COPYRIGHT TRIBUNAL OF AUSTRALIA

Application by Isentia Pty Limited [2021] ACopyT 2

COMMONWEALTH OF AUSTRALIA

Copyright Act 1968

THE TRIBUNAL DIRECTS THAT:

1.    The applicant submit to the Tribunal proposed orders giving effect to the Tribunal’s decision as set out in the Reasons for Determination generally but having particular regard to [862] and [863] of the reasons.

2.    The question of the costs of the proceeding for the purposes of s 174 of the Copyright Act 1968 (Cth) are reserved for later determination.

3.    The Tribunal’s decision and reasons for determination be published from the Chambers of the President, the Hon Justice Andrew Greenwood.

COMMONWEALTH OF AUSTRALIA

Copyright Act 1968

THE TRIBUNAL DIRECTS THAT:

1.    The applicant submit to the Tribunal proposed orders giving effect to the Tribunal’s decision as set out in the Reasons for Determination generally but having particular regard to [862] and [863] of the reasons.

2.    The question of the costs of the proceeding for the purposes of s 174 of the Copyright Act 1968 (Cth) are reserved for later determination.

3.    The Tribunal’s decision and reasons for determination be published from the Chambers of the President, the Hon Justice Andrew Greenwood.

REASONS FOR DETERMINATION

GREENWOOD J (PRESIDENT), DR RHONDA SMITH AND MS MICHELLE GROVES:

Procedural and general contextual background

Introduction

1    These proceedings engage a broad sea of dispute between each of the applicants, Isentia Pty Limited (“Isentia”) and Meltwater Australia Pty Ltd (“Meltwater”) on the one hand, and the respondent, Copyright Australia Limited (“CA”) on the other hand, in relation to separate applications made to the Tribunal by Isentia and Meltwater under s 157(3) of the Copyright Act 1968 (Cth) (the “Act”).

2    In each application, the applicant claims, put simply for present purposes, that it requires a licence to exercise the reproduction and communication right comprised in the bundle of exclusive rights subsisting in the copyright in literary works (articles) published in print editions of newspapers and magazines (although “press clips” are now supplied digitally as PDF images of hard copy articles), or electronic online editions of newspapers and magazines (the “online” services) published by a number of well-known and some less well-known publishers. The well-known publishers include News Limited and its related entities (“News Corp” or “News Corp Australia”), Fairfax Media Limited and its related entities (“Nine Publishing” or “Nine/Fairfax”), Australian Provincial Newspapers Pty Ltd (“APN”), Rural Press Ltd (“Rural”) and West Australian Newspapers Ltd. Those publishers publish editions of their publications under prominent and some less prominent Mastheads. Some of the more prominent Mastheads, among others, are these: The Australian, The Sydney Morning Herald, The Age, The Financial Review, WA Today, The Daily Telegraph, The Herald Sun, Courier Mail.

3    The publishers are the owners of the copyright subsisting in the works they cause to be originated including works they then publish (putting to one side the rights they enjoy to publish the works of other (foreign) publishers under particular arrangements). That being so, they enjoy the exclusive right to reproduce each work in a material form (s 31(1)(a)(i) of the Act); to communicate the work to the public (s 31(1)(a)(iv)); and to exercise any of the other exclusive rights set out in s 31(1)(a) of the Act.

4    It almost goes without saying that the copyright in the relevant work is infringed by a person who, not being the owner of the copyright, does in Australia or authorises the doing in Australia of “any act comprised in the copyright”, without the licence of the owner of the copyright: s 36(1) of the Act. The reference to the doing of an act comprised in the copyright includes the doing of that act in relation to a substantial part of the relevant work: s 14(1) of the Act. The term “communicate” means “make available online or electronically transmit (whether over a path, or a combination of paths, provided by a material substance or otherwise) a work or other subject matter …”: s 10(1) of the Act. It follows that in order for anyone to engage in an act of reproducing or communicating to the public the whole of a published article, it is necessary to obtain the licence of the publisher.

Extracts

5    Where the doing of an act in relation to a part or portion or a snippet of an article (using those terms in a non-technical and non-definitional sense for present purposes), such as a headline from an article plus a sentence or a headline and some limited number of characters is said to constitute the doing of an act comprised in the exclusive reproduction or communication right, such an act may, or may not, be properly characterised as the doing of an act in relation to a “substantial part” of the relevant work. Whether it is, or is not, an act in relation to a substantial part of the relevant work engages a question of fact and degree and consideration of the extent of the copying, the quality of that which is copied and the importance of the part copied to the work in which the copyright subsists.

6    If the relevant part, portion or snippet is not a substantial part of the relevant work, the doing of the act will not engage an act of infringement and thus the licence of the copyright owner is not required in order for the person to undertake that act, as a licence is not necessary in order to render the act a non-infringing act.

7    If the part or portion is a substantial part of the relevant work, doing the act will require the licence of the copyright owner as the act will otherwise be an infringing act.

8    It also follows, at least as a foundational matter of principle, that in order to determine whether an act of reproducing and/or communicating a part, portion or snippet of a published article engages the doing of an act in relation to a substantial part of a published article (whether a press or online article) requires an evaluative judgment to be made concerning each and every act in issue and each and every corresponding article.

