COMPETITION TRIBUNAL OF SOUTH AFRICA
Case No: 48/CR/Jun04
In the matter between:
MANDLAMATLA PUBLISHING (PTY) LTD Complainant
And
INDEPENDENT NEWSPAPERS (PTY) LTD Respondent
Panel : D H Lewis (Presiding Member), Y Carrim (Tribunal
Member), and M Madlanga (Tribunal Member)
Heard on : 57 June 2006, 2628 July 2006, and 11 September 2006
Decided on : 6 November 2006
Reasons for Decision
Decision
1] On 57 June 2006, 2628 July 2006 and 11 September 2006, the Tribunal
heard the application brought by MandlaMatla Publishing (Pty) Ltd against
Independent Newspapers (Pty) Ltd. The application is dismissed. The reasons
follow.
Background
2] This is a complaint brought by MandlaMatla Publishing (Pty) Ltd (henceforth
“MM”) against Independent Newspapers (Pty) Ltd (“IN”). 1
1 Independent Newspapers conducts business in KwaZulu Natal as Independent Newspapers
KwaZulu Natal.
3] Before 1 April 1987 the isiZulu language newspaper called Ilanga was owned
and published by Natal Newspapers (Pty) Ltd, the predecessor in title to the
respondent, IN. Natal Newspapers published Ilanga, printed it on its presses
and distributed it throughout the province of KwaZulu Natal (“KZN”).
4] Ilanga was founded in 1903 and for the following century was the only mass
circulation isiZulu language newspaper in the country. It is published and
distributed twice a week on Monday and Thursday. 2
5] On 15 April, with effect from 1 April 1987, the complainant, MM, purchased the
right, title and interest in and to Ilanga from Natal Newspapers. 3 Also on 15
April 1987 the complainant concluded a service agreement with Natal
Newspapers in terms of which Natal Newspapers was to provide certain
services to the complainant, with effect from 1 April 1987. 4 In terms of the
service agreement Natal Newspapers was to provide services to MandlaMatla
in relation to Ilanga which included: 5
[5.1] printing;
[5.2] distribution;
[5.3] sale of advertising; and
[5.4] administration of the editorial department
6] Further, the service agreement contained the following material provisions:
[6.1] all matters of editorial policy would be determined and decided by MM; 6
2 Transcript p310, line 15.
3 Sale agreement, joint bundle p2 clauses 3 and 4.
4 Service agreement, joint bundle page 16.
5 Service agreement, joint bundle page 16 clause 3.
6 Service agreement, joint bundle page 16 clause 4.1.
2
[6.2] the agreement would continue for 15 (fifteen) years provided that Natal
Newspapers, or its successors in title, would be entitled at any time to
terminate the agreement on six months written notice to MM; 7
[6.3] MM would pay Natal Newspapers a service fee of 60% of the annual gross
profit of Ilanga excluding editorial costs, before taxation and after all costs incurred by
Natal Newspapers in its production of Ilanga;8
[6.4] during the term of the agreement Natal Newspapers would not publish
an isiZulu language newspaper in KZN without the prior consent of MM. 9
7] On 28 July 1998 the parties concluded an amending agreement to alter the
60/40 split in costs to a 50/50 split. 10
8] For approximately 15 years Natal Newspapers published and distributed llanga.
The service agreement was due to expire on 31 March 2002. During the
currency of the service agreement, the respondent, IN, succeeded to the rights
of Natal Newspapers as embodied in the service agreement.
9] During the currency of the service agreement with MM, IN utilized, for the
purposes of distributing Ilanga, a network comprising four categories of
distributors. Firstly, there were direct retail outlets or ‘agents’ to which IN
directly distributed newspapers. These included department stores,
convenience stores, filling station forecourts and cafes. 11 Secondly, IN utilised
subscription contractors to service subscribers. 12 Thirdly, IN made use of
street contractors who sell to commuters on the streets. Finally, IN made use of
a group of independent distributors, dubbed ‘country distributors’. The country
distributors, whose services are at the heart of the present matter, are
described below.
7 Service agreement, joint bundle page17 clause 5.1.
8 Service agreement, joint bundle page 20 clause 7.
9 Service agreement, joint bundle page 23 clause 12.1.
9 Service agreement, joint bundle page 23 clause 12.1.
10 Amending agreement, joint bundle page 30.
11 These largely fall into the greater Durban and Pietermaritzburg areas and a few fall within
the former African townships of Durban. See page 575 of the transcript.
12 Currently these are only found in Durban and Pietermaritzburg.
3
10] The term “country distributors” was used by IN to describe the fifteen
independent contractors who were contracted to it to distribute its newspapers
in areas prescribed in the contracts with IN. The country distributors covered
the greater area of KZN, that is, those large areas of the province falling
outside the metropolitan areas of Durban and Pietermaritzburg. The most
prominent of these were:
[10.1] Ollason’s News Agency based in Stanger and covering the coastal area
from Stanger to Gingindlovu and inland from Eshowe to Nongoma
including Hlabisa;
[10.2] Umlazi News Agent (owned by Mr Mbatha), covering the Umlazi
Lamontville area and the African townships south of Durban;
[10.3] Central Media Distributors (owned by Mr Smith) based in Empangeni
and covering the area north of Gingindlovu to Kosi Bay;
[10.4] JC Delport Distributors (owned by Mr Delport) based in Newcastle and
covering the area north of Ladysmith including Dundee, Vryheid and
Newcastle; and
[10.5] ZJ Distributors (owned by Mr Zondi) based in Claremont and covering
the African townships to the north and west of the greater Durban area.
11] In ABC region 48, covering the greater Durban area, and in Pietermaritzburg,
(area 51) IN used its own staff and employees to distribute Ilanga to retail
outlets.13 In area 48 there were also some independent contractors
distributing to street vendors and informal outlets. Note that the densely
populated former African townships surrounding Durban fell, for the most part,
into ABC region 47 although it seems that some of these ‘township’ areas
13 ABC is the Audit Bureau of Circulation which audits newspaper sales. In performing its
13 ABC is the Audit Bureau of Circulation which audits newspaper sales. In performing its
audit function it divides the country up into a number of regions in which circulation is
measured.
4
(notably those served by Mr. Zondi), fell into area 48. The distributors serving
these urban township areas – these being the distribution businesses of Mr
Zondi and Mr Mbatha are also included in the list of ‘country distributors’. IN
also engaged in some direct distribution into these areas.
12] Discussions were held during the course of 2000 between MM and IN
concerning the renewal of the service agreement. However, they failed to reach
agreement on the conclusion of a new contract. 14
13] On 19 November 2000 MM entered into a service agreement with the Natal
Witness Printing and Publishing Company (Pty) Ltd (“NW”), a Pietermaritzburg
based publisher, to publish and distribute Ilanga.15 NW, which was founded in
1846, is a 50/50 joint venture between Media 24, which is a subsidiary of
Naspers, and a family shareholding, which is brought together through a
company called Lexshell (Pty) Ltd. 16 NW’s flagship title is the Natal Witness
newspaper.17
14] The material terms of the agreement between MM and NW stipulate that:
[14.1] the agreement would commence on 1 April 2002;
[14.2] MM’s responsibilities would include the format of the newspaper, the
appointment and control of editorial staff and content;
[14.3] NW would be responsible for printing, advertising and distribution;
[14.4] the fees payable to NW would be calculated as an agreed percentage of
pretax profits of MM in relation to the latter’s newspaper business
during any annual period;
[14.5] during the currency of the agreement neither party would publish or in
14 IN wanted the agreement to be renewed on the same terms but this was rejected by MM
which viewed the terms of this agreement as excessively onerous.
15 Memorandum of agreement, joint bundle page 43.2.43.
16 Transcript p282.
17 Ibid
5
any other way control or have an interest in any other newspaper
distributed in KZN which might be seen to compete with one of the
newspapers listed in the agreement without written approval from the
other party. The listed newspapers included Ilanga.
15] At the time of the conclusion of the service agreement NW did not have
sufficient capacity to print Ilanga. Thus it had to incur capital expenditure in the
amount of R32 million for the purchase and installation of a new printing press,
an investment which increased NW’s printing capacity some tenfold.
16] NW, like IN, operated its own distribution division. However in much the same
way that IN’s direct distribution service was centred on Durban, NW’s core
direct distribution area centred on Pietermaritzburg. It appears that NW relied
on IN’s network of contracted country distributors to distribute its titles in their
respective areas of operation. Mr. Delport testified that, on being approached to
distribute NW’s titles, he had sought permission of IN because this was
required by his contract with IN:
So, in the real sense of the word, yes, we are independent contractors.
We are not Independent Newspaper contractors. We are independent
contractors. We’ve got different clauses in all contracts and people are
fully aware of it. If my thoughts don’t let go of me, it’s 95/96 Natal
Witness have contacted us via the Caxton Group, which were doing
the local papers, or the Penrose Group at that time, and asked us if we
can’t look in distributing Witness for them in the area. Obviously it’s a
small number of papers we are talking about. So they requested us.
