by selling at prices below average variable costs, for example during a recession (which is the example Professor Yamey gives). However, such hypothetical circumstances are far removed from the situation in the present case. In this 124 case the evidence is that the whole strategy pursued by the Herald & Post was to eliminate a competitor, namely the Independent . That is what gave rise to the Herald & Post being sold below average variable costs, not the effect of a recession or other extraneous circumstances.

[410]We are satisfied on the evidence that the Herald & Post was sold below average variable cost during March 2000. Was the Herald & Post sold at below variable cost in the last week of March 2000?

[411]In reliance on Appendix B to its first notice of appeal, Aberdeen Journals argues that its revenue for the last publication in March 2000, that of 29 March 2000, exceeded its variable costs by £221.

[412]Following further investigation, further items of cost, not taken into account in Appendix B to the first notice of appeal, have now been identified: see paragraphs 178 and 179 of the decision. Averaging the additional costs figures now provided by Aberdeen Journals over the five weeks of March 2000, the alleged “surplus” in the last week of March 2000 disappears, on our calculation. Since on that basis revenue was below variable costs in the last week of March 2000, we reject Aberdeen Journals’ argument.

[413]In any event, even if the figures had not been adjusted, the Director in our view acted reasonably to take average variable costs on a monthly, rather than a weekly basis. In this case, a month is likely to be the minimum period for which figures can meaningfully be presented for the purpose of taking management decisions. From the point of view of a competitor under attack from a dominant undertaking, it is cold comfort that the dominant undertaking may have managed by a fraction to have covered variable costs in one week when revenue remained below variable cost over the period of a month taken as a whole.

[414]We have already pointed out that cost rules should not be applied mechanistically, and that it should not be assumed that pricing above average variable cost but below average total cost represents a safe haven for a dominant undertaking. In our view, that is likely to be particularly so when a prolonged period of pricing below average variable costs is then followed by a period of pricing between average variable costs and total costs. As we have already said, such pricing by a dominant undertaking may well force an equally efficient competitor to exit the market. 125

[415]In those circumstances, even if it had been the case, we would not have regarded the fact that the Herald & Post had on certain assumptions marginally covered its average variable costs in the last week of March 2000 as having any material impact on the Director’s finding of abuse for the whole of that month. Did the conduct of the Herald & Post change materially before the end of March 2000?

[416]A principal part of Aberdeen Journals’ defence is that it was already taking steps before 1 March 2000 to put its house in order and that, by that date, any link with its conduct prior to 1 March 2000 had been broken.

[417]As already foreshadowed at paragraphs 248 to 252 above, we reject that submission.

[418]As regards, first, the reduction of the Herald & Post ’s distribution from over 120,000 to some 107,000 in October 1999, Herbert Smith’s letter of 4 April 2000 points out that such reduction affected outlying areas and had little effect on revenues. Equally, such efforts that we are told were made at this time to increase advertising rates seem to have had very little effect on the average advertising rates of the Herald & Post . The same is true of such efforts as may have been made in January and February 2000 to increase advertising rates. Prices continued to remain below average variable cost, and even below the cost of newsprint and distribution, throughout the period to the end of March 2000. Annex 2, Graph 7, of the decision does not seem to us to show any material change in advertising yields. Average rates appear to have been about £1.22 per sccm in January 2000, £1.18 per sccm in February 2000 and £1.26 per sccm in March 2000: see Annex 2, Graph 7 and page 483 of the bundle . Thus advertising yield in February 2000 was, in fact, marginally below that of January 2000 and the yield in March was only slightly higher. These rates may be compared with the rate of £3. 29 per sccm achieved by the Herald & Post i n the six months prior to April 1996 at the time of the launch of the Independent (page 481 of the bundle). Although admittedly slightly higher than the exceptionally low rates prevailing from October 1998 to July 1999 (less than £1 per sccm) advertising rates for the Herald & Post remained in our view far below normal competitive levels in the last quarter of 1999 and the first quarter of 2000, including the whole of March 2000.

[419]As to the other steps relied on by Aberdeen Journals, we are told that distribution was further reduced to 84,000 copies per week, but that reduction took effect only with the edition of 29 March 2000. Average distribution in March 2000 was still 102,600 (paragraph 169 of the decision). Pagination, we are told, was reduced to 56 pages per week, but again that was only 126 the edition of 29 March 2000. Pagination in March 2000 was 388 pages (paragraph 167 of the decision) which gives a weekly average of 78 pages during that (five-week) month, which is comparable to the number of pages being produced in each of the previous four months.

[420]As noted above, even the belated changes in distribution and pagination made in the edition of 29 March 2000 did not enable the Herald & Post to cover its average variable costs (on the basis of the revised figures in the decision) even for the last week of that month, let alone for the month of March 2000 as a whole. Finally, the move of the Herald & Post ’s organisation into the Aberdeen Journals’ head office in February 2000 would have had no effect on variable costs.

[421]It is true that a memo dated 10 March 2000 from Mr Davidson of Northcliffe to Mr Scott, Managing Director of Aberdeen Journals, about the profitability of Aberdeen Journals as a whole includes, relatively low down on a list of action points:

“Progress to break even on the Herald & Post ”

. However, that statement does not indicate to us the existence of any plan to raise the revenues of the Herald & Post above average variable costs prior to 1 March 2000, or even before the end of that month. As we understand it, a new business plan for the Herald & Post was not in fact produced until July 2000. Although by then the Herald & Post was, on the Director’s calculations, covering average variable costs, that business plan did not envisage the Herald & Post making a positive contribution in the current or following financial years, (page 539 of the bundle). The figures submitted by Aberdeen Journals in response to the first Rule 14 notice forecasted a positive monthly contribution on the basis of the management accounts from February 2001.