9    If a person seeks to engage in acts which do involve the reproduction of a work or the communication of a work to the public (for example, the whole of an article) which thus requires the licence of the copyright owner (or, for present purposes, a licence from CA), and the licensor seeks to bring within the licence (that is, within the scope of the grant and the price and non-price terms) as a condition of the grant concerning those acts that do require a licence, other acts that do not, a question arises as to whether bringing those other acts within the licence (and particularly as a basis for the payment of charges), is “reasonable”. A further question is, can a licence in relation to acts of reproduction and communication for which a licence is necessary, also bring within the scope of the grant and the price and non-price terms, activities (acts or uses) which do not involve an act of reproduction or communication (such as value added services provided by an organisation dedicated to media analysis and media intelligence) but which might be said to be “enabled” by “access” to the copyright works of a publisher or “use” of those works in the sense that but for acts of reproduction and communication (at one level, perhaps a threshold “ingestion” level), the value added services simply could not be undertaken. If those circumstances were to be the fact, is a licence adopting such an approach one which is “reasonable” in the circumstances of the particular licensee?

10    Questions related to the principles concerning acts said to involve a “substantial part” of a work have been the subject of extensive evidence in these proceedings in the context of particular conduct and services. We will return to the specific content and context of the matter later in these reasons. For present purposes, it is sufficient to simply note, at least as a matter of principle, that when it comes to extracts of online articles, on some occasions an evaluative judgment might result in a conclusion suggesting that an act engages the notion of a substantial part of the article and on other occasions not.

11    We do not propose to try and answer in these proceedings the question of whether the acts of extracting a portion of an article whether called a “Portion” in the context of the evidence or called a “snippet” or a “teaser” or some other term such as “ingress” is or is not an exercise of the reproduction and/or communication right on the footing that the extract constitutes an act in relation to a substantial part of a corresponding article. Such a task is not possible and, in any event, impractical without looking at each article and the corresponding extract. What can be said is that, portion by portion, article by article, a relevant extract, however described, when forensically examined in the context of the article from which the extract is taken or assembled, may or may not, engage an act in relation to a substantial part of the article. We will, however, return to this topic.

12    However, it should also be noted that whatever the evaluative judgment may be, article by article, should such a task be undertaken, Isentia and Meltwater contend that there must necessarily be a qualitative difference of some substance between exercising a reproduction or communication right in relation to an entire article (that is, the whole of the article) as compared with an act engaging, for example, a headline plus the first sentence or a headline and a limited number of characters from the text of the article.

13    For present purposes, we note that CA contends that the true qualitative character of the extracts in issue in the evidence suggests an act in relation to a substantial part of the corresponding article and thus an exercise of the reproduction and communication right by acts undertaken by each applicant. CA makes a broader point about these acts to the effect that even if an extract does not engage an act in relation to a substantial part of the relevant article, there is nevertheless a “use” of the works the subject of the article which needs to be taken into account in determining the “value” of the works and the value of that use to the applicants.

Two preliminary matters

14    At this point, it is convenient to note two matters.

15    First, much of the material in evidence is either confidential or highly confidential or enjoys a particular characterisation within a hierarchy of confidentiality. In these decision reasons, the Tribunal makes reference to evidence without seeking to isolate whether the references are confidential or not. Previously in these proceedings, the Tribunal has given decision reasons which removed confidential references into a separate confidential schedule. That exercise is a time consuming one in the course of writing. Having regard to the scale of the matter and the extensive evidence put on in the proceedings, it is simply not possible to take that step in the course of writing the decision reasons. Accordingly, the decision reasons will be published to the principal solicitor for each party who enjoys access to the highest level of confidentiality under the prevailing arrangements. The principal solicitor can then determine which persons within the representative cohort for the client can properly be given access to the reasons. In due course, the decision reasons will be reviewed with a view to either redacting particular information or removing it to a confidential schedule. The decision reasons will not be published on the Tribunal’s website or the Federal Court’s website until the process just described is undertaken.

16    Second, set out below is a list of the witnesses who gave evidence in the proceedings either by affidavit or orally or both and a reference to the party who “put on” or otherwise “relied” upon the evidence of the relevant individual. The schedule below identifies the relationship between the witness and the party. Subject to later addressing a particular contention advanced by the applicants concerning the evidence given by Mr Andrew Ross which we address separately, we are satisfied that all of the witnesses who gave evidence were doing the best they could to assist the Tribunal in the resolution of the issues (and, for present purposes, by referring to Mr Ross we do not mean to convey that Mr Ross was actively seeking not to assist the Tribunal, but rather a question needs to be later addressed in relation to his evidence).

The publishers

17    As to the publishers that own the copyright in the particular works, some of the publishers have elected to establish a vehicle, CopyCo Pty Ltd (“CopyCo”) to represent their interests in relation to the licensing of specified works in which they (or their related entities) own the copyright. CopyCo is described by CA as a “joint venture” entity. Its current shareholders are Fairfax Digital Pty Limited, News Corp Australia, Bauer Media Pty Limited (now known as Are Media Pty Limited), Rural (or a Rural entity) and APN.

18    Various publishers have entered into agreements with CopyCo granting it a non-exclusive licence to sub-licence specified rights in certain of their publications. Those publishers include Fairfax Media Limited, News Corp, Rural, Bauer Media Pty Limited, West Australian Newspapers Ltd, Elliott Newspaper Group Pty Ltd, The Border Watch Pty Ltd and Torch Publishing Company Pty Ltd. The scope of the non-exclusive licence conferred by each publisher on CopyCo is a function of the terms of each “Publisher Agreement”. Aspects of these matters have been previously addressed by the Tribunal: Application by Isentia Pty Limited [2020] ACopyT 2 (“July 2020 Tribunal decision”).

19    Each publisher that has conferred such a non-exclusive licence on CopyCo is referred to in the proceedings as a “CopyCo Publisher”.