So, we went to Independent and asked, being basically our biggest
contractor at that time, the Independent Group, obviously being from
KZN, they had a clause in their contracts that we had to get permission
KZN, they had a clause in their contracts that we had to get permission
from them. Basically they came to us, being fully aware that we are
doing the Johannesburg papers in KZN. Now we are approached by a
KZN company to do KZN papers. We approach Independent, which
didn’t have a problem, because obviously there were also some scale
6
economies for them involved. So they came through. 18
17] At the time of concluding the new service agreement with NW, MM did not
inform IN that it would not be renewing their agreement. In fact it only
communicated its decision not to renew their service agreement to IN in a letter
dated 8 October 2001, that is, some eleven months after concluding the new
agreement with NW and six months before the expiry of the service agreement
with IN which was due to terminate on 31 March 2002. 19 Both MM and NW
were intent on ensuring – and understandably so – that IN should not get wind,
at too early a stage, of the new agreement. MM was anxious to ensure that
IN’s service levels did not deteriorate in the remaining period of the
agreement.20 NW did not want knowledge of its massive investment in new
printing capacity brought to the attention of its competitors before it was ready
to deploy these assets. 21
18] Once IN had been informed by MM that their service agreement was not to be
extended, a series of transitional arrangements was agreed upon between
them. A task team comprising senior employees of both companies was
formed to oversee the transition of the functions in question from IN to NW. The
first meeting of the task team was on 15 November 2001. At that meeting, Mr
Christianson, an MM executive member, requested certain information from IN
on behalf of MM. This information included:
[18.1] Guide date statistics for the previous 12 months and records going
forward to 31 March 2002;
[18.2] A listing of all Agents, giving physical and postal addresses, telephone,
and fax numbers; and
[18.3] Details of commissions paid both to the retail agents and contractors.
18 Transcript pp 656
19 Letter by Mr Konigkramer to Mr King, dated 8 October 2001, joint bundle page 44.
20 See page 313 of the transcript and page 43.2.9 of the joint bundle.
20 See page 313 of the transcript and page 43.2.9 of the joint bundle.
21 See page 362 line 423 – page 363 line 17 of the transcript, 26 July 2006.
7
19] It appears that IN remained intent upon salvaging some of the lucrative
distribution business that it was set to lose as a result of the termination of the
nonrenewal of its agreement with MM. Accordingly at the meeting of 15
November 2001 IN raised with MM the possibility of future cooperation on
distribution. IN subsequently made an offer to MM to provide a distribution
service in respect of ABC regions 47 and 48 (which cover the greater Durban
Metropolitan area including the former African townships). 22 This offer was
ultimately rejected by MM on 24 January 2002. 23
20] On 29 November 2001 IN informed MM that it had rejected its request to
provide the information listed above. IN claimed that it was acting on legal
advice to the effect that it was not obliged to provide MM with all the information
that it had requested. IN proposed what it referred to as an “amicable
conclusion” which would enable it to share information on circulation history as
well as a list of agents’ names and addresses but only on condition that IN and
MM concluded a suitable distribution agreement – apparently referring to ABC
districts 47 and 48 in respect of Ilanga. IN stated that it was not, under any
circumstances, prepared to give details of commissions paid to retail agents
and contractors and it was not prepared to give any of the advertising
information other than a list of advertisers that had advertised in Ilanga over the
past twelve months.
21] IN proceeded to instruct its staff to ensure that all material relating to
advertising and promotions of Ilanga was securely stored away since it was
proprietary information. 24 The staff was expected to sign the instructions as a
gesture of acceptance.
gesture of acceptance.
22] In its letter dated 8 February 2002 IN made it clear to MM that no information
would be forthcoming since it had turned down IN’s offer to distribute Ilanga in
22 Letter from Mr Maclaine to Mr Christianson dated 27 November 2001, joint bundle pages
59.159.2.
23 Letter from Mr King to Mr Konigkramer dated 8 February 2002, joint bundle page 79.
24 See letter from Mr Taylor to Ms De Klerk dated 3 December 2001, joint bundle page 66.
8
areas 47 and 48. 25
23] When it became clear that the information requested would not be forthcoming,
MM responded by initiating arbitration proceedings on 11 February 2002. The
arbitration took place in the middle of April and the arbitration award was
delivered on 30 April 2002 and was in favour of MM. 26 IN partially complied
with the arbitration award by furnishing some of the information that had been
requested. Subsequent correspondence between IN and MM resulted in further
information being provided. 27
24] After expiry of the service agreement with MM, IN launched its own publication
called Isolezwe that was intended to be a direct competitor to Ilanga. The initial
intention was for Isolezwe to be launched on the 1 st April 2002, the date from
which respondent would no longer be publishing or distributing Ilanga.
However the launch was ultimately delayed by one week.
25] Recall that the agreements in force between IN and the country distributors
provided that the latter required IN’s permission to distribute nonIN
publications. This is illustrated in JC Delport’s contract with the respondent in
which it is stated that:
The Wholesaler shall not be entitled, within the Territory to distribute
any publications other than those delivered to it by Natal Newspapers
in terms hereof, without Natal Newspapers prior written consent which
consent shall not be unreasonably withheld in respect of publications
which do not compete with those distributed by Natal Newspapers .28
26] In October 2001 IN held a meeting with the country distributors. The meeting
was chaired by Mr Lorne Maclaine, IN’s Circulation Director. He informed the
country distributors that from the beginning of April 2002 IN would no longer be
25 Letter from Mr King to Mr Konigkramer dated 8 February 2002 (wrongly dated 2001), joint
bundle page 79.
26 The arbitration award is on page 162 of the joint bundle.
bundle page 79.
26 The arbitration award is on page 162 of the joint bundle.
27 There is a series of letters on pages 169, 170, 175, 176, 177, 178, and 179 of the joint
bundle.
28 Agreement between Natal Newspapers and JC Delport Distributors CC, Exhibit 2 p1.
9
distributing Ilanga. He also told them that the IN was considering launching its
own isiZulu language title in direct competition with Ilanga. A second meeting
was held with the country distributors in January 2002 at which Mr. Maclaine
confirmed that IN was indeed to launch its own isiZulu language title, to be
named Isolezwe, at the beginning of April 2002. The intended launch was
timed to coincide with the termination date of the contract between IN and MM.
Mr. Maclaine also informed the country distributors that distributors of
Isololezwe would not be permitted to continue distributing Ilanga.
27] On 5 March 2002 IN wrote letters to the country distributors confirming its
requirement that no distributor under contract to IN was permitted to provide
distribution services to Ilanga. It also advised the country distributors that it
would only be launching Isolezwe in the second week of April. It tendered a
cheque to each of the distributors in lieu of distribution fees lost in respect of
this week. The cheques were of differing amounts determined largely by the
proportion of each distributors’ income earned from the distribution of Ilanga
and that would be forfeited as a result of the exclusivity agreement. The country
distributors were asked to accept the cheques together with the terms and
conditions specified in the letter. They were requested to sign the letter and
return it to the respondent.
28] The clause in the letter of 5 March 2002 which prohibited the country
distributors from distributing a ‘direct competitor’ of Isolezwe read:
“No Independent Newspaper’s contractor will be permitted to provide
any services for a direct competitor of Isolezwe after 1 April, 2002.
No Independent Newspapers’ contractor may launch, sponsor,
No Independent Newspapers’ contractor may launch, sponsor,
underwrite or assist any contractor distributing for a competitor of
Isolezwe after 1 st April 2002. (This exclusion is to discourage any
contractor from considering settingup a parallel distribution network to
be operated by a friend or a family member).” (sic)
29] This clause was directed primarily at Ilanga. In fact the country distributors
10
continued to distribute other titles not published by IN and which are in
competition with some of its other titles. For example, Mr Delport and other
distributors distributed The Witness and The Mercury , which are direct
competitors.29
Submission of complaint
30] On 13 February 2003 MM filed its first complaint with the Commission. This
complaint was based on IN’s refusal to furnish certain information to MM. The
Commission declined to prosecute this complaint and issued a notice of ‘non
referral’ in its notice dated 13 May 2003.
31] Upon receiving the nonreferral notice from the Commission, MM wrote to the
Commission alleging further anticompetitive behaviour. The Commission
advised MM to file a fresh CC1 complaint with the Commission. MM then filed
its second complaint on 24 July 2003. The new complaint incorporated both of
what have come to be referred to as the information complaint and the
exclusivity complaint. The Commission again nonreferred the matter in its
notice dated 25 May 2004.
32] MM then referred the complaint to the Tribunal on 25 June 2004.
The complaint
33] The complainant’s case is based on two causes of action. The first complaint
was based on the refusal by IN to furnish certain information to MM during the
period December 2001 to April 2002. This is what is referred to as “the
information complaint”. The second complaint is based on IN’s refusal to permit
the country distributors to distribute a direct competitor of Isolezwe, effectively
Ilanga. This is what is referred to as ‘the exclusivity complaint”.