[422]The overall picture we have is that no effective action to bring the revenues of the Herald & Post above average variable costs was taken prior to the end of March 2000. In our view, there is no effective break, prior to the end of March 2000, with the chain of conduct that the Herald & Post had been following since mid-1996. In our view, the correct factual basis on which to approach this case is that of a continuous period of pricing below average variable costs which lasted without an effective break from mid-1996 to at least the end of March 2000. We so find. Did Aberdeen Journals intend to eliminate competition after 1 March 2000?

[423]It is clear from AKZO , at paragraph 71, and Tetra Pak II , at paragraph 41, cited above, that pricing below average variable cost by a dominant undertaking is presumed to be an abuse of a dominant position without the need to prove an intention to eliminate a competitor. In 127 those circumstances it was unnecessary for the Director to establish any specific eliminatory intent on the part of Aberdeen Journals during March 2000.

[424]That conclusion is, in our view, wholly consistent with the general thrust of the case law we have already cited, which is to the effect that an abuse of a dominant position is an objective concept which requires a dominant undertaking to observe a special responsibility not to distort competition by having recourse to methods that diverge from what is normal in competitive markets. As already indicated, in our view pricing below average variable cost is not a normal business practice in normal competitive conditions. Such a practice inevitably distorts competition when undertaken by a dominant undertaking. In our view, that suffices to establish the abuse alleged, without it being necessary to examine the question of intention.

[425]In any event, in our view an intent to distort competition in a way which threatened, or at least risked, the elimination of the Independent may properly be imputed to Aberdeen Journals during the month of March 2000.

[426]The documents cited in paragraph 181 of the decision, and set out at paragraph 48 above, plainly show that the original strategy of Aberdeen Journals was to eliminate the Independent , either by buying it, or by forcing it out of business. As late as January 2000 Mr Davidson, in his memo of 5 January 2000, asked Mr Ezzat to construct scenarios which included either buying the Independent or “denying the Independent all commercial oxygen”. Mr Ezzat’s review during January 2000 recommended the strategy of buying the Independent , in the meantime following a strategy which would “not allow the Independent to break even” and would “[maintain] the pressure on Barwell by attacking the Independent more aggressively.”

[427]It appears that Mr Ezzat’s memo, apparently dated 18 January 2000, which refers to “resolving the OFT queries”, was written after the Director’s letter of 10 January 2000 showing that the Independent ’s complaint was under active investigation. However that may be, it is surprising to find such remarks recorded in internal memos in circumstances where the OFT had already received a complaint from the Independent in July 1999. Aberdeen Journals had already been required to give undertakings to the Director on 23 December 1999 regarding allegations of bundling. The entry into force of the 1998 Act, on 1 March 2000, was only a short time away. 128

[428]Despite the fact that Mr Davidson met the OFT on 17 January 2000, Aberdeen Journals’ letter of 10 February 2000 (see paragraph 234 above) in response to the Director’s letter of 10 January 2000, does not evince any intention on the part of Aberdeen Journals to alter its pricing strategy, either before the entry into force of the 1998 Act, or at all. Instead that letter largely blamed Mr Barwell for the situation that had arisen, unfairly in our view, and sought a further meeting with the OFT.

[429]A further meeting with the OFT apparently took place on 24 February 2000, as a result of which the Director requested further information from Aberdeen Journals on 28 February 2000, having regard to Aberdeen Journals’ “continuing conduct” in relation to an allegation of predatory pricing. There is a letter of 3 March 2000 from Herbert Smith to the OFT promising a paper from Aberdeen Journals “showing the steps they have been and are taking to reduce losses at the Herald & Post and a plan showing what steps they propose to take with a view to enabling the Herald & Post to revert to profitability”, the plan to be submitted in draft to the OFT for comment.

[430]As noted above, a business plan for the Herald & Post for the year 2000/2001 was not prepared before July 2000, and even then did not foresee a rapid return to profitability. All that in fact occurred was that on 4 April 2000 Herbert Smith wrote to the OFT summarising the matters we have already considered at paragraphs 416 et seq above.

[431]In our view, if a dominant undertaking knowingly adopts a predatory pricing policy with the intention of eliminating a competitor, it will be presumed to have that intention for as long as it knowingly continues with that pricing policy, unless there is clear evidence that the dominant undertaking intends to cease that policy, and has taken active and effective steps to do so.

[432]In this case there is no evidence that Aberdeen Journals took any effective steps to comply with the Chapter II prohibition, prior to the entry into force of the 1998 Act on 1 March 2000. Such steps that were taken prior to that date, seem to have had no material effect on average yields, and further steps were not taken until the end of March 2000. There is, in our view, nothing to rebut the presumption that Aberdeen Journals knowingly allowed the Herald & Post to continue pricing far below costs during March 2000, and knowingly allowed a strategy originally conceived with the intention of eliminating the Independent to continue throughout that month. 129 Was there a distortion of competition or a threat to the Independent in March 2000?

[433]Aberdeen Journals argues that its conduct had no demonstrable effect on competition during March 2000, and posed no risk to the Independent , notably since, by then, Aberdeen Journals no longer had the intention, or potential ability, to eliminate the latter.

[434]We have already held above that during March 2000 Aberdeen Journals knowingly allowed the Herald & Post to continue with a strategy of pricing below cost that had been originally designed for the purpose of eliminating the Independent . Although, in the event, the Independent did not go out of business, it is not necessary to show that a competitor has actually gone out of business in order to establish the abuse of predatory pricing (see: Tetra Pak II , cited above, at paragraph 44). In this case it does not seem to us that one can discount the risk that the Independent might have ceased publication during March 2000, faced as it had been with predatory pricing for nearly four years.