20    There are, of course, many publishers that have not entered into a Publisher Agreement with CopyCo. They are referred to as “non-CopyCo Publishers” and their publications are described as “non-CopyCo Publications”. They consist of a large number of smaller and independent Australian publishers including, for example, the McPherson Media Group, Star News Group and Private Media which is the publisher of Crikey and The Mandarin. They include publishers of specific publications (such as Australian Mining, Australian Photography, Fishing World, Australian Doctor, Medical Observer). The non-CopyCo Publishers and Publications also include publishers that are members of CA directly.

CA

21    As to CA, it describes itself relevantly for these proceedings, as a non-exclusive licensor of “certain reproduction and communication rights with respect to works contained in: (a)  a printed edition of a newspaper, magazine or similar periodical; and (b)  an electronic edition of a newspaper, website or other electronic news service (other than a journal)”. It offers “voluntary licences” on both a “blanket” and a “transactional” basis to a range of users.

22    As to CA’s relationship with CopyCo Publishers, CopyCo by an Agency Agreement dated 20 March 2000 appointed CA to act as its “agent” in granting to classes of users, including media intelligence organisations otherwise known as media monitoring organisations (“MMOs”), (such as Isentia and Meltwater), “sub-licences” (falling within the scope of the primary grant by CopyCo to CA) to exercise particular exclusive rights of CopyCo publishers.

23    CA describes its “primary role” with respect to CopyCo licensed content as one of proactively suggesting potential licences, negotiating licences within its mandate and administering those licences. Mr Suckling (for CA) describes CA’s “function” as one by which CA “develops” and “negotiates” copyright charging models and licence terms for consideration by CopyCo against the background of market conditions and a function of calculating, collecting and distributing licence fees to CopyCo.

24    As to non-CopyCo Publishers, CA says that it has been authorised by its publisher “members”, “other rights holders” (the large number of smaller and independent Australian publishers mentioned earlier), the publishers of “over 5,000 NLA titles” (including The Financial Times, the Economist and others) and the publishers of “4,000 CFC titles” (including Le Monde and Le Figaro) to grant licences to MMOs of the right to copy and communicate “over 17,000 websites, newspapers and magazines” (which includes CopyCo material).

25    CA describes its “repertoire” for licensing as spanning:

… some of the most trusted sources of quality journalism in Australia as well as more popular news sources including The Sydney Morning Herald, The Age, Brisbane Times, WA Today, The Daily Telegraph, The Herald Sun, Courier Mail and news.com.au; regional titles owned by Rural Press Limited; independent editions including Crikey and The Mandarin; and [the 5,000 titles mentioned earlier including dailymail.co.uk and The Guardian, including the UK and Australian editions of each].

26    Apart from the role already described, CA is also a declared collecting society that manages certain statutory licences under the Act (such as educational and government copying licences). Although CA’s role as a statutory collecting society is not engaged by these proceedings, it seems clear enough that by reason of its role as a statutory collecting society (and thus its familiarity with the notion of “equitable remuneration”) and its role (exercised as early as 1999/2000) in negotiating and acting as a licensor of “certain reproduction and communication rights” (as earlier described), CA would necessarily be astute to the statutory concept (and the corresponding statutory constraint upon CA in proposing and negotiating licences), that a person claiming to require a licence from CA in relation to those reproduction and communication rights enjoys a right to test whether certain terms relating to the “payment of charges” or the “conditions” of such a licence put to that person are reasonable or unreasonable, in the particular circumstances.

27    The circumstance that the source of the rights the subject of the proposed licence (a sub-licence to be granted by CA as “licensor” under its grant) is one or more of a bundle of exclusive rights conferred on the publisher copyright owner (with CA standing in the shoes of the copyright owner within the limits of the grant) does not mean that CA is unconstrained, as a statutory matter, in calling for or “subjecting” a licence to terms and conditions that are, in the relevant circumstances, “unreasonable”.

The statutory provisions and aspects of the claims of the applicants

28    The source of the statutory constraint upon such a licensor in this regard is to be found in s 157 of the Act.

29    Relevantly for these proceedings, s 157(3) provides that a person “who claims” that they require a licence, and one of either s 157(3)(a) or (b) is engaged, may “apply to the Tribunal”, and if the claim is shown to be “well-founded”, the Tribunal must either make an order under either of s 157(6B)(a) or s 157(6B)(b) of the Act.

30    The “licence” contemplated by s 157(3) is a licence to be granted “by or on behalf of the owner or prospective owner of the copyright in a work … to do an act comprised in the copyright”: s 136(1) of the Act.

31    The term “licensor” is a defined term under the Act and it is common ground that CA is a licensor for the purposes of s 157(3) recognising of course that s 157(3) is not concerned with a case to which a “licence scheme” (as defined in s 136(1) of the Act) applies and nor, as mentioned, is CA engaged in its role as a statutory collecting society when engaging with Isentia and/or Meltwater or any other potential licensee of works in respect of which it has a mandate from the copyright owner.

32    As to s 157(3)(a) and (b), the first of the two alternatives (in circumstances where a person claims that they require a licence), is that a licensor has refused or failed to grant, or to procure the grant of the licence, and that in the circumstances it is unreasonable that the licence should not be granted: s 157(3)(a).

33    The second limb of s 157(3) is that a licensor proposes that the licence should be granted subject to the payment of charges, or to conditions, that are unreasonable: s 157(3)(b).