34] MM contended that the exclusivity clauses constitute a prohibited restrictive
vertical practice as contemplated by section 5(1) of the Act, in that it is an
29 See pages 717, 783, 784 and 785 of the Transcript.
11
agreement between parties in a vertical relationship that has the effect of
substantially preventing or lessening competition in a market. Section 5(1) of
the Act states that:
An agreement between parties in a vertical relationship is prohibited if
it has the effect of substantially preventing or lessening competition in
a market, unless a party to the agreement can prove that any
technological, efficiency or other procompetitive gain resulting from
that agreement outweighs that effect.
35] MM further contended that the conduct was a prohibited abuse of dominance
as contemplated by section 8(d)(i) of the Act. Section 8(d)(i) states that:
It is prohibited for a dominant firm to engage in the following
exclusionary act…. unless the firm concerned can show technological,
efficiency or other procompetitive gains which outweigh the anti
competitive effect of its act, …(i) requiring or inducing a supplier or
customer to not deal with a competitor .
36] With regards to the information complaint MM alleged that the refusal by IN to
furnish it with certain information amounted to an abuse of dominance by a
dominant firm and is prohibited by section 8(c) of the Act which reads:
It is prohibited for a dominant firm to … engage in an exclusionary act,
other than an act listed in section 8(d), if the anticompetitive effect of
that act outweighs its technological, efficiency or other procompetitive
gain.
37] An ‘exclusionary act’ is defined in Section 1(1)(x) of the Act which reads that:
’exclusionary act’ means an act that impedes or prevents a firm
entering into, or expanding within a market.
12
The relief sought
38] MM seeks an order declaring the exclusive terms in the agreement between IN
and independent distributors to be void, and to be declared a prohibited
practice in terms of the Act for the purposes of section 65 of the Act. Secondly,
MM seeks an order declaring IN’s refusal to furnish certain information a
prohibited practice in terms of the Act for purposes of section 65 of the Act. In
addition to these orders MM is asking for an order as to costs
39] Section 65 of the Act provides that a party who has suffered damages arising
from conduct prohibited in terms of the Act can only commence action for
damages in the civil court if the Tribunal has declared the conduct to constitute
an infringement of the Act. If the party commences civil action for damages
based on a breach of the Act, the civil courts can refer the matter to the
Tribunal to determine if there was in fact a breach of the Act. It seems the
complainant intends to institute civil action for damages and so has asked the
Tribunal to declare the above conduct a prohibited practice in terms of the Act.
Competition Analysis
40] It is as well to recount, even at the risk of some repetition, the essential facts
and allegations.
41] This is a complaint brought by the publisher of a single newspaper title, namely,
Mandla Matla (MM), the owner of Ilanga, an isiZulu language newspaper, the
vast bulk of whose large market is in the province of Kwazulu Natal (KZN). MM
owns none of the assets required to print and distribute a mass circulation
newspaper and so it is obliged to enter into a contract with a service provider
possessed of the requisite assets and skills.
42] The complaint is directed at the alleged conduct of Independent Newspapers
(IN), a major national newspaper group, which owns a number of newspaper
(IN), a major national newspaper group, which owns a number of newspaper
titles prominent in the province of KZN (and in the rest of the country) but none,
it appears, that, at the time of the filing of the complaint, competed headto
13
head with Ilanga. IN – or rather its various predecessors in title – had
previously owned the Ilanga title which it sold to MM in 1987. At the time of the
sale MM and IN concluded an agreement in which IN retained contractual
responsibility for the printing and distribution of Ilanga. Pertinently, the terms of
this agreement prevented IN from publishing another isiZulu language
newspaper.
43] The complaint originates in MM’s decision to move the highly lucrative contract
for the printing and distribution of Ilanga from IN to the Natal Witness (NW)
group. NW is also a significant publishing group based in KZN (it is, in fact,
jointly controlled by Media 24, one of the country’s largest publishing groups).
NW produces a major newspaper title (the Natal Witness) circulating in KZN, a
title which competes, to a certain extent, with an IN title (the Mercury), although
it too does not own a title that competes directly with Ilanga.
44] Because MM’s decision to move the contract to print and distribute Ilanga from
IN to NW released IN from the contractual restraint that prevented it from
publishing a newspaper in the same market segment as Ilanga, a veritable
windstorm of competition was unleashed in the KZN newspaper market with the
launch, immediately upon termination of the printing and distribution contract, of
IN’s isiZulu language title, Isolezwe.
45] A mere five years later, the upshot of this outbreak of competition is a market
segment – that is the market segment for isiZulu language newspapers – in
which IN’s new title, Isolezwe, is the largest participant, and in which the
venerable 100 year old Ilanga, though still a very significant presence, is the
smaller of the two newspapers competing for the custom of isiZulu language
newspaper readers. In the period January to June 2002, Ilanga had a weekly
newspaper readers. In the period January to June 2002, Ilanga had a weekly
circulation of 191 634 and a market share of 58%, whereas Isolezwe had a
weekly circulation of 137 375 and a market share of 42%. In the period July to
December 2005 Ilanga had a weekly circulation of 221 960 and a market share
of 34%. Isolezwe had a weekly circulation of 431 160 and a market share of
66%.30 Ilanga was published only on Monday and Thursday whereas
30 See pleadings bundle page 59 where these figures are extracted from Audit Bureau of
Circulations of Southern Africa (ABC).
14
Isolezwe was published from Monday to Friday. Note that this means that
although Isolezwe’s weekly sales volumes are significantly larger than Ilanga’s,
the latter’s daily average sales are still quite significantly higher than those of
their rival.
46] MM contends that IN could not have achieved this without the expedient of
denying Ilanga, or, more accurately, NW, Ilanga’s newly contracted printer and
distributor, access to the market for the distribution of newspapers in key areas
of the province. MM alleges that this foreclosure of the distribution market was
achieved, firstly, by a contract entered into between IN and several
independent distributors which forbade the group of distributors from
distributing Ilanga on pain of forgoing the right to distribute IN titles (henceforth,
the ‘exclusivity complaint’) and, secondly, in denying MM, or, more accurately,
NW, access to certain information that would have enabled NW to establish its
own distribution network (henceforth, the ‘information complaint’). 31 In the
words of MM’s counsel:
The complainant’s case is that the respondent abused its market
dominance with the purpose of inflicting harm on the complainant and
in order to ease the process of growing the respondent’s new product,
Isolezwe.32
And again:
‘Although the conduct complained of has been, and continues to be,
identified and analysed as two distinct types of conduct, both were
directed to and had the same impact, i.e. to substantially prejudice the
complainant in its efforts to distribute Ilanga.’ 33
47] Note the intimate connection between the exclusivity complaint and the
31 We stress that contractual and operational responsibility for distributing Ilanga lay with NW
and not with MM. The access to the independent distributors and/or the information that is
and not with MM. The access to the independent distributors and/or the information that is
requested would effectively go to NW rather than MM. As elaborated below, the evidence
suggests that this clearly accounts for some of IN’s resistance to the demands for access and
information.
32 Complainant’s Heads of Argument para 2
33 ibid para 4
15
information complaint. In the words of MM’s counsel:
Moreover, and as will be seen, each type of conduct was as effectively
prejudicial as it turned out to be because of the other conduct; that is to
say, if there had been no exclusivity imposed on the distributors the
information would not have been needed as badly as it was and vice
versa. There is accordingly some artificiality in analysing the conduct
separately, and this must be borne in mind in what follows.
With regard to the information complaint, had it not been for the
respondent’s complete dominance of the relevant distribution market
the information would not have been so important to the complainant
(it could have accessed the distribution market without it) and there
would not have been any appreciable value to the respondent in
denying information to the complainant. That is what makes the
information complaint the proper subject of a competition complaint. 34
48] In fact, as already noted, the complaint first surfaced before the Commission as
an information complaint in its entirety. When the Commission declined to
prosecute this complaint, a second complaint, the exclusivity complaint, was
filed. When this too was ‘nonreferred’ by the Commission, MM elected to bring
both complaints directly to the Tribunal and it is now before us as a complaint
with two separate but, as is conceded by the complainant, intimately related
causes of action, namely, the exclusive dealing element and the denial of
information element, with the former conduct, the exclusivity complaint, clearly
in the foreground.
49] We agree with IN’s counsel that these are properly considered as alternative
causes of action. We will follow the path taken in both parties’ heads of
argument and focus our competition analysis on the exclusivity complaint –
which is alleged to contravene Sections 5(1) and 8(d)(i) of the Act and then
consider the information complaint (which is alleged to contravene Section
8(c)), to the extent that any additional issues are raised by separate
34 ibid paras 4 and 5 our emphasis
16
consideration of this latter cause of action.
50] IN’s counsel has sought to make something of the intimate connection between
the two causes of action and of its transformation from a complaint initially
based on the alleged withholding of information, to one in which the exclusivity
complaint is clearly foregrounded. It is suggested that there is some ‘clutching
at straws’ in the belated discovery, in the ‘afterthought’ that brought the
exclusivity complaint to centre stage. MM’s counsel is, of course, quite correct
in asserting that, even if the filing of the exclusivity complaint was belated, even
if it only occurred to the complainant after the Commission’s rejection of the
information complaint, this has no bearing on the legitimacy of the exclusivity
complaint. It is properly before us and is entitled to be examined and decided
on its own merits.