[435]In any event, by continuing to price in a predatory way in March 2000, in our view Aberdeen Journals was distorting competition, hindering the maintenance of an effective competitive structure and hindering the growth of competition, by forcing the Independent to incur losses, and denying it business that it could otherwise have been expected to obtain. The purpose, or at least the effect, of that strategy was to continue to protect Aberdeen Journals’ dominant position in the market for local newspapers in Aberdeen during March 2000. In our view those effects on competition are amply sufficient to engage the Chapter II prohibition in accordance with the principles of Hoffman-La Roche and the subsequent case-law set out above. The issue of recoupment

[436]Aberdeen Journals argues that its conduct was not abusive during March 2000 because by that date it had no realistic possibility of recouping the losses it was incurring on the Herald & Post .

[437]The issue of recoupment arises from the assumption that the essential reason for selling at “predatory” prices, thereby incurring losses, is to eliminate a competitor. Once the competitor is eliminated, the dominant firm will then be able to recoup its losses by charging high prices. But, so the argument runs, if the alleged predator would not be able to recoup its losses by raising its prices after the competitor is eliminated, the alleged predation has, in fact, no detrimental effect. On the contrary, such conduct brings consumers the positive benefit of low prices with no risk of having to pay high prices later on. 130

[438]Recoupment is particularly significant in the context of US anti-trust law. For example, in Brooke Group v Brown & Williamson Tobacco 509 US 209 (1993) , a small tobacco company failed to establish that below-cost sales by a competitor with a market share of 12 per cent violated the relevant provisions of the Robinson-Patman Act, essentially because the competitor would have had no subsequent ability to raise prices above competitive levels, even if its strategy of selling below cost had been successful. However, it is important to bear in mind that US anti-trust legislation is drafted differently from the Chapter II prohibition. Hence, concepts developed in US law that enable the courts to distinguish competitive activit y which is beneficial to consumers from cases of monopolistic or quasi-monopolistic exploitation are not necessary transposable to the Chapter II prohibition.

[439]To take, for example, Brooke Group v Brown & Williamson Tobacco : although the alleged ‘predation’ harmed the plaintiff in that case, there was no detrimental effect to consumer welfare, only the benefit of low prices, because the defendant lacked any market power to recoup its losses by subsequently raising prices. However, a case such as Brooke Group v Brown & Williamson Tobacco would simply not arise under the Chapter II prohibition because the defendant, with a market share of 12 per cent, would not have had a dominant position.

[440]In paragraph 78 of his opinion in Tetra Pak II Advocate General Ruiz-Jarobo Colomer said at

[1996] ECR I-5983

“78. I do not consider it desirable that the Court of Justice should lay down the prospect of recouping losses as a new pre-requisite for establishing the existence of predatory pricing contrary to Article [82], for a number of reasons: — selling at a loss in order to eliminate a competitor would be suicidal if it were used by a dominant undertaking with no prospect of recouping the losses incurred; — the economic potential of the dominant undertaking and the weakening of competition on the dominated or related market will in principle ensure that losses are recouped; — proof of a prospect of recouping losses is difficult to define and requires complex market analyses, as is clear from the US Supreme Court’s own case-law; — recouping losses is the result sought by the dominant undertaking, but predatory pricing is itself anti-competitive regardless of whether it achieves that aim.”

[441]In its judgment in that case the Court said at paragraph 44:

“44. Furthermore, it would not be appropriate, in the circumstances of the present case, to require in addition proof that Tetra Pak had a realistic 131 chance of recouping its losses. It must be possible to penalise predatory pricing whenever there is a risk that competitors will be eliminated. The Court of First Instance found, at paragraphs 151 and 191 of its judgment, that there was such a risk in this case. The aim pursued, which is to maintain undistorted competition, rules out waiting until such a strategy leads to the actual elimination of competitors.”

[442]In his opinion in Compagnie Maritime Belge , cited above, Advocate General Fennelly said at paragraph 136:

“136. The sharing of loss of revenues prompts me to revert briefly to the possible need to establish an intention or a possibility of recoupment. The process of sharing revenue losses is in essence a form of recoupment. The strategic purpose of the fighting rates carries with it the unspoken implication that rates will not be reduced for any sailings, current or future, where that is not necessary to meet competition. Furthermore, once the competitor was eliminated, they would clearly no longer be justified. Thus, to the extent that it is necessary, I believe that the present case passes the test of recoupment. At the same time, I would say that some such requirement should be part of the test for abusively low pricing by dominant undertakings. It is implied in the first paragraph of the quotation from AKZO (see paragraph 126 above). It is inherent in the Hoffman-La Roche test (see paragraph 124 above). The reason for restraining dominant undertakings from seeking to hinder the maintenance of competition by, in particular, eliminating a competitor is that they would thus be enabled to charge abusively high prices. Thus, an inefficient monopoly would be reinstated and consumers would benefit only in the short run. If that result is not part of the dominant undertaking’s strategy it is probably engaged in normal competition.”

[443]We do not read Mr Fennelly’s opinion in Compagnie Maritime Belge as throwing any doubt on the Court’s judgment in Tetra Pak II , at paragraph 44, that there is no need to prove the possibility of recoupment whenever there is a risk that competitors will be eliminated.

[444]We have already found that the risk that the Independent might be eliminated existed in March 2000, even if it did not eventuate. Aberdeen Journals’ whole pricing policy as regards the Herald & Post had been conceived with the strategic purpose of eliminating the Independent from the market. In our view, as long as that policy continued, there was a risk that the Independent would be eliminated. Looking at the situation as at 1 March 2000, no-one could have been certain how long the Independent could survive.

[445]Moreover, as long as the Herald & Post was continuing to be priced below cost, Aberdeen Journals was protecting the market share and revenue of the Evening Express , denying market share and revenue to the Independent , and establishing a reputation likely to deter other would-be entrants. As the Tribunal has already pointed out in Napp , cited above, 132 predatory pricing may be a rational policy by a dominant undertaking when it is pursued in one sector in order to protect profits and market share in another sector. That, in our view, is a form of recoupment. In the circumstances of this case, we would not expect the Director to adduce any further evidence of “recoupment” in order to prove an abuse of a dominant position.