34    The case propounded by each applicant in these proceedings is that they require a licence to exercise the reproduction and communication right in literary works of the relevant publishers for the purpose of providing a “press monitoring service” and an “online monitoring service” (and, having regard to earlier licences, they required such a licence from 1 July 2018). They say that, in the case of Isentia, the licence proposed by CA to Isentia on 21 May 2018 contained terms relating to the payment of charges, and non-price conditions, that are unreasonable. That licence as so proposed, otherwise called the “CA Licence” (CA’s Further Amended Statement of Points in Answer, 15 December 2020 at [46]) was attached to an email and letter both dated 21 May 2018 from Mr Suckling (CA’s CEO) to Mr Croll (Isentia’s CEO at the relevant time).

35    Thus, Isentia contends that s 157(3)(b) is engaged. CA does not contend that the Tribunal’s jurisdiction is not properly enlivened.

36    Isentia also proposed a licence subject to the payment of charges and particular conditions. It contended that the terms and conditions as proposed were reasonable. It did so by the document annexed to its Statement of Points in Support of the Applicant’s Case dated 16 August 2019. By CA’s Statement of Points in Answer filed on 1 October 2019, CA confirmed that it would not grant a licence in the terms proposed by Isentia. Isentia claims that s 157(3)(a) is thus also engaged. Isentia has amended the terms and conditions of its proposed licence to take account of the findings set out in the July 2020 Tribunal decision. The terms of Isentia’s proposed licence are now set out in “Isentia’s Proposed Licence” annexed to Isentia’s Second Further Amended Statement of Points dated 1 December 2020: see [132]-[146] of that document. Isentia contends that the terms are reasonable.

37    It seeks an order under s 157(6B)(b) that it be granted a licence in the terms proposed by it.

38    Alternatively, it seeks an order specifying the “charges” and the “conditions” the Tribunal considers reasonable “in the circumstances in relation to Isentia”.

39    As to Meltwater, CA proposed to Meltwater on 18 May 2018, a licence in the terms of the “CA Licence” by the document described as Schedule 1 to CA’s Statement of Points in Answer as to which, see [33] of the document. Meltwater contends that the terms as to the payment of charges (which it says increases the fees payable by it by (redacted) are unreasonable and so too are the non-price conditions. Thus, s 157(3)(b) is said to be engaged. Meltwater also proposed a licence in the same terms, in effect, as the licence proposed by Isentia. That document is an annexure to Meltwater’s Further Amended Statement of Points in Support of the Applicant’s Case dated 1 December 2020; [49] and [49A]. It too says that the terms are reasonable.

40    Both Isentia and Meltwater contend that they have a well-founded claim pursuant to s 157(3)(b) because the 2018 CA Licence proposed to each of them (and which CA continues to support) contains terms as to the payment of charges and non-price conditions which are unreasonable. Alternatively, they both say that s 157(3)(a) is made good and well-founded.

41    It is convenient to now set out the text of s 157(3) and its relationship with s 157(6B) of the Act. The provisions are in these terms:

…

No licence scheme and licensor refuses or fails to grant reasonable licence

(3)    A person who claims that he or she requires a licence in a case to which a licence scheme does not apply (including a case where a licence scheme has not been formulated or is not in operation) and:

(a)    that a licensor has refused or failed to grant the licence, or to procure the grant of the licence, and that in the circumstances it is unreasonable that the licence should not be granted; or

(b)    that a licensor proposes that the licence should be granted subject to the payment of charges, or to conditions, that are unreasonable;

may apply to the Tribunal under this section.

…

Order dealing with application under subsection (2) or (3)

(6B)    If the Tribunal is satisfied that the claim of an applicant under subsection (2) or (3) is well-founded, the Tribunal must either:

(a)    make an order specifying, in respect of the matters specified in the order, the charges, if any, and the conditions, that the Tribunal considers reasonable in the circumstances in relation to the applicant; or

(b)    order that the applicant be granted a licence in the terms proposed by the applicant, the licensor concerned or another party to the application.

The ACCC guidelines

42    It is also convenient to note at this point s 157A of the Act in relation to the Tribunal’s obligation to have regard to guidelines issued by the Australian Competition and Consumer Commission (the “ACCC”) should a party to an application request the Tribunal to have regard to the guidelines. The applicants have made such a request. Section 157A is in these terms:

157A    Tribunal must have regard to ACCC guidelines on request

(1)    In making a decision on a reference or application under this Subdivision, the Tribunal must, if requested by a party to the reference or application, have regard to relevant guidelines (if any) made by the [ACCC].

(2)    To avoid doubt, subsection (1) does not prevent the Tribunal from having regard to other relevant matters in making a decision on a reference or application under this Subdivision.

43    Later in these decision reasons, the Tribunal will address the ACCC guidelines and their relevance for the purposes of these proceedings. We propose to examine the guidelines in a little detail as we understand that this is the first proceeding before the Tribunal where parties have requested the Tribunal to take the guidelines into account.

Streem

44    Streem Pty Ltd (“Streem”) is also a media intelligence organisation or media monitoring organisation.

45    On 29 May 2018, Streem was also offered a licence by CA in terms reflecting the “CA Licence” as described by Mr Forbes (Streem’s Commercial Director) as the “CA Proposed Licence”). Mr Forbes took the view that the CA Proposed Licence contained unreasonable terms. Mr Forbes sets out his criticism of the terms of CA’s Proposed Licence at [112]-[152] of his affidavit affirmed on 26 June 2020. That affidavit was filed in support of Streem’s application to the Tribunal under s 157(3)(b). On 9 September 2019, Streem served on CA a licence on terms and conditions proposed by it together with its Statement of Points in Support of its case in the Tribunal. On 4 June 2020, Streem was given leave to file an Amended Statement of Points in Support annexing an “Amended Proposed Licence” which is substantially in the same terms as the licence proposed by Isentia. Thus, s 157(3)(a) was also said to have been engaged.