51] And yet there is an inference to be drawn from the sequencing of the
complaints. IN and NW are not only publishers and printers of newspaper
titles, some of which are in direct, albeit muted, competition. They are also
established distributors of newspapers and other printed matter. We have not
been told much about their general distribution businesses although we do
know that their respective distribution strengths are IN’s network in the greater
Durban area and NW’s distribution network in the greater Pietermaritzburg area
through which they distribute, inter alia, their notionally competing flagship
titles, respectively the IN’s Mercury and the NW’s Witness.35 We also know of
IN – and we presume the same is true of NW – that the distribution business is
a discreet profit centre, the services of which are available for hire to
newspaper and magazine titles outside of IN and NW stables. 36
newspaper and magazine titles outside of IN and NW stables. 36
52] There is evidence to suggest that this initially surfaced as an information
complaint because NW, which, with the winning of the contract to distribute
Ilanga with its large Durban readership base, had gained the material basis for
spreading the reach of its distribution business into Durban, the very heartland
35 We use ‘ notionally competing’ advisedly because although the Witness has a limited
presence in Durban and the Mercury in Pietermaritzburg and both have some presence in
greater KZN beyond both cities, it is clear that in their respective core metropolitan markets
competition between the titles remains limited at best.
36 Transcript page 681
17
of IN’s distribution business. Moreover, this would not only have enabled the
NW’s distribution business to compete with IN’s distribution business in Durban,
but it would have enabled NW’s Witness title to compete more effectively in
Durban with IN’s Mercury. Certainly we glean from the evidence of Mr. Graham
King, at the time the CEO of IN’s Natal operation, to the arbitration hearing into
the withholding of information that this is what IN apprehended and it is a
perfectly reasonable apprehension. 37
53] Now as a body charged with defending and promoting competitive market
structures, we undoubtedly welcome competition between what had been two
geographically separated distribution businesses, just as we welcome the
prospect of enhanced competition between the Mercury and the Witness.
Indeed in our view one of several manifestly positive outcomes of MM’s
decision to sever ties with IN in favour of a contract with NW is that it has
ultimately compelled NW to establish a distribution network in Durban. But it is
quite one thing to welcome competition and another to agree that competition
must be nurtured by a business – in this case IN – handing over critical
competitive information developed over many years and with much investment
of time and money to a rival distribution business and the owner of a major
competing newspaper title. 38 There is indeed a manifestly anticompetitive
claim that is advanced here – why would any business invest in a distribution
network, particularly one that was making these distribution services available
for sale in the market, if it reasonably apprehended that it would be obliged to
hand over the critical asset that is information to a rival supplier of distribution
hand over the critical asset that is information to a rival supplier of distribution
services? Such claims are not unknown in competition law but to be
successfully prosecuted it would have to be established that the distribution
asset in question was an ‘ essential facility’ defined in our Act as ‘ an
infrastructure or resource that cannot reasonably be duplicated, and without
37 See arbitration transcript at page 281 of the record of these proceedings:
Adv. Pammenter : ...Part of your concern about giving that information is because …
that these agents don’t only sell the Ilanga, they sell other newspapers.
Mr. King: Correct.
38 See arbitration transcript at pp 2817 of the record of these proceedings. King’s insistence
that his group had over the very many years of its existence built up a distribution system, in
which information was a critical competitive asset, one that to some extent was of particular
pertinence to the circulation of a particular publication but which was also a critical component
of the competitive advantage underpinning a successful newspaper distribution in general, is,
despite the apparent scepticism of the arbitrator, perfectly persuasive.
18
access to which competitors cannot reasonably provide goods or services to
their customers.’ 39 This is a considerable hurdle, one which was clearly
contemplated by the complainant but which it ultimately elected to steer well
clear of. 40
54] In short IN appears to have apprehended that NW intended using its newly
acquired distribution contract with MM as a beachhead for the easy
establishment of a rival distribution network in Durban. It is a reasonable
apprehension and one rendered all the more plausible by the initial
foregrounding of the information complaint rather than the exclusivity complaint.
That is, it is reasonable to infer that NW was not so much interested in gaining
access to IN’s Durban distribution network as it was in easing the establishment
of its own Durban distribution network by gaining access to IN’s distribution
information through the expedient of hanging on to the coattails of the MM
complaint. We infer further that it was when IN (reasonably, in our view,
despite the finding of the arbitrator) refused to hand over this critical
competitive information, that it occurred to NW and MM that, absent the
information, NW’s ability to effectively distribute Ilanga in the Durban African
townships and the country areas of KZN – the areas operated by the country
distributors was in jeopardy, and hence that it required the next best thing, viz,
access to IN’s established network of distributors. It is, in fact, precisely by
withholding the information and then denying access to the distributors
themselves that IN forced NW to set up its own competing distribution network,
in our view one of the most positive outcomes of this entire saga.
in our view one of the most positive outcomes of this entire saga.
55] But let us return to our narrative in order to identify the competition theory
underlying the complaint.
56] In the passages from the complainant’s heads of argument cited above there is
one clear indication of the competition theory posited when counsel argues that
39 Competition Act Section 1(1)(viii)
40 Note the European case of Oscar Bronner where a fledgling Austrian title attempted to
gain access to the national distribution network of a rival newspaper group by invoking the
essential facilities doctrine. Suffice to say that in this instance the complainant’s claims were
roundly and, in our view, persuasively rejected by the European courts. Oscar Bronner GmbH
& Co. KG v. Mediaprint Zeitungs und Zeitschriftenverlag GmbH & Co. KG and Others (Case
C7/97).
19
it is ‘ respondent’s complete dominance of the relevant distribution market’ that
made the acquisition of the information in dispute so essential. The implicit
theory here is that through its ‘complete dominance of the relevant distribution
market’ the respondent was able to leverage market power from the distribution
market, which it allegedly dominated, in order either to extend or acquire
market power in the newspaper market by denying rival newspapers access to
– ‘foreclosing’ in competition lexicon – all available distribution facilities.
57] Now it is clear that while IN (as with NW) actually owned a significant
distribution business, the use of this internal IN facility and these assets is not
in contention here. 41 In contention are the services of a group of independent
distributors who, though contracted to IN, had, in addition to their task as
distributors of IN titles, regularly distributed newspapers and magazines that did
not emanate from IN’s stable of titles, including Ilanga when its distribution was
the responsibility of IN. Recall that their contracts with IN incorporated a
‘moonlighting’ clause which required the distributors to seek permission of IN
were they to distribute a nonIN publication. It appears that this permission
had, hitherto, been regularly given. Hence certain of these distributors were
distributing both the Witness and the Mercury as well as other competing
titles.42
58] The independent distributors – who IN refers to as ‘country distributors’ –
included established distributors serving the rural areas and small towns of
KZN as well as two distributors responsible for large swathes of the vast,
formerly African townships of Durban. These latter were obviously particularly
formerly African townships of Durban. These latter were obviously particularly
important conduits for the distribution of an isiZulu language newspaper.
41 Were it to be placed in contention, the complainant would have been obliged to rely on the
difficult essential facilities doctrine. This was effectively conceded by MM’s counsel. See
transcript p950 at lines 1323.
42 Mr Le Roux, the Managing Director of NW, testified:
“ I’d met, on a couple of occasions, with the management of Independent in Durban
and there certainly was a mindset that actual distribution of product is not what is the
core of your business. The core of your business is producing newspapers and
producing the content that goes with those newspapers, and there had been
discussions that perhaps we should look at ways of working together to try and reduce
the costs of distribution to both parties. And with those kind of discussions behind and
we were travelling, and using Delport who was carrying both the Mercury and the
Witness on the same vehicle every day and had been doing that for several years.”
(Transcript p287)
20
Certain of these country distributors were also, it appears, responsible for
servicing the street vending networks in Durban itself.
59] The evidence suggests that the distribution task taken on by IN itself in
Durban’s city and suburbs (or by NW in the city and suburban areas of
Pietermaritzburg) is distinguished from the task undertaken by the ‘country
distributors’ by the fact that the former essentially involves distribution to well
established, stable retail outlets. The task of the country distributors, on the
other hand, involves a significant degree of distribution through many small,
often informal or semiformal, retail outlets. 43 This appears to involve a degree
of local knowledge regarding, for example, actual knowledge of the retail
outlets whose ownership and location may change quite frequently and their
reliability from a payment point of view.
60] A question that immediately arises is ‘how was IN able to persuade the country
distributors to deal exclusively with it and thereby potentially forego additional
income earning opportunities by taking on the distribution of other, even directly
competing publications?’
61] The decision by many of those independent distributors who were prepared to
forego income from the distribution of Ilanga was dictated by individual
43 Of course in their areas of operation the country distributors also distributed through many
of the same formal outlets – the large department and convenience stores, cafes, garage
forecourt stores, etc – served by IN and NW in their city and suburban bases. This is why
there is a substantive distinction between two types of retail outlets serviced by the country
distributors: on the one hand, those retail outlets to whom the country distributors distributed
but in respect of which the ‘publishing’ function (this seems to encompass the business of
deciding on the numbers of newspapers to be left at the retail outlet, the payment
arrangements, etc) was carried out by IN itself; and, on the other hand, the smaller, often less
formal outlets in respect of whom the country distributors assumed the ‘publishing’ function.