[446]Finally, we point out that in this case it is principally the intervention of the authorities that caused Aberdeen Journals to modify its pricing polices from April 2000 onwards. Had the Director not intervened, there is no reason to suppose that Aberdeen Journals would not have continued with the tactics outlined in Mr Davidson’s memo of 5 January 2000 and Mr Ezzat’s memorandum in reply. In our view it is not permissible for a dominant undertaking to invoke an argument based on lack of “recoupment” where the alleged inability to recoup arises, not because of lack of market power, but because the Director has intervened to put a stop to the abuse in question. Is the period of the predation too short to constitute an abuse?

[447]Aberdeen Journals argues that a period of pricing below average variable costs for one month is too short to constitute an abuse for the purposes of the Chapter II prohibition.

[448]We can see that if, from time to time, the prices of a dominant undertaking fall below average variable or average total costs for a short period, for extraneous reasons, with no suggestion that the dominant undertaking has been targeting a competitor, it may be possible, in some circumstances, for the dominant undertaking to rebut the presumption of abuse. That, however, is not what has happened in this case.

[449]As we have already found, we do not consider that anything in the conduct of Aberdeen Journals broke the link between its conduct in the period before 1 March 2000 and its conduct during March 2000. In our view the facts of this case present a continuum, lasting from mid-1996 to at least the end of March 2000, in which the Herald & Post pursued a policy of pricing below cost. The true period of predation is in our view not one month but nearly four years. It is true that, legally speaking, Aberdeen Journals’ conduct was caught by the Chapter II prohibition only for the last month of that period. But the coming into force of the 1998 Act does not affect the economic analysis of the underlying conduct. We see no reason why the Chapter II prohibition should not apply, as from 1 March 2000, to the last month of a continuous period of nearly four years of pricing below average variable costs, particularly in 133 circumstances where Aberdeen Journals signally failed to put its house in order before the coming into force of the 1998 Act, despite having ample time to do so. Was the Independent an in efficient entrant?

[450]For the reasons already given at paragraphs 194 et seq above, there is in our view no evidence that the Independent was an inefficient entrant, nor a “fireship”, such as might give rise to a possible “objective justification” for the conduct of Aberdeen Journals. US and Australian cases

[451]While, of course, decisions in other anti-trust jurisdictions are of general interest to the Tribunal, for the reasons already given (paragraphs 438 and 439 above) caution is needed in seeking to transpose decisions from the United States of America to cases arising under the Chapter II prohibition or Article 82 of the Treaty, because the conceptual basis of the legislation is not the same. Although section 46(1) of the Australian Trade Practices Act 1974 is similar in concept to aspects of the Chapter II prohibition, Aberdeen Journals has not explained to us the relevance to its arguments of any Australian decisions. D. Conclusion

[452]For the foregoing reasons we conclude that the Director has established that Aberdeen Journals abused a dominant position in the supply of advertising in local newspapers in Aberdeen by pricing the Herald & Post below average variable costs during the month of March 2000. VII EFFECT ON TRADE WITHIN THE UNITED KINGDOM A. Arguments of the parties

[453]As regards “effect on trade”, Aberdeen Journals argues that section 60 of the 1998 Act obliges the Director to prove a causal connection between the actual abuse and a material effect on trade within the United Kingdom. However, according to Aberdeen Journals, the Director does not link Aberdeen Journals’ failure to cover average variable costs by £17,670 in March 2000 to any resultant harm to the Independent (which was in fact benefiting from new business from advertisers who were not prepared to pay the increased rates in the Herald & Post ), or to any effect on customers, third parties, or otherwise. Moreover, the Director’s assertion at paragraph 211 of the decision that it acquired a reputation for predatory 134 behaviour is unfounded since the chain of conduct was indisputably broken after 1 March 2000. In any event, this latter argument was not raised by the Director in the administrative procedure.

[454]The Director notes that Article 82 of the EC Treaty provides that an abuse of a dominant position is prohibited “in so far as it may affect trade between Member States”, whereas the Chapter II prohibition provides that an abuse is prohibited “if it may affect trade within the United Kingdom”. According to the Director this is a “relevant difference” for the purposes of section 60(1): there is no requirement under section 60(2) of the 1998 Act to show that an abuse has an appreciable effect on trade within the United Kingdom, in a manner analogous to the test under EC law. It is a purely jurisdictional test to establish that the conduct concerned has some effect in the United Kingdom, rather than a substantive competition test. As a result the threshold for satisfying it is low.

[455]According to the Director, pricing below average variable costs will inevitably affect the business of a competitor. For example, the effect which Aberdeen Journals’ conduct had on the Independent’s business is provided by Aberdeen Journals’ own assertion that, when it did finally take steps to increase rates and decrease pagination and distribution, it lost customers to the Independent . Aberdeen Journals’ conduct in March 2000 therefore had an effect on trade in the United Kingdom. In any event, as set out at paragraph 210 of the decision, Aberdeen Journals’ conduct also had the potential to eliminate the Independent as a viable competitor on the market. Pricing below average variable costs was also capable of creating or enhancing a reputation for fighting entry, and therefore of deterring market entry by third parties. Contrary to Aberdeen Journals’ assertion, this reputational effect was raised in the administrative procedure. B. The Tribunal’s findings

[456]We reject Aberdeen Journals’ submission that pricing the Herald & Post below average variable cost to the extent of £17,670 (see paragraph 179 of the decision) in March 2000 is de minimis for the purposes of the application of the Chapter II prohibition.