Some procedural matters

46    Procedural orders were made by the Tribunal that the Isentia, Meltwater and Streem applications under s 157(3) be heard together. It will be necessary to examine later in these decision reasons the particular history of the licence arrangements (including the interim licence arrangements) between CA and Isentia and also CA and Meltwater (and aspects of the terms of the interim arrangements between CA and Streem).

47    The hearing of the three applications which engaged extensive evidence including expert opinion evidence was set down for hearing for three weeks commencing on Monday, 12 October 2020. The proceeding as between Streem and CA settled on the preceding Friday, 9 October 2020. The settlement of the Streem proceeding had a profound effect upon the proceedings generally.

48    On the morning of the second day of the hearing on Tuesday, 13 October 2020, CA contended that the arrangements between CA and Streem gave rise to a licence agreement which represented terms and conditions agreed upon as between CA and a media monitoring entity concerning the grant of a non-exclusive licence in relation to articles published in a printed or electronic edition of a nominated newspaper, magazine or Masthead publication to undertake acts described as (redacted). We will return to the scope of the grant later in these reasons.

49    The significance of CA and Streem having agreed to enter into the licence on 9 October 2020 as just described is that CA contended before the Tribunal on 12 October 2020 that the Streem licence is “very indicative” of a directly comparable market transaction to which the Tribunal ought to have regard in deciding the questions arising in the present proceedings under ss 157(3) and (6B) of the Act. CA maintained that contention throughout the proceeding.

Aspects of the Streem licence of 9 October 2020

50    It is convenient to identify some aspects of the licence.

51    The Licence Agreement entered into between CA and Streem, described as a “Full Service Media Monitoring Licence” (the “FSMM Licence”), provides for the payment of a “Licence Fee” (cl 4; Schedule 1, cl 1) as specified in Schedule 2. (redacted)

52    (redacted)

53    (redacted)

54    (redacted)

55    (redacted)

56    (redacted)

57    (redacted)

58    (redacted)

59    (redacted)

60    (redacted)

61    (redacted)

62    (redacted)

63    (redacted)

64    (redacted)

65    (redacted)

66    (redacted)

67    (redacted)

Three documents related to the Streem Licence of 9 October 2020

68    As to the FSMM Licence between CA and Streem, it became clear during the course of opening submissions on 12 October and 13 October 2020 that entry into that licence needed to be understood in the context of three other documents: a Deed of Settlement of the Streem proceedings dated 9 October 2020 (the “Settlement Deed”); a Deed Poll dated 9 October 2020 given by News Corp Australia to Streem; and a Deed Poll also dated 9 October 2020 given by Nine/Fairfax to Streem.

69    (redacted)

70    (redacted)

71    (redacted)

72    (redacted)

73    The background to the licence arrangements between CA and Streem is this.

74    In July 2017, CA and Streem entered into a Press Monitoring and Online Monitoring Licence which expired on 30 June 2018. On 29 May 2018, CA offered Streem the “CA Licence” (again described in CA’s proposal as a Press Monitoring and Online Monitoring Licence). As mentioned earlier, Streem contended that terms of that licence as to the payment of charges and other conditions of that licence were unreasonable, leading to Streem’s application to the Tribunal under s 157(3) of the Act.

75    On 11 July 2018, the Tribunal made interim orders under s 160 of the Act pending the final determination of Streem’s application. The interim orders granted Streem an interim licence applying the terms of the previous agreement (which had expired on 30 June 2018) on a continuing interim basis, commencing on 1 July 2018. The interim orders contained, put simply, an adjustment mechanism such that the terms of the licence as to the payment of charges as determined by final orders of the Tribunal would also apply during the interim period from 1 July 2018 with the result that if the final terms proved to be more favourable to CA than the prevailing interim terms, Streem would pay CA an amount representing the quantum of the difference, and should the final terms prove to be more favourable to Streem than the interim charges, CA would pay Streem the difference.

76    (redacted)

77    (redacted)

78    (redacted)

79    (redacted)

80    (redacted)

81    (redacted)

82    (redacted)

83    (redacted)

The procedural consequences of CA and Streem having entered into the Streem Licence, the Deed of Settlement and the Deed Polls

84    As a result of CA’s proposition put to the Tribunal on Monday, 12 October 2020, that it would seek to rely on the FSMM Licence struck with Streem (contextualised by the Settlement Deed and the two Deed Polls all of the same date) as significant evidence of an immediately comparable transaction reflecting a market price concerning participants in the same market in relation to a licence of the same subject matter, Isentia and Meltwater contended that they would need to take instructions on many aspects of the arrangements between CA and Streem and would likely want to put on further evidence including further expert evidence. Isentia and Meltwater made those enquiries and confirmed that they would wish to put on further evidence. CA contended that it would wish to respond to that evidence.

The 11 November 2020 CA proposed licences

85    In the result, the hearing was necessarily adjourned at the request of Isentia and Meltwater without objection from CA.

86    On 11 November 2020, CA filed a document described as: Alternative licence proposed by [CA] to Applicants.

87    By that document, CA put two separate proposals to Isentia and Meltwater for the purposes of s 157(3) of the Act. The two proposals are true “alternatives” in the sense that CA maintains its contention that the terms as to the payment of charges and the non-price conditions of the CA Licence (the subject of s 157(3)(b) contentions by each applicant) are “not unreasonable”.