That is, in respect of these latter, the number of newspapers that the country distributors took
from IN was based on their own knowledge of what their retail outlets could sell. The country
distributors took a ‘bulk’ supply from IN and broke down this bulk based upon their own
‘publishing’ data which they were responsible for generating. Hence this is referred to as the
‘bulk distribution’ business. It appears that IN did not have detailed knowledge of this ‘break
bulk’ publishing data, because, it seems, the ‘publishing’ task in respect of their bulk
distribution was truly entrusted to their experienced country distributors. Hence when,
following the arbitration decision, the information in IN’s possession was handed over to
MM/NW, its utility to NW, for the purpose of building a new distribution network, was limited
and thus necessitated a demand to get access to the bulk distributors themselves, the
effective possessors of the information in question. Hence the exclusivity complaint assumed a
first order importance with the information complaint substantially relegated to the background.
21
commercial considerations. In this case, at first blush the appropriate
competitive response by NW or anyone else who desired to overcome the
exclusive dealing element in the distribution contracts was simply to offer the
distributors a premium over IN’s offer. There is no evidence to suggest that
NW attempted to gain access to the independent distributors by offering them a
more attractive distribution contract than that on offer by IN.
62] MM’s likely rejoinder to the above answer is: ‘we cannot pay a premium large
enough to compensate for the income that the distributors would forego by
deserting IN because the distributors would lose the right to distribute the entire
IN stable of publications in exchange for our single publication or much smaller
stable of publications.’
63] Implicit in the latter answer is a second competition theory that differs subtly
from the one sketched above, that is, from the theory that holds that dominance
in the distribution market was leveraged into the newspaper market. This
second theory holds implicitly that it is the IN group’s dominance over the
market for the purchase of newspaper and magazine distribution services, a
dominance that must naturally derive from its dominance over the market for
the sale of newspapers and magazines in the relevant geography, that is
leveraged in order to achieve domination over the distribution market. This
theory, though explicitly acknowledged, is never fully developed by MM
although it has been placed on the table. There is also some evidence of its
operation in practice. For example, Mr. Delport’s articulated fear that by
foregoing IN he would be sacrificing his ‘seven day income’ in exchange for
something less attractive is suggestive of IN’s power as a purchaser of
something less attractive is suggestive of IN’s power as a purchaser of
distribution services and, by logical extension, its power in the market for the
sale of newspapers and magazines. 44
44Mr Delport justified his decision to remain with IN when he said “Basically it was economical
based. I had a team that had to be paid, drivers, I had an infrastructure that I needed to be
financed and it basically comes down to that. Yes although I would like to go with Ilanga I have
to go where money was at that stage and to supply my people and provide my people with
jobs”.(See page 61 line 22 page 62 line 2 of the transcript). Mr Delport also stated that “I
think, as I said earlier, its economical decision that was taken at that point as a 7 day
operation where you are stepping into a 2day operation for many people and therefore it was
basically economical to say we’ve got a 7day income from Independent and should we go
with Ilanga, we only go 2 days?” (See page 94 line 1115 of the transcript).
22
64] However we should say at once that there are elements of this latter theory that
are highly questionable. The theory that IN leveraged dominance from the
newspaper publishing market into the distribution market posits that because
the IN group of newspapers is allegedly dominant in respect of all newspaper
sales in the province, it follows that it is the dominant purchaser of all
newspaper distribution services in the province, and hence that it is able to
exercise buyer power in the market for newspaper distribution services through
its provincial domination of newspaper sales. However this fails to take account
of the essential fact that newspaper distribution services are neither offered nor
demanded on a provincewide basis. Hence the question of buyer power in a
particular distribution area would have to be determined by the composition of
newspaper sales in that area and cannot simply be inferred from sales at a
provincial level. 45
65] For example, if Mr. Zondi earns the lion’s share of his income in his relatively
confined, but densely populated, region from the distribution of Ilanga – and the
evidence is that this is indeed overwhelmingly so 46 – then, in respect of this
distribution area, the market power that adheres to the publisher qua purchaser
of newspaper distribution services belonged to MM rather than IN, regardless
of IN’s superior market position in the province as a whole. Not to belabour the
point, Mr. Zondi had relatively little to fear from exclusion by IN. In fact all that
he had to fear was the prospect that Isolezwe may ultimately usurp the position
of dominance enjoyed by Ilanga in his region. His behaviour bears out his
dilemma which he attempted to resolve by hedging his bets, a strategic choice
that involved no small degree of deceit: he accepted IN’s terms and cashed the
that involved no small degree of deceit: he accepted IN’s terms and cashed the
compensatory cheque tendered and then carried on clandestinely distributing
45 This much is conceded by NW’s managing director, Mr. Le Roux, in the following exchange
with the Tribunal:
Chairperson: But in the areas in which Ilanga was important and certainly 47 and 48
were clearly those, and I’m not sure, but I think the evidence is that even in Delport’s
area Ilanga was a big, big seller. In those areas, is the dominant paper not in fact
Ilanga? Is this little Mandla Matla not really the dominant group in those areas, in the
areas in which Ilanga is important? Because over the whole of Natal, yes,
Independent Newspapers may be the biggest shot in town, but in Umlazi and
Kwamashu they are not. It’s Ilanga that’s the important title.
Mr. Le Roux: Mr. Chairman, the way you are defining that, if you use just Umlazi and
obviously KwaMashu, you are correct. (Transcript p3556)
46 Mr Zondi testified that the circulation of Ilanga contributed 85% of his income. See page
104 line 22 of the transcript.
23
Ilanga.47
66] As for Delport, while we do not know precisely the division of his revenue as
between Ilanga and the IN stable of publications, 48 it appears that he was
rather more vulnerable to IN’s power as a purchaser of distribution services,
hence his stated anxiety about losing his ‘seven day’ source of revenue. Again
one can reasonably hypothesise that his vast, relatively underpopulated area
of operation suggests a large number of small but heterogeneous customers
and a relatively large investment in physical assets in order to operate it.
Under these circumstances there is unlikely to be a single title large enough to
amortise his investment and so he would be more likely to be in thrall to the
group with the largest spread of titles and customers. Our point is that this
theory of competition cannot be tested at the level of the province – it has
rather to be tested in each contested distribution area.
67] The two theories posited and outlined above are not mutually exclusive. That
is to say it is entirely possible to construct a theory that is consistent with IN
possessing market power in both the newspaper distribution market and the
market for the publication of magazines and newspapers, where its dominance
in Market A is leveraged to protect and extend its dominance in Market B and
vice versa. Although there is little clarity on the definitions of the relevant
markets, in particular the newspaper publishing market, it appears that MM
does indeed contend that IN is dominant in both the newspaper distribution
market and the newspaper publishing market. It appears, however, that the
principal theory that is advanced by the complainant charges that the
respondent has abused its dominance in the distribution market – a dominance
respondent has abused its dominance in the distribution market – a dominance
that is represented by IN’s ability to induce or compel the independent
distributors to enter into exclusive dealing arrangements with it – in order to
protect and extend the position of Isolezwe at the expense of Ilanga.49 That is,
47See page 115 line 323 of the transcript
48 Delport did testify stated that Ilanga did not constitute a majority of his income. He
estimated that it contributed less than15% of his income. Transcript page 94 line 20 – 95 line
7.
49 See complainant’s heads of argument para 68: ‘ The conduct complained of was exerted in
the distribution market, and it was felt by the complainant in that market’ . Strictly speaking it
should rather read that the dominance leveraged by the respondent in the distribution market
was felt by the complainant in the newspaper market.
24
dominance in distribution services is leveraged to achieve or extend market
power in the newspaper market, implicitly in the market for isiZulu language
newspapers. This theory clearly underpins the complainant’s claim of a
contravention by IN of Sections 5(1) and, particularly, Section 8(d)(i).
68] The parties agree that in order to sustain a claim under Section 5(1) , it must be
established that:
[68.1] the parties are in a vertical relationship
[68.2] they have entered into an agreement
[68.3] the agreement has the effect of substantially preventing or
lessening competition.
69] If these three elements are established, the respondent is, in its defence,
entitled to seek to establish the existence of procompetitive gain flowing from
that agreement sufficient to outweigh its anticompetitive consequences.
70] Although expressed in terms slightly different from those used here, both
parties concur with this characterization of the elements of Section 5(1).
71] However, the parties’ understanding of the elements of the Section 8 , abuse of
dominance complaints exhibits an important divergence in the interpretation of
the Act.
72] In respect of Section 8(d)(i) , the MM’s counsel argues that the essential
elements that must be proved are
[72.1] that the respondent was a dominant firm within a relevant market; and
[72.2] that the exclusionary conduct was an act by the respondent requiring or
inducing a supplier or customer to not deal with a competitor.