[457]First, pricing below average variable cost, even for a short period, can have a significant effect on competitors. In the present case, as we have already indicated, it is not a question of an isolated period of one month, but the continuation of a prolonged period of below-cost pricing lasting nearly four years. Even though pricing the Herald & Post below average variable cost was not illegal prior to 1 March 2000, the fact that such pricing took place 135 means that the impact of predation in March 2000 was greater than it would otherwise have been. In addition, the figure of £17,670, which represents the extent to which the Herald & Post ’s revenue fell below variable costs as defined by the Director, does not in our view represent the full competitive impact of the pricing policy of the Herald & Post . As paragraph 174 of the decision states, the negative contribution of the Herald & Post in March 2000 was £48,038, representing 58 per cent of revenue. Even that figure takes no account of the printing costs of the Herald & Post , as we have already pointed out, and no contribution to Aberdeen Journals’ general overheads. Moreover, the period after March 2000 is then followed by a prolonged period when the Herald & Post remains unprofitable (see paragraphs 364 et seq above) thus delaying the Independent ’s recovery from the consequences of the predation suffered in that month.

[458]Had Aberdeen Journals not priced at a predatory level in March 2000, the Independent would have incurred fewer losses, and would have gained significant business. On all those facts, it seems to us that the effect on trade within the United Kingdom is plainly established.

[459]More generally, we are not satisfied that we should read into the statutory wording of section 18(1) of the 1998 Act a requirement that the effect on trade should be appreciable. It is true that, ever since the decision of the Court of Justice in Case 5/69 Volk v Vervaecke [1969] ECR 295, it has been the rule that the prohibition of Article 81 of the EC Treaty applies only if there is an “appreciable” effect on competition and trade between Member States: see also Case 22/71 Béguelin v Commission [1971] ECR 949. The requirement that there should be an “appreciable ” effect on inter-state trade is, however, largely understood as a jurisdictional requirement which demarcates the boundary line between the application of Community competition law and national competition law: see e.g. Cases C-215/96 and C-216/96 Bagnasco v Banco Popolare di Novara

[1999] ECR I-135

, a case under Article s 81 and 82, and Case 22/78 Hugin v Commission [1979] ECR 1869, a case under Article 82.

[460]We accept the Director’s submission that, since we are already dealing, under domestic law, with conduct which takes place within the United Kingdom, there is no need to import into section 18(1) of the 1998 Act the rule of “appreciability” under Community law, the essential purpose of which is to demarcate the fields of Community law and domestic law respectively. In terms of section 60(1) of the 1998 Act, that seems to us to be a “relevant difference” between the 1998 Act and the provisions of Community law. 136

[461]The conclusion that there is no de minimis rule to be applied under section 18(1) is in our view supported by the fact that section 40 of the 1998 Act enables the Secretary of State to prescribe “conduct of minor significance” in relation to which a person is immune from any penalty that might be imposed for breach of the Chapter II prohibition under section 36(2). Such conduct has been prescribed by the Competition Act 1998 (Small Agreements and Conduct of Minor Significance) Regulations 2000, S.I. 2000 no. 262. Essentially those Regulations take conduct by an undertaking with an applicable turnover of less than £50 million in the previous business year outside the scope of the Chapter II prohibition. Aberdeen Journals has – in our view rightly – not contested the Director’s view that “the undertaking” for the purposes of those Regulations is the economic entity of which Aberdeen Journals ultimately forms part, namely not only Northcliffe with an annual turnover (not consolidated with its subsidiaries) of £59.3 million in the year ended 30 October 1999, but also Daily Mail & General Holdings Limited with an annual turnover of £1,620 million in the year ended 3 October 1999. An undertaking which does not benefit from section 40 cannot, in our view, claim to benefit from a further de minimis rule to be read into section 18(1).

[462]It follows, in effect, that conduct which(i) amounts to an abuse under section 18(1) of the 1998 Act, and

(ii) does not benefit from the Regulations made under section 40, will infringe the Chapter II prohibition unless it can be shown either that the only effect of that conduct is on trade outside the United Kingdom or, perhaps, that what is affected by the conduct is not “trade” within the meaning of section 18(1). Neither of those exceptions applies here.

[463]It follows that the Director has duly established the infringement of the Chapter II prohibition set out in the decision. VIII THE PENALTY A. Arguments of the parties

[464]Aberdeen Journals denies that any infringement of the Chapter II prohibition it may have committed was done so intentionally or negligently. According to Aberdeen Journals, the fact of committing an infringement negligently is a mitigating factor: see OFT 423, The Director General of Fair Trading’s Guidance as to the Appropriate Amount of a Penalty (“the Guidance ”). However, the decision does not address the distinction between intention and negligence. 137

[465]In the present case, submits Aberdeen Journals , it could not be said that its actions were intentional. During March 2000, there was no effect on competition, the Herald & Post had no prospect of recoupment and its conduct could not be said to be “deliberate”. The only evidence in relation to the “deliberateness” of the Herald & Post ’s behaviour are certain memoranda pre-dating 1 March 2000, which are irrelevant.

[466]As regards negligence, Aberdeen Journals refers to the guidelines entitled Enforcement , OFT 407, at paragraph 46, and states that it was taking drastic steps in the period before and during 1 to 29 March 2000 to revert the Herald & Post to profitability, and had no current intention to eliminate the Independent . There is no reason why Aberdeen Journals “ought to have known” that it was committing an infringement in March 2000.

[467]Moreover, the Director has imposed a starting point of 10 per cent on the basis of a statement in paragraph 2.4 of his Guidance , without any analysis of whether Aberdeen Journals’ allegedly predatory conduct had, or was likely to have, a “particularly serious” impact. According to Aberdeen Journals, the Director was wrong to take 10 per cent as an automatic starting point without further thought or explanation.

[468]Moreover, in accordance with paragraph 4.3 of the Guidance the Director should have analysed the nature of the titles included in the market definition, the structure of the relevant market, the market share of Aberdeen Journals and Aberdeen Independent, the conditions of entry, and the effect of the Herald & Post ’s activities on advertisers, all in the relevant period 1 to 29 March 2000.