88    Thus, there are now three proposals put to the applicants.

89    As to the 11 November 2020 proposals, the first alternative is that a licence be granted to each applicant (which is open to acceptance by either applicant independently of the other) on the terms of a Pro Forma Licence which, (redacted) (including a “Revenue Fee” for the purposes of Schedule 2, cl 2 of 23% of “Revenue”) for a term of two years commencing on 1 October 2020. CA contended in the 11 November 2020 document that if such a licence were to be granted in respect of that period, it would resolve each application but leave unresolved questions relating to the interim period up to 1 October 2020. The Tribunal had also made interim orders in the case of Isentia and Meltwater providing for an adjustment mechanism along the lines mentioned earlier concerning the Streem Interim Licence.

90    CA’s first proposal of 11 November 2020 is described in these proceedings as the Alternative CA Licence.

91    The second alternative of 11 November 2020 proposed that a licence be granted to each applicant (again open to acceptance by either or both independently of the other) on the terms of the Pro Forma Licence (redacted) but with the Schedule 2, cl 2 Revenue Fee (redacted) from 23% to (redacted) by operation of a proposed Deed of Settlement of the relevant proceeding in each case and that the applicant enter into a proposed Pro Forma Deed of Settlement of the relevant proceeding (redacted) except that any additional payment to be made in relation to the interim licences in the period from the grant of those interim licences to 1 October 2020 is to be either agreed between the parties or determined by the Tribunal.

92    The second alternative of 11 November 2020 is described in these proceedings as the Amended Alternative CA Licence.

93    The Amended Alternative CA Licence is a proposal for the grant of a licence on terms that the relevant proceeding be settled on the terms of a proposed deed with a question held over.

94    (redacted)

95    As to “assurances of the kind given (redacted) CA recites in the 11 November 2020 document that it “enjoys no guarantee of a perpetual right to licence this content” (para 6) and “such assurances can only be offered voluntarily by the publishers”: para 7. At para 8, CA recites that having made enquiries of the Publishers, CA “understands” that if Isentia or Meltwater accepted the terms of either of the two proposals of 11 November 2020, that is, the Alternative CA Licence or the Amended Alternative CA Licence, and similar assurances were sought, then those assurances would be given by each Publisher effective from 1 October 2020 on “substantially the same terms” as the respective (redacted) “and subject to any necessary adjustments to reflect individual supply arrangements”.

96    It follows that the Amended Alternative CA Licence proposal comprises a proposal to enter into a Pro Forma Deed of Settlement of the relevant proceeding coupled with the grant of, in effect, (redacted) as contemplated by the proposed Settlement Deed (annexing the proposed Pro Forma FSMM Licence) giving effect to (redacted) in the percentage of Revenue calculation of 23% to (redacted) and taking into account CA’s “understanding” that “assurances” “similar” to those (redacted) by the third party Publishers would be given, not on the same terms, but “substantially” on the same terms “subject to any necessary adjustments to reflect individual supply arrangements”. Those assurances, as so put, would endure presumably for the two year term of the Pro Forma Licence commencing 1 October 2020.

97    CA says that each of the two alternatives are made for the purposes of, and engaged by, s 157(3) of the Act and the terms as to the payment of charges and other non-price conditions of each alternative are “not unreasonable”.

98    Isentia and Meltwater contend that, in effect, CA has abandoned the “CA Licence” proposal and now, in truth, only relies upon the Alternative CA Licence and the Amended Alternative CA Licence. Each applicant says that the Amended Alternative CA Licence is nothing more than a settlement offer rather than a licence proposal. They say that the terms of the CA Licence to the extent that that licence remains a proposal at all are unreasonable and so too are the price and non-price terms of each of the two proposals of 11 November 2020.

99    The applicants urge the Tribunal to grant a licence to each of them on the terms and conditions proposed by each of them.

100    The CA alternative licence proposals of 11 November 2020 are taken up in CA’s Further Amended Statement of Points in Answer to each application. CA says this at [55V] of that document:

… the Alternative CA Licence, the amended Alternative CA Licence, the Streem Licence and the Streem Arrangements (as defined in paragraph 1(b) of the Comparison Document) are all comparable bargains. The Streem Licence and the Streem Licence as amended by the Streem Deed [of Settlement] represent contemporaneous bargains struck between a willing buyer and a willing seller such as to render those bargains the most up to date and applicable evidence of market value available to the Tribunal.

The written submissions

101    In these proceedings, there are very many issues of fact and expert opinion in contest going to the issues raised by s 157(3) and s 157(6B). Much evidence has been filed on the various issues. The parties have filed extensive written submissions supported by schedules such as CA’s schedules: Overview of Acts comprised in Copyright by Isentia; Copyright Acts in Meltwater’s Services; Overview of the Parties Positions on Non-Price Terms in the CA Licence and the Alternative CA Licence. Meltwater relies on two schedules cross-referenced to its closing submissions. The submissions and schedules are extensive, before also taking into account the folders of closing (and opening) tender bundles handed up in support of the submissions.

102    Apart from the primary submissions, the following additional material has been put to the Tribunal:

(a)    Isentia’s Reply Note and Corrections, 22 March 2021;

(b)    Isentia’s Evidence References in relation to its closing submissions, 22 March 2021;

(c)    Isentia’s Note in Reply to CA’s Oral Submissions, 1 April 2021;

(d)    Supplementary CA Note in Reply to Isentia’s Note dated 1 April 2021 (as adopted by Meltwater on 7 April 2021), 29 April 2021;

(e)    Isentia’s Note in Response to CA’s “Rejoinder Submissions” of 29 April 2021, 21 May 2021;

(f)    Meltwater’s Note in Reply to CA’s Note in Reply, 31 May 2021;

(g)    A joint submission by the parties in relation to aspects of the affidavit of Mr John Fairbairn dated 16 August 2021.