73] If these elements are proved, avers MM’s counsel, it is then open to IN to show
that the anticompetitive effects of that act are outweighed by its technological,
25
efficiency or other procompetitive gain. 50
74] In respect of Section 8(c) (the basis alleged for the information complaint),
MM’s counsel contends that the essential elements to be proved are
[74.1] dominance; and
[74.2] that the refusal by the respondent to furnish to the complainant the
relevant information had the effect of preventing the complainant and/or
The Natal Witness from entering into or expanding within a market (i.e.
it was an ‘exclusionary act’); and
[74.3] that the anticompetitive effect of the act outweighs its technological,
efficiency or other procompetitive gain. 51
75] IN’s counsel however submits that even if it were assumed that each of the
elements listed above were proved ‘they do not make out a case of “anti
competitive effect”’. 52 Relying on our decision in South African Airways 53, IN’s
counsel sets out ‘ the proper approach to this requirement’ in his replying heads
of argument. 54
76] The written submissions of MM’s counsel avoids all mention of the necessity, or
otherwise, to prove anticompetitive effect. However he has dealt with this
issue in his closing oral submission where he, too, relies on South African
Airways in support of his contention that Section 8 does not require a showing
of ‘anticompetitive effect’. He insists that once the existence of an
‘exclusionary act’ is established, anticompetitive consequences are assumed
to flow from that act. He relies on the following passage of the South African
Airways judgment:
What this excursion into the case law and commentary suggests is that
50 Complainant’s heads of argument para 37
51 ibid para 38
52 Respondent’s heads of argument para 44.
53 Competition Commission v SAA (Pty) Ltd [2005] CPLR 303 (CT).
54 Respondent’s replying heads of argument paras. 26ff.
26
there is respectable authority for the notion that exclusionary practices
should not require evidence of actual competitive harm for a finding of
abuse. The finding is still possible if there is evidence that the
exclusionary practices, substantial or significant or expressed
differently, have the potential to foreclose the market to competition. If
it is substantial or significant, it may be inferred that it creates,
enhances or preserves the market power or the dominant firm. If it
does the latter, it will be assumed to have an anticompetitive effect .55
77] However the paragraph cited does not support MM’s contention. This passage
outlines what, in the words of IN’s counsel already cited, is the ‘proper
approach’ to measuring ‘anticompetitive effect’. South African Airways holds
that the anticompetitive effect of exclusionary conduct may be proven by
evidence of ‘actual competitive harm’ or by evidence of market foreclosure.
However, as the final sentence of the paragraph cited above confirms, all this is
to the end of meeting the requirement to prove ‘anticompetitive effect’ – it
cannot simply be assumed to flow from the existence of exclusionary conduct.
This is put to rest by the following passage from South African Airways which is
worth quoting at some length:
One approach is to say that if the act is exclusionary, it is deemed to
have an anticompetitive effect. On this approach the only issue that
remains to be decided is the balancing of the efficiency justification
against this deemed anticompetitive effect.
The problem with this approach is that it can lead to the outlawing of
conduct that has no anticompetitive effect. The definition of an
exclusionary act is very broad indeed. Discussing not dissimilar
exclusionary act is very broad indeed. Discussing not dissimilar
language, Areeda and Hovenkamp, in their treatise have this to say:
“In defining undesirable conduct, we are concerned mainly with
exclusionary behaviour, that which prevents actual or potential
rivals from competing or impairs their opportunities to do so
55 Competition Commission v SAA(Pty)Ltd Para.. cited at transcript p850 lines 110 our
emphasis
27
effectively. But this term and the root idea are also too broad,
for they embrace all competitive behaviour: All successful
competitive moves tend to exclude, particularly in oligopoly
markets”.56
The same observation by the authors can be made in respect of our
Act’s definition of an exclusionary act. The term is not a useful filter for
determining whether conduct is competitive or anticompetitive; both
can sensibly be included in the definition. If, however, we do not regard
the notion of anticompetitive effect, referred to in both paragraphs (c)
and (d), as inferentially linked to an exclusionary act, this danger can
sensibly be averted . It means that that the notion of what it anti
competitive is something different to an exclusionary act.
At a purely textual level they appear to be notionally different. If they
were congruent notions, then the legislature need not have troubled
itself with introducing the language anticompetitive effect into the
paragraphs, but would instead have referred to exclusionary effects.
We suggest that there is a difference between an exclusionary act as
defined and the inference that it has an anticompetitive effect. Without
some notion of what the anticompetitive effect is it would be
impossible to perform the weighing with the efficiency justification that
both (c) and (d) require. In order to perform a weighing of the anti
competitive effect on the one side of the scale to the efficiency gain we
need to have some notion of its quantitative effect. But the definition of
an exclusionary act is descriptive of a conduct’s ‘type’, not its ‘gravity
or extent’. Thus by way of example the refusal to supply one customer
with a de minimus market share and the refusal to supply a substantial
number of customers, representing a large proportion of the rest of the
number of customers, representing a large proportion of the rest of the
market are both exclusionary acts in terms of d(ii), but they have very
different competitive consequences.
For this reason the Act requires a finding both in terms of paragraphs
56 See Antitrust Law, Areeda and Hovenkamp Volume 3, paragraph 651b
28
(c) and (d) that the complainant not only establishes that there has
been an exclusionary act, but also that it has an anticompetitive
effect.57
78] We note that not all of the conduct described in Section 8 requires the showing
of an ‘anticompetitive effect’. Section 8 describes two ‘ per se ’ abuses. These
are Section 8(a) which prohibits the charging of an ‘excessive price’, and
Section 8(b) which prohibits a refusal to give a competitor access to an
‘essential facility’. In respect of these two sections there is manifestly no
requirement to prove an anticompetitive effect – as with Section 4(1)(b) (the
prohibition of horizontal price fixing, market allocation and bid rigging), the
offences described in 8(a) and 8(b) are presumed to embody anticompetitive
consequences. We note however that the elements of each of these Section 8
per se offences – excessive price and essential facility – are defined in the Act
and the hurdles that the complainant is required to cross in order to prove that
the elements of the impugned conduct conform to those defined in the Act are
considerable. However section 8(c) and 8(d) require that both the elements of
the exclusionary act and its alleged anticompetitive effect are proved in order
to undertake the balancing required by the procompetitive defence that is
permitted in respect of these exclusionary acts but which is not provided for in
respect of the conduct described in Sections 8(a) and 8(b).
79] Notwithstanding MM’s denial of the necessity to prove anticompetitive effect, it
has made some attempt to show that there is indeed an anticompetitive effect
resulting from the IN’s alleged exclusionary conduct. We have examined the
evidence for this and found it to be wanting and we will detail these efforts and
explain the reasoning for our finding.
explain the reasoning for our finding.
80] Antitrust scholars, Gellhorn, Kovacic and Calkins, have distilled numerous
decisions of US courts and scholarly writings into a threepart test for
evaluating the reasonableness of exclusive dealing arrangements. 58 The first
is the extent of market foreclosure. The second is the duration of exclusive
57 SAA paras 107111 our emphasis
58 Ernest Gellhorn, William E. Kovacic, Stephen Calkins – Antitrust Law and Economics in a
Nutshell (Thomson West)
29
agreements. The third is the height of entry barriers into the affected market.
81] The high water mark of MM’s claim regarding the extent of foreclosure is that
the information withheld and the exclusive dealing arrangements implicated
26% of Ilanga’s market. We note that this figure represents 26% of the sales of
isiZulu language newspapers of which Ilanga, immediately prior to the launch of
Isolezwe, enjoyed a share of 100%.
The effect of the exclusive dealing was that the complainant was
immediately excluded from accessing retail outlets to which 26% of its
papers had been delivered ; these were retailers that did not even
appear on the IN’s agents list that was supplied to the respondent a
week after the arbitration award. That is a very substantial
foreclosure.59
82] If however the relevant newspaper market is broadened to include all English
language newspapers available in KZN – and this would be consistent with
MM’s own contention the proportion of the market implicated by the exclusive
dealing arrangement (that is, the extent of the foreclosure) shrinks significantly.
83] However let us accept, for the sake of this argument, that the relevant market is
indeed that for isiZulu language newspapers, an assumption which strongly
favours MM’s contention regarding the extent of foreclosure while, ironically,
conflicting with its own contention regarding the actual relevant market for
newspaper publishing. 60 While a foreclosure effective over 26% of a market
would normally be grounds for concern, it is our view that this significantly
exaggerates the reach of the foreclosure claimed here. It cannot be that the
59 Complainants heads of argument para 50
60 MM requires a newspaper publishing market that is as broad as possible in order to support
its contention that IN is dominant in that market and so support the theory that IN had
leveraged a dominant position in the publishing market in order to achieve or extend
dominance in the distribution market. If the newspaper publishing market is the market for all
English and isiZulu language newspapers published or, even wider, sold in the province of
KZN then it is possible that IN, with its extensive portfolio of titles, may emerge as dominant
within the meaning of Section 7. However, if the relevant market for newspaper publishing is
the market for isiZulu language newspapers (the empirical basis for the complainant’s
contention that 26% of the market was foreclosed), then at the time that the alleged anti
competitive conduct is said to have taken place, MM’s title, Ilanga, was not merely dominant
but was an out and out monopolist.