[469]In any event, Aberdeen Journals considers that the penalty is entirely disproportionate to Aberdeen Journals’ failure to cover average variable costs in the period 1 to 29 March by £17,670. The duration of the infringement was only three weeks at the maximum; the only influence the Herald & Post exerted on the market was to ask its advertisers to pay increased advertising rates; Aberdeen Journals derived no profit from the alleged practices; and the products concerned were four editions of a free weekly newspaper. The chain of conduct was indisputably broken after 1 March 2000, Northcliffe was not “intimately” involved in any abuse in the period 1 to 29 March 2000 as shown by Mr Davidson’s memorandum to Mr Scott of 10 March 2000. According to Aberdeen Journals, there is no justification for applying a factor of four to the Director’s starting percentage which is, in any event, too high. 138

[470]Finally, the Director should have reduced the penalty by more than he already has to take account of mitigating factors. Aberdeen Journals actively assisted the OFT in its investigation by disclosing in good faith all background papers relevant to the activities in question. Moreover, the memoranda provided to the OFT contributed materially to the investigation and constitute the principal evidence of the existence of the infringement.

[471]The Director submits that, in spite of the existence of an Office investigation since July 1999, and the Director’s letter requesting detailed information of 10 January 2000, in spite of the benefit of a lengthy transition period, and in spite of the knowledge that pricing below average variable costs would be very likely to constitute a breach of the Chapter II prohibition, Aberdeen Journals failed to take meaningful action to increase its average revenue until after the entry into force of the 1998 Act on 1 March 2000.

[472]According to the Director, his Guidance is not a binding legal code. Section 38(3) of the 1998 Act only requires the Director to “have regard to” guidance which has been published by him in accordance with section 38(1). Furthermore, according to that Guidance , there is no need to consider the distinction between intentional and negligent conduct until the issue of mitigation arises (if at all).

[473]According to the Director, the starting point of 10 per cent for calculating the penalty is justified in this case by the fact that predation is a very serious infringement. The Independent was Aberdeen Journals’ only direct competitor. Had Aberdeen Journals been successful, it would have obtained a monopoly. As regards the “limited duration of the infringement”, the Director took specific account of the fact that Aberdeen Journals took steps to cease its infringement under the heading of “adjustment for mitigating factors”. It would be inappropriate to give credit for this factor twice over.

[474]The fourfold increase of the penalty level was not disproportionate, but was intended to ensure that the penalty “acts as an adequate deterrent to predation to this undertaking, to the broader newspaper publishing industry, and more generally” (paragraph 224 of the decision). The level of deterrence is justified on the particular facts of the case by virtue of(i) the size of the annual turnover of Aberdeen Journals (£33.9 million), Northcliffe (£59.3 million) and the Daily Mail Group (£1,620 million);

(ii) the fact that the management of Northcliffe was intimately involved in Aberdeen Journals’ predatory conduct, but took no effective steps to ensure compliance with the 1998 Act before 1 March 2000; and

(iii) the fact that Northcliffe 139 publishes over 50 separate regional titles. The acquisition of a reputation for predation by Northcliffe could have far-reaching adverse effects on competition in several markets.

[475]With regard to other mitigating factors, the Director argues that, although Aberdeen Journals did cooperate throughout the investigation, it nevertheless put forward vigorous arguments to defend its conduct. Furthermore, had it not co-operated, the relevant information could have been obtained by the Director using his enforcement powers. Contrary to Aberdeen Journals’ suggestion that the Director had not considered whether Aberdeen Journals acted “negligently” as opposed to “intentionally”, in paragraph 218 of the decision the Director finds that Aberdeen Journals’ conduct was engaged in deliberately. B. The Tribunal’s findings The duration to be considered

[476]We observe, first, that at paragraph 214 of the decision the Director finds that Aberdeen Journals has “provided evidence to the Director that it has ceased predating”. At paragraph 225 the “rapid steps to cease its infringement” taken by Aberdeen Journals are found to be a mitigating factor. We have already indic ated our doubts about certain aspects of the figures relied on by the Director, and our view that pricing below average total cost may also constitute an abuse of a dominant position contrary to the Chapter II prohibition (paragraphs 361 et seq above). We note that in this case the revenues of the Herald & Post remained well below total cost, and its contribution was negative, for a considerable period after March 2000. Average advertising yields of the Herald & Post actually fell in June and July 2000 from the level reached in May, (see Annex 2, Graph 7).

[477]In this judgment, however, we make no finding, one way or another, as to any period after March 2000, and we deal with this case strictly on the basis found by the Director, namely an infringement of one month’s duration in March 2000. We also remind ourselves that Aberdeen Journals’ conduct prior to 1 March 2000 was not prohibited by the 1998 Act. The penalty imposed

[478]Section 36(2), (3) and (8) of the 1998 Act provide:

“36. (2) On making a decision that conduct has infringed the Chapter II prohibition, the Director may require the undertaking concerned to pay him a penalty in respect of the infringement. 140 (3) The Director may impose a penalty on an undertaking under subsection (1) or (2) only if he is satisfied that the infringement has been committed intentionally or negligently by the undertaking. (8) No penalty fixed by the Director under this section may exceed 10% of the turnover of the undertaking (determined in accordance with such provisions as may be specified in an order made by the Secretary of State).”

[479]An order has been made under section 36(8): the Competition Act 1998 (Determination of Turnover for Penalties) Order 2000: S.I. 2000 no. 309. Nothing turns on the provisions of that Order in this appeal. A penalty imposed under section 36 is recoverable as a civil debt: section 37.

[480]Section 38(1), (4) and (8) of the 1998 Act provide:

“38.–(1) The Director must prepare and publish guidance as to the appropriate amount of any penalty under this Part. (4) No guidance is to be published under this section without the approval of the Secretary of State. (8) When setting the amount of a penalty under this Part, the Director must have regard to the guidance for the time being in force under this section.”