103    Against this background, we simply want to note one further preliminary matter. The Tribunal only asked one thing of the parties in these proceedings which was that the submissions on the many contested questions should, in effect, exhibit a degree of synthesis and reduction (to the extent possible) so as to enable the decision to be made and the decision reasons, explanatory of the decision, to be written as efficiently as possible. However, we now move on to address the many issues raised by the proceedings.

The Tribunal’s approach to determining the questions in issue

104    As to the method of deciding the questions in issue, the Tribunal has adopted the following approach. The Tribunal has considered the very extensive and detailed written submissions, heavily footnoted, supported by comprehensive oral submissions over three days from the parties. The written submissions advance the case sought to be made by each party and, point by counterpoint, extensively seek to answer each contention put against the party by its opponent.

105    We do not propose to develop a narrative of all of the detailed evidence we have heard. Rather, we propose to provide reasons for the decision we have reached on the key topics in dispute. We will, of course, address the evidence going to these topics and the relevant context. We do not propose to address in these decision reasons any evidence that has not been expressly relied upon by a party in the course of the written or oral submissions. The key topics are designed to be broad topics which capture the discussion before us and the various points of contention between the parties on so many fronts.

The key topics

106    The key broad topics which we propose to address are these:

(1)    The history of the licence arrangements between CA and Isentia and CA and Meltwater.

(2)    In that context, the 2016 Licence between CA and Isentia, the circumstances in which it came to be entered into, and its relevance.

(3)    Aspects of the statutory question to be determined by the Tribunal under s 157 of the Act.

(4)    The elements of the ACCC guidelines and their application.

(5)    The activities of and services offered by Isentia and Meltwater.

(6)    The question of whether the Streem FSMM Licence in the context of the arrangements between CA and Streem made on 9 October 2020 is a comparable market bargain or market rate.

(7)    The extent to which Dr Eisenach’s benchmarking exercise and robustness check is of assistance to the Tribunal in deciding the questions in issue.

(8)    The Alternative CA Licence and the Amended Alterative CA Licence and aspects of the evidence of Dr Pleatsikas.

(9)    The licence proposals of the Isentia and Meltwater.

(10)    The evidence of Mr Samuel and the critique of aspects of that evidence by Mr Ross.

(11)    The non-price conditions of the licences.

107    These topics, of course, overlap but we will seek to address each topic and the resolution of the issues raised by each topic in the context of the statutory questions to be determined by the Tribunal. There is nothing rigid about these categories.

108    Before turning to each of the topics the following further important matters of fact can be conveniently noted now.

The withdrawal of titles

109    On 23 January 2020, the solicitors for CA sent a letter to the solicitors for Isentia, Meltwater and Streem, referring to interim orders made and interim licences granted by the Tribunal on 23 May 2018, 11 July 2018 and 23 April 2019 in the Meltwater, Streem and Isentia proceedings respectively, advising that News Corp had notified CA that The Australian and The Weekend Australian would no longer be publications licensed by News Corp to CopyCo. Formal notification would follow once CopyCo gave notice to CA under the provisions of the Agency Agreement.

110    On 5 February 2020, formal notice was given by CA’s solicitors to Isentia, Meltwater and Streem, that CopyCo had notified it that as from 20 March 2020 CA would cease to have the right to sub-licence The Australian and the Weekend Australia “for press clipping or webscraping”. Thus, the source of the right to copy and communicate “press clips” of those titles or “webscrape” those titles was no longer to be found in a licence (sub-licence) from CA.

111    As well, by the same letter, CA notified the three MMOs that CopyCo had notified it that CA would cease from 17 January 2021 to have the right to sub-licence the following titles of Nine/Fairfax for “press clipping or webscraping”: The Australian Financial Review, The Financial Review Smart Investor, The Financial Review BOSS, The Australian Financial Review Magazine and The Weekend Financial Review.

112    Thus, the nominated News Corp and Nine Publishing titles were withdrawn from the interim licences ordered by the Tribunal as from 20 March 2020 and 17 January 2021, respectively. (redacted) CA does not contest the proposition that these titles are important titles for a licensee seeking to provide media monitoring services. (redacted)

113    (redacted)

114    (redacted)

115    (redacted)

116    (redacted)

117    (redacted)

118    (redacted)

119    (redacted)

120    (redacted) In closing submissions, Meltwater notes that (redacted). Meltwater’s submissions also note that (redacted)

121    (redacted)

122    (redacted)

Topic (1): The history of the licence arrangements between CA and Isentia and CA and Meltwater

123    The proposals put by CA in terms of the CA Licence to Isentia on 21 May 2018 and Meltwater on 18 May 2018 did not come out of thin air. There is a history of licence arrangements between CA and each applicant which provide examples of transactions reflecting price and non-price terms in relation to related subject matter as between the licensor and licensee which illustrate agreements CA and each applicant have entered into over time. It is necessary to recognise these various methods of licensing between the same parties and, where relevant, take them into account.