30
new distribution service which NW was in the process of establishing (and
which, as we shall see below, was immediately operational) was ignorant of the
whereabouts of all of the agents through whom this 26% of Ilanga sales was
retailed. Indeed it is quite conceivable that the NW distributors had knowledge
of access to retail agents who were not known to the IN distributors. Recall too
that Mr. Zondi and Mr. Mbatha, despite having accepted the exclusive dealing
arrangement required by IN and having cashed the compensatory cheques
issued to them, continued distributing Ilanga, in their areas of operation.
84] In summary then even assuming – in MM’s favour in this instance – that the
relevant market is that for isiZulu language newspapers and hence that the
share of the market from which Ilanga was allegedly excluded equates to the
share of the total market that was foreclosed, there are strong reasons for
doubting that the actual foreclosure achieved was anything near as large as
26%.
85] What of the duration of the foreclosure? NW’s distribution manager, Ms.
Naidoo, testified that sales of Ilanga started to stabilise as early as the first
month after her network assumed responsibility for distribution, an outcome
that she attributed to the hard work she and her team had put into setting up
the network from the time – some two months before April 2002 – when it
became clear that the information that NW/MM demanded was not going to be
forthcoming, information, which, in any event, was made available immediately
after the arbitrator’s decision was handed down. 61 Ms. Naidoo also conceded
that sales started to steady up into the 90 000’s after the first month. Mr
Christianson stated that it took them three to four months to reestablish the
Christianson stated that it took them three to four months to reestablish the
street vending operations in the greater Durban area alone (excluding the
areas for which Mr Mbatha and Mr Zondi were responsible). 62 Mr Le Roux,
NW’s managing director, testified that the network was up and running a few
months after arbitration. 63 The duration of the foreclosure – such as it is – can
properly be described as minimal.
61 See transcript page 389 line 20; page 396 line 1324;and page 413414. Note that
because IN did not possess the detailed information that they required it did not, when actually
provided, prove to be as useful as MW had hoped.
62 See transcript page 192 line 916.
63 See transcript page 349350 .
31
86] And what of the extent of barriers to entry ? The rapidity with which NW
managed to get its own distribution facility up and running is, in itself,
thoroughly persuasive evidence of very low barriers to entry indeed. If more is
needed then surely the evidence and assessment of the distributors
themselves disposes of any possible doubt on this score.
Adv. Stewart : Now we know that when Isolezwe was launched in April
2002 you were not allowed by Independent Newspapers to distribute
Ilanga. In those circumstances, as you see it, what could the Natal
Witness or Mandla Matla do in your area to distribute Ilanga? What
options were open to them?
Mr. Delport: In my specific area honestly is to get a different distributor or startup
distributor to do deliveries in our area.
Adv. Stewart: And is that something… or let me put it like this. What would need to
be done to do that, to start up a new distributor? What are the challenges?
Mr. Delport: Depending on what you are going to distribute, it’s basically a vehicle. If
it’s larger, you are going to need some office space and personnel to do the
distribution and create the infrastructure. 64
87] And further:
Adv. McNally : And would you accept as I think you have accepted,
that’s its not particularly difficult to set up a an alternative distributor
provided its still economical to do so:
Mr. Delport: Well to be quite honest, we experienced a lot where people are coming in
to compete in our market. So it is fairly, I must admit on that point to say yes, it is
possible for them to set up a…
88] Mr. Zondi effectively concurs with Delport’s assessment:
64 Transcript page 57
32
Adv. McNally : Mr. Zondi just so I understand the role that play as a
distributor. Is it correct that you know the area that you work in?
Mr. Zondi: Very well.
Adv. McNally: And you for that reason, get to know the retail outlets that you can
place a newspaper at?
Mr. Zondi: Yes.
Adv. McNally: And what else do you have that makes you a useful distributor, apart
from your knowledge of the area, what do you need to be a distributor?
Mr. Zondi: It’s hard work, that’s all.
Adv. McNally: So anyone who knows the area and is prepared to work hard, they can
do the job?
Mr. Zondi: Exactly. 65
89] Mr. Maclaine testified that just prior to the launch of Isolezwe IN was obliged to
terminate the services of its principal distribution contractor in KwaMashu. 66
IN itself took over the distribution in this area and was able to get the service up
and running within 2 weeks. 67
90] However we would go further than this and say that the establishment of NW’s
distribution facility was so rapid that it constitutes evidence of the operation of
supplyside substitutability . We do not know the precise character and
composition of NW’s new distribution facility in Durban itself or in the areas of
IN’s country distributors. 68 However, to set up an effective distribution network
within, at most, a few months supports the contention – advanced by IN’s
65 Transcript pp 1134. Zondi later clarified that his business comprised 4 drivers, 4 assistant
drivers, 6 street vendors, one administrator and 4 vehicles. (Transcript page 124).
66 This does not refer to Mr. Zondi who handled part, but not all, of KwaMashu.
67 Transcript page 582.
68 In respect of Mr. Zondi’s distribution area, it seems that IN is distributing Isolezwe through
its internal distribution division. Zondi is distributing Ilanga and all other NW publications and,
he testifies, earning more than he would have earned had he elected to accept IN’s demands
he testifies, earning more than he would have earned had he elected to accept IN’s demands
to distribute its publications exclusively. (Transcript pages 1223)
33
expert witness – that the requisite skills and assets needed for the distribution
of newspapers in the areas in question already existed in the market. They
were not deployed for purpose of distributing newspapers for the simple reason
that IN gave its country distributors exclusivity within their distribution areas and
that, until the launch of Isolezwe, it also gave them permission to distribute non
IN titles. Hence there was no effective demand for the services of additional
newspaper distribution services. But once, with the launch of Isolezwe, this
situation no longer pertained, the assets required for establishing an alternative
distribution network were easily found, either from within NW’s existing portfolio
of skills and assets or from others who were performing closely related
functions. The evidence is that in one of the most important areas affected by
IN’s exclusive dealing requirement, namely the large area of Mr. Delport, NW’s
new distributor of Ilanga and presumably other NW titles is Baldwin’s
Distributors, an established distributor of pamphlets and other advertising
material in the area in question. 69 The evidence is that this service was
operational from Ilanga’s first issue in April. 70
91] It is our finding then that MM has clearly failed to establish anticompetitive
effect. IN has nevertheless attempted to preempt a finding that its conduct is
anticompetitive by asserting a procompetitive defence. Mr. Maclaine avers
that its insistence that the distributors of Isolezwe not also distribute Ilanga was
designed to ensure that they focused their energies and resources on getting
this new title into the market.
92] MM has identified particular and general inconsistencies in this argument. In
particular it points out that, despite the stated importance of exclusive focus by
particular it points out that, despite the stated importance of exclusive focus by
its distributors, IN attempted to secure a distribution contract for Ilanga in areas
47 and 48 even when it had already decided to launch Isolezwe. Mr.
Maclaine’s attempts to cast this as the efforts of the distribution centre to
salvage part of what had been a valuable contract, an attempt whose success
he partly feared because of the difficulties that it portended for the distribution
function, are not entirely persuasive.
69 Transcript page 93
70 Transcript Page 80.
34
93] On the other hand, we note that NW had indicated that it had no intention of
using one of IN’s country distributors, namely Mr. Smith who was responsible
for a large swathe of northern Natal ‘ because Mr. Smith is a former employee
of Independent Newspapers and was regarded by us as being too close to
Independent Newspapers’. 71 Clearly then there were circumstances in which
NW itself believed that joint distribution would compromise the competitiveness
of its titles which is precisely the basis upon which IN demanded an exclusive
dealing arrangement with its distributors.
94] In general, MM insists that joint distribution of competing titles is a common
phenomenon in the newspaper distribution market. While this is true, we are
persuaded that this is because the market is characterized by competition
between mature titles, several of them more than a century old. In any event,
the fact that joint distribution occurs frequently does not, in and of itself,
establish that it is compatible with robust competition – that is, competition in
the newspaper market may well be compromised by the practice of joint
distribution of competing titles. But more important we are persuaded by the
argument that the launch of a new title against a direct competitor of some 100
years standing demands extraordinary measures – that joint distribution is the
norm in the mature segments of the market does not constitute a justification
for extending it to the extraordinary event that is the launch of a new title. This
was conceded by Mr. Le Roux, the managing director of NW:
Chairperson:… Would it not be reasonable on [IN’s] part to apprehend
that if they were bringing out a competitor, that indeed the distributors
who have for so long relied on Ilanga, from whom so much of their
income is derived, too would favour Ilanga over this new competitor?
income is derived, too would favour Ilanga over this new competitor?
Was there not a reasonable apprehension on their part that if they
were going to reasonably penetrate this market, they could not do it
through a distribution network that was so dependent upon and
accustomed to distributing Ilanga already? I mean is that not a
reasonable apprehension in your view as a sort of newspaper
manager?