[481]The Director’s Guidance , already referred to, is published under section 38.

[482]In this case, at paragraphs 217 to 226 of the decision the Director has calculated the penalty by following the various steps set out in the Guidance . At Step 1 (paragraphs 2.3 to 2.6 of the Guidance ) he first takes the combined turnover of the Evening Express and the Herald & Post in the supply of advertising space in the previous business year (£4,743,000) and then applies the figure of 10 per cent to that amount on the ground that the infringement is very serious (making £474,300). At Step 2, adjustment for duration, (paragraph 2.7 of the Guidance ), the Director has not increased the penalty because the abuse lasted less than one year. At Step 3, adjustment for other factors, (paragraphs 2.8 to 2.9 of the Guidance ) the Director has applied a multiple of four to the figure of £474,300, essentially to ensure that the penalty acts as a deterrent to the broader newspaper publishing industry and more generally (paragraph 224). Under Step 4, adjustment for further aggravating and mitigating factors, (paragraphs 2.10 to 2.12 of the Guidance ), the resulting amount of £1,897,200 has then been reduced by 10 per cent because Aberdeen Journals co-operated fully throughout the investigation and also by a further 20 per cent because Aberdeen Journals took “rapid steps” to end the infringement (paragraph 225). Those calculations give rise to the penalty imposed of £1,328,040. 141

[483]That penalty represents approximately 4 per cent of the turnover of Aberdeen Journals, 2.2 per cent of the turnover of Northcliffe and 0.08 per cent of the turnover of Daily Mail & General Trust. Intentionally or negligently

[484]It follows from section 36(3) of the 1998 Act that, as a threshold matter, a penalty may be imposed only if the Director (and on appeal the Tribunal) is satisfied that the infringement was committed intentionally or negligently. However, as the Tribunal has already held at [453] to [455] of Napp , cited above, in order to impose a penalty the Director does not have to decide whether the infringement was committed intentionally or negligently, so long as he is satisfied that the infringement is either intentional or negligent. The question of whether the infringement was intentional or negligent goes, at most, to mitigation.

[485]As to the meaning of “intentionally” and “negligently”, at [456] and [457] of Napp the Tribunal said: “456. As to the meaning of “intentionally” in section 36(3), in our judgment an infringement is committed intentionally for the purposes of the Act if the undertaking must have been aware that its conduct was of such a nature as to encourage a restriction or distortion of competition: see Musique Diffusion Français , and Parker Pen , cited above. It is sufficient that the undertaking could not have been unaware that its conduct had the object or would have the effect of restricting competition, without it being necessary to show that the undertaking also knew that it was infringing the Chapter I or Chapter II prohibition: see BPB Industries and British Gypsum , cited above, at paragraph 165 of the judgment, and

Case T-29/92

SPO and Others v Commission

[1995] ECR II-289

, at paragraph 356. While in some cases the undertaking’s intention will be confirmed by internal documents, in our judgment, and in the absence of any evidence to the contrary, the fact that certain consequences are plainly foreseeable is an element from which the requisite intention may be inferred. If, therefore, a dominant undertaking pursues a certain policy which in fact has, or would foreseeably have, an anti-competitive effect, it may be legitimate to infer that it is acting “intentionally” for the purposes of section 36(3). 457. As to “negligently”, there appears to be little discussion of this concept in the case law of the European Community. In our judgment an infringement is committed negligently for the purposes of section 36(3) if the undertaking ought to have known that its conduct would result in a restriction or distortion of competition: see United Brands v Commission , cited above, at paragraphs 298 to 301 of the judgment. For the purposes of the present case, however, we do not need to decide precisely where the concept of “negligently” shades into the concept of “intentionally” for the purposes of section 36(3), nor attempt an exhaustive judicial interpretation of either term.” 142

[486]In the decision, the Director found that the infringement was “deliberate” (paragraph 218) which we take to be a finding that the infringement was committed intentionally, although at paragraph 208 he characterises the infringement as “at best, negligent”. In our view, as we have already held, the infringement is correctly characterised as “intentional” during the month of March 2000. Even if, under the pressure of the Director’s investigation, some attempts to begin to restore the Herald & Post towards at least “break even” were undertaken prior to March 2000, Aberdeen Journals must have known that its prices were far below cost during March 2000. Aberdeen Journals thus knowingly continued with conduct that was highly likely to have an anti-competitive effect during March 2000.

[487]In our view the position would be different only if Aberdeen Journals could demonstrate that, prior to the coming into force of the Chapter II prohibition, it had put into effect a clear plan to cease pricing below normal competitive levels. In those circumstances the infringement might arguably be characterised as “negligent” rather than “intentional”. However, that is not the case here. There is no evidence that Aberdeen Journals had any plan to cease pricing far below competitive levels before 1 March 2000, and no measures that might have led to that result were instigated before the end of the month.

[488]Accordingly we hold that the Director had jurisdiction to impose a penalty in this case, since the infringement was intentional. That would still be so even if he had found that Aberdeen Journals’ infringement was negligent, rather than intentional. The gravity of the infringement

[489]As the Tribunal pointed out in Napp , cited above, at [497] to [499], the Tribunal is not bound by the Director’s Guidance . The duty of the Tribunal is to make its own assessment of the amount of the penalty, subject only to the overall cap imposed by the Competition Act 1998 (Determination of Turnover for Penalties) Order, already cited. The Tribunal adopts a “broad brush” approach, with each case dependent on its own circumstances: see Napp at [535].

[490]As regards predatory pricing, the Tribunal said in Napp at [518] to [519]:

“518. We agree with the Director that predatory pricing, even of short duration, falls into the category of a serious abuse. Although it may, at first sight, seem anomalous that the application of competition law should result in higher, rather than lower prices, the present case vividly illustrates that the reason for predatory pricing is typically to exclude or neutralise competitors with a view to maintaining market share and/or high prices in sectors that would otherwise be threatened by competition. The “benefit” that some consumers (in this case hospital purchasing 143 authorities) receive from below-cost predatory prices is wholly outweighed by the “disbenefit”, in terms of high costs and lack of choice, which flows from the monopoly (in this case in the community segment) that the predatory pricing is designed to protect or strengthen. Unless predatory pricing, and especially pricing below average variable cost, by dominant undertakings is rigorously penalised by competition law, new competitive entry may be thwarted, with the result that consumers never receive the benefit of competitive conditions, nor the lower long-run price levels, wider choice and better quality which, in general, competition brings. 519. We therefore agree with the Director’s view, at paragraph 2.4 of his Guidance that predatory pricing by a dominant undertaking is one of the most serious infringements of the Act.”

[491]In our view the predatory pricing which occurred in this case constitutes a serious infringement of the Chapter II prohibition. It is particularly serious because it was practised by an incumbent monopolist against its sole competitor. It is probable in this case that, had it not been for the determination of Mr Barwell and the intervention of the Director, the Independent would have been driven out of business, allowing Aberdeen Journals to maintain its monopoly on local newspaper advertising in Aberdeen.

[492]In Napp at [502], the Tribunal said that the penalty imposed “must be such as to constitute a serious and effective deterrent, both to the undertaking concerned and to other undertakings tempted to engage in similar conduct. The policy objectives of the Act will not be achieved unless this Tribunal is prepared to uphold severe penalties for serious infringements.” We thus agree with the Director that an important purpose of the penalty imposed in this case is not only to deter Aberdeen Journals but also to deter others, both in the newspaper industry and other sectors of the economy from similar behaviour. With that end in view, the Tribunal accepts that a substantial penalty is warranted in the present case.

[493]We think a substantial penalty is warranted despite the short duration of the infringement found in this case. In addition to the intrinsic seriousness of predatory pricing, as the Tribunal pointed out at [509] of Napp , cited above, once a period of predatory pricing has terminated, it still takes time for the market to return to normal. In this case, the competitive structure in the relevant market remained in a weakened state, to say the least, for a prolonged period, even if the Director did not find an infringement after March 2000.

[494]In our judgment, the penalty should not differ according to whether the market in which Aberdeen Journals is dominant is taken to be local advertising in both paid-for and free newspapers in Aberdeen, or in paid-for newspapers only. Even if the relevant product market 144 were to comprise only the Evening Express , we would think it right to include, for the purpose of calculating the penalty, not only the turnover of the Evening Express but also the turnover of the “fighting title” defending it, namely the Herald & Post .

[495]As regards mitigating factors, the steps taken by Aberdeen Journals to bring the infringement to an end, and its co-operation with the Director’s investigation, have already been taken into account by the Director, and led him to reduce the penalty that he would otherwise have imposed by some 30 per cent. Aberdeen Journals’ co-operation with the Director does not seem to us to have gone beyond the normal: see Napp at [527]. Our own analysis does not suggest any further mitigating factors under these heads.

[496]In the course of this appeal Aberdeen Journals has not acknowledged that any criticism could be made of its conduct, but has strenuously contested almost every point in the case, despite the inferences that are plainly to be drawn from its own internal documents. The Tribunal would be prepared to take into account, even at the appeal stage, any mitigation that arose from the way in which an appeal had been conducted, but no such mitigation arises in this case.

[497]We note, however, that in this case the starting point for the Director’s calculations under Step 1 was at the top end of the scale – 10 per cent of Aberdeen Journals’ relevant turnover – multiplied by four for deterrence under Step 3. Although Aberdeen Journals knowingly allowed its infringement of the Chapter II prohibition to continue throughout March 2000, we are prepared to accept that Aberdeen Journals probably realised, shortly before 1 March 2000, that its aggressive campaign to eliminate the Independent would have to be modified. Although the Herald & Post ’s prices remained far below normal competitive levels at the end of March 2000, at least something was done to reduce distribution and pagination from the end of March 2000 onwards. The result is a finding of infringement, albeit serious, of one month. In those circumstances, in our view the Director pitched his calculations under Steps 1 and 3 of his Guidance , on the high side.

[498]We also note that Step 2 of the Director’s Guidance permits the Director to increase the starting point under Step 1 to take into account the duration of the infringement, in particular where the infringement ha s lasted more than one year. However, there is no comparable provision in the Guidance , at least explicitly, enabling the Director to take into account a duration of less than one year. Although in this case the short duration may be partly taken into account, indirectly, in the reduction of the penalty for mitigating factors made by the 145 Director under Step 4, we think some more explicit recognition should be made of the fact that the infringement in this case, as found by the Director, lasted only for one month.

[499]In all those circumstances, and looking at the matter in the round, we think it right to reduce the penalty imposed on Aberdeen Journals to £1,000,000, a reduction of just under 25 per cent.

[500]There will be interest on the penalty pursuant to Rule 27 of the Tribunal’s Rules from such date and at such rate as the Tribunal may determine in default of agreement between the parties: see Napp v Director General of Fair Trading (interest and costs)

[2002] CAT 3

[2002] CompAR 160. We will also deal with any applications for costs. IX CONCLUSION

[501]On those grounds the Tribunal unanimously decides:

[1]The penalty to be paid by Aberdeen Journals to the Director is fixed at £1,000,000.

[2]Save as aforesaid, Aberdeen Journals’ application to the Tribunal dated 18 November 2002 is dismissed. Christopher Bellamy Andrew Bain Patricia Quigley Delivered in open court Charles Dhanowa 23 June 2003 Registrar 146

Cited in 21 later judgments

and 12 more of the most senior