71 See Witness Statement of Mr. Le Roux cited in transcript at page 353
35
Mr. Le Roux: It is. It is Mr. Chairman. It is reasonable. 72
95] Nor do the available data suggest that even MM, much less competition itself,
suffered unduly in consequence of Isolezwe’s entry. Developments in the
newspaper market suggest that MM had, for long, been content with living the
‘quiet life’ of a monopolist. We say this because clearly the success of
Isolezwe in a remarkably short space of time is attributable not to the
destruction of Ilanga, to what would have amounted to the replacement of one
monopolist with another, through the foreclosure of the system of distribution.
Isolezwe’s arrival appears rather to have tapped into a rich vein of unsatisfied
demand as is manifest in the truly startling expansion of the market for isiZulu
language newspapers. 73 In the period between January to June of 2002, the
period when Isolezwe was introduced, the weekly circulation figures for isiZulu
language newspapers was 329 009. In the period July to December of 2003,
the weekly circulation figures for Zulu language newspapers had grown to 472
959. By the end of 2005 (the period between July to December) the weekly
circulation figures for Zulu language newspapers had climbed to a high of 652
850.74 Indeed it is reasonable to conclude that Ilanga and MM actually
benefited from the arrival of competition in the shape of Isolezwe. This is
evidenced by the strong assertion of Mr. Konigkramer, MM’s managing director,
that MM’s profitability has shown a marked improvement as a result of moving
its printing and distribution contract from IN to NW. 75
96] In any event, in our view the narrowly based procompetitive arguments for
72 Transcript p355
73 It appears that this development is not confined to the isiZulu language segment of the
market. Mr. Maclaine avers: ‘…the best thing that has happened to this industry is Media 24
launching the Daily Sun, Die Son in Cape Town also from Naspers and ourselves doing
Isolezwe. It’s opened up a whole new category from a previously docile, static, not doing
anything market. And there is great vitality. Again it’s the wrong word to say the bottom end.
That’s not what I mean, but the lower priced end of the market. There is enormous vitality and
things are happening.’ ( Transcript page 634)
74 See pleadings bundle page 59 where these figures are extracted from Audit Bureau of
Circulations of Southern Africa (ABC).
75 In Mr. Konigkramer’s own words: “Let me just sum it up and say that we had done an
analysis and we believe that we would be better off financially, much better off financially than
we were under Independent”. (Transcript Page 312 line 59). We understand that this is partly
a consequence of entering into a more favourable arrangement with NW but part of MM’s
current success is clearly due to expanded sales within a dynamic and growing market.
36
which IN contends are trivial when set alongside the manifestly procompetitive
chain of events triggered by MM’s decision to refuse to accept what it viewed
as IN’s onerous contract terms in favour of a better deal from an alternative
service provider. The exclusive dealing arrangement with the country
distributors may not have had a large role to play in these procompetitive
outcomes – but to the extent that it has, it is to be lauded for its contribution to
what is, from a competition perspective, an outcome far preferable to that which
has prevailed for the last very many years.
97] Note the chronology of events beginning with MM’s decision not to renew the
printing and distribution contract with IN in favour of a contract with NW. This
gives an indication of the manifestly procompetitive outcomes that flowed from
this decision. Consider the following:
[97.1] MM, dissatisfied with the service provided – and, particularly, the price
charged by IN, approached NW, a firm competing with IN in the
publishing, printing and distribution of newspapers in KZN province, with
a proposal that it assume responsibility for the printing and distribution of
Ilanga;
[97.2] NW, by agreeing to take on the printing of Ilanga, was obliged to make a
considerable investment in new printing capacity, capacity which would
not only release MM from its dependence on IN but which would
compete in the general printing market in KZN and probably further
afield;
[97.3] By agreeing to take on the distribution of Ilanga, NW recognised that it
would have to complement its strong distribution network in greater
Pietermaritzburg with distribution capacity in other areas in which Ilanga
had a significant presence, including greater Durban but also covering
country areas and small towns in other parts of KZN. This naturally
country areas and small towns in other parts of KZN. This naturally
portended the establishment of a rival distribution network covering parts
of the province where newspaper distribution appears to have been
dominated by IN, these being the city and suburbs of Durban itself where
IN utilised its own staff and assets for distribution purposes, and the
37
previously African townships of Durban and the country areas of KZN
where IN relied on a contracted network of independent distributors;
[97.4] MM, by electing to move its printing and distribution contract from IN to
NW, effectively freed IN to introduce a rival to the only isiZulu language
newspaper then in existence. While it is true that this market sharing
restraint simply moved from IN to NW, there is no evidence that NW had,
in the fifteen years in which IN was effectively shut out of this market
segment, contemplated producing its own isiZulu language publication
whereas IN, freed from its restraint, did immediately enter the market
with Isolezwe. The data suggest strongly that this not only established a
rival to a monopoly of some 100 years’ standing, but it massively
expanded the size of the market for isiZulu language newspapers.
98] In summary then, we have, thanks to MM’s decision to terminate its contract
with IN, a significantly more competitive printing market with the 10 fold
expansion of one player, NW; we have, in NW’s construction of a network to
parallel IN’s Durban and ‘country areas’ network, intensified rivalry in the
distribution market; and we have, thanks to the entry of Isolezwe, significantly
enhanced competition in the market for isiZulu language newspapers. By the
single decision to move the contract for the printing and distribution of Ilanga
from IN to NW, MM has effectively animated a significant improvement in the
competitive structure of three important markets, namely the markets for
newspaper printing, distribution and publishing.
99] Our finding then is that MM complainant has failed to make out a critical
component of a claim under Sections 5, 8(c) or 8(d), namely, that the conduct
generated an anticompetitive effect. Given MM’s failure to establish anti
generated an anticompetitive effect. Given MM’s failure to establish anti
competitive effects arising from the conduct complained of, IN is not obliged to
put up a procompetitive defence. It has nevertheless elected to do so. While
the narrowly based defence that it has chosen – namely, that the exigencies of
entering a market against a longstanding monopolistic provider of a
newspaper, a product to which consumers could reasonably be assumed to
have a strong degree of attachment, required the exclusive focus – is not
particularly persuasive, the respondent cannot be faulted for taking this
38
precaution which, on balance, probably contributed little to the core competitive
strategies it employed these being a well resourced and well executed
marketing campaign and the launch of a fresh, more contemporary product
offered to consumers. The success of Isolezwe was clearly abetted by MM’s
striking hubris which appeared to dictate its failure to respond with any
apparent urgency to the launch of a competitor. 76 However, in addition to the
narrow procompetitive defences asserted by the respondent is to be added the
manifestly improved competitive structure in at least three markets. It would be
a travesty for a body mandated to uphold competition to impugn conduct which
has made a contribution, albeit probably small, to an outcome that deserves
high praise rather than opprobrium. Be that as it may, the respondent is only
obliged to invoke procompetitive defences in order to counter the
complainant’s successful assertion of anticompetitive effect and this it has
failed to do.
100] The failure to show anticompetitive effect is sufficient ground for the dismissal
of the complaint.
Order
101] The application is dismissed. The complainant is ordered to pay the costs of
the respondent including the costs of two legal representatives.
76 MM’s witnesses were deeply offended by the allegation of ‘managerial slack’ that was
levelled at them by IN’s expert witness. In fact the evidence is that MM’s (and NW’s) response
to the impending launch of Isolezwe was remarkably complacent and by no means only in the
area of distribution. Pressed on this Mr. Konigkramer insisted that “But secondly Ilanga, as
you well know, is founded by a very eminent South African Dr John Dube and its an extremely
powerful brand. Its got great brand loyalty, so from our perspective what we would have done
powerful brand. Its got great brand loyalty, so from our perspective what we would have done
and which we did do is simply to make sure that we produced a good newspaper, which would
supply information needs of people and that is how we would grow circulation…” (page 310
line 1420 of the transcript). There can surely be no clearer illustration of the complacency to
which monopoly gives rise. This is made all the more remarkable when one considers the
commercial damage to IN that was caused by MM’s refusal to renew the printing and
distribution contract. In the course of the arbitration Mr. King revealed that the contract with
MM constituted approximately onefifth of IN’s profitability in KZN (transcript page 642). This
was clearly known to MM and, we presume, NW. When Mr. Konigkramer proposed to MM’s
board that it support moving the contract from IN to NW he suggested that the move would
jeopardise IN’s commercial viability in KZN. See Transcript p313 and joint bundle p 43.2.9.
Surely under these circumstances a robust response from IN must have been expected. And
yet MM (and NW) apparently preferred to conduct business as usual relying on the longevity of
Ilanga’s brand and the continuation of cosy, cooperative arrangements that appear to
characterise ‘competition’ between mature newspaper titles.
39
____________________
D H Lewis
Presiding Member
Y Carrim and M Madlanga concur in the judgment of D H Lewis.
Tribunal Researcher: R Kariga
For the complainant: Adv. Steward instructed by Bigby Woodhead Attorneys
For the respondent: Adv McNally instructed by Cliffe Dekker Attorneys
40