CA and Isentia

124    On 27 August 2001, CA and Isentia entered into an agreement described as: Press Clipping Service Licence: Newspaper and Magazines: Copying & Communication Rights (the “2001 Isentia Licence”). (redacted). The 2001 Isentia Licence provided for a payment of a licence fee of $1.00 per article communicated to the customer. It provided for annual increases in the $1.00 rate to take account of increases in the Consumer Price Index (“CPI”). (redacted)

125    The 2001 Isentia Licence contained an acknowledgement that articles (works) of a publisher might become “Excluded Works” and thus removed from the licence.

126    In 2001, CA and Isentia also entered into an agreement described as: Press Clipping Service Licence Newspapers and Magazines Fax Transmission Licence (the “2001 Isentia Fax Licence”). That licence provided for the payment of a fee by Isentia to CA based on a revenue percentage of (redacted) of the total amount invoiced by Isentia to its customers for transmitting articles to customers by facsimile during the previous quarter. The end of the process of transmitting material by facsimile resulted in no transmission and no further payment of fees under that licence.

127    The 2016 Isentia Licence addresses not only “press monitoring” but also “online monitoring”. The notion of “scraping” or “extracting” the works of publishers from electronic editions of Mastheads and communicating a portion of such works to a customer of Isentia was the subject of a previous agreement in 2014 (the “2014 Scraping Licence”). Under that licence, CA granted Isentia a non-exclusive licence to systematically copy or extract (that is, “scrape”) works of publishers (“Licensed Works”) from electronic editions of the “Mastheads” set out in Annexure A to the licence and to communicate a portion of the scraped work to Isentia’s customer together with a link to the publisher’s website where the customer (if they elected to engage with the link) could see the entire article displayed on the publisher’s website.

128    The term of the 2014 Isentia Scraping Licence was 1 July 2014 to 30 June 2015. It provided for (redacted). Notwithstanding the term, the agreement continued to operate until 30 June 2016. It too contained an “Excluded Works” provision.

129    Some clients of an MMO receiving press clippings as a result of Isentia’s upstream licence with CA would want to internally distribute and store the clippings either by email or by means of an intranet. In order to do so, the customer must also have a licence from CA to reproduce and communicate the works (solely within the customer’s organisation). These “downstream licences” were administered by Isentia on behalf of CA.

130    As mentioned, in 2016 Isentia and CA entered into a new licence for the period 1 July 2016 to 30 June 2018, the 2016 Isentia Licence. It is necessary to examine the terms of the licence and the circumstances in which it came to be concluded, for a number of reasons.

131    First, the licence represents a significant change to the charges for the grant of the licensed rights the subject of that agreement compared with the previously prevailing charges between CA and Isentia. That matter is material as an example of a licence between the parties conferring a grant of reproduction and communication rights over relevant subject matter relatively close in time to the licence CA proposed in 2018, the CA Licence. The 2016 Isentia Licence was in place for two years to 30 June 2018.

132    (redacted)

133    Third, the circumstances in which a significant change occurred to the basis for the terms as to charges to be paid by Isentia for the grant of a licence bear on the question of the extent to which the exclusive rights subsisting in the works of News Corp conferred by the Act find expression in the market positions taken by CA in the negotiations to obtain particular terms as to charges and, ultimately, the terms and conditions of the licence. Isentia and Meltwater contend that the exclusive rights conferred by the Act on News Corp (compounded, in effect, by aggregation of those rights by the publishers acting together through CA) give rise to a substantial degree of market power which was unmoderated by CA in its dealings with Isentia in relation to the 2016 Isentia Licence. Isentia and Meltwater contend that the terms as to charges and non-price terms of the proposals put by CA to each of them are unreasonable and represent the expression of the market power of News Corp which was aggregated with the market power of other CopyCo publishers.

134    Fourth, Mr Suckling and Mr Gray gave evidence about the factors leading to the position adopted by News Corp in CA’s negotiation of the 2016 Isentia Licence. In addition, Mr Suckling accepts that, put simply, the negotiations did not represent CA’s finest hour. (redacted). No attempt was made to identify a reasoned basis for the quantum of the fee demands made of Isentia which might have enable Isentia to assess the reasonableness of the proposal.

135    (redacted)

136    Before examining the 2016 Isentia Licence, aspects of the historical arrangements between CA and Meltwater ought to be noted.

CA and Meltwater

137    CA and Meltwater were parties to two licence agreements both of which expired on 30 November 2017. The first is an agreement described as: Scraping Licence. It commenced on 1 December 2014 and was amended on 7 October 2016. The second agreement is described as: Press Clipping Service Licence Newspapers and Magazines: Copying and Communication Rights (the “Press Clipping Agreement”). It was made on 12 October 2015.

138    (redacted)

139    (redacted)

140    Meltwater, in its service offerings when displaying the results of online monitoring to a customer, describes a portion of an article as an “ingress”. It represents the same thing as a portion as it consists of details of the publication in which the article appears (from which the extract is taken) and includes a limited amount of text and a link to the article.

141    (redacted)

142    (redacted)

143    (redacted)

144    (redacted)

145    (redacted)

146    (redacted)

147    The two licences expired on 30 November 2017. On 23 May 2018, the Tribunal made interim orders granting Meltwater a licence pending the determination of the proceeding on the terms of those earlier licences. (redacted)

The elements of the 2016 Isentia Licence

148    By this licence, CA grants Isentia a licence to copy and communicate Licensed Works for the purpose of providing a Press Monitoring Service and store those works for up to 12 months in a “Secure Portal”.

149    The licence authorises Isentia to make a “Scraped Copy” and to communicate a Portion to its customers (including by making that Portion available for up to 12 months in a Secure Portal) for the purpose of providing an Online Monitoring Service to its customers. This right includes the right to communicate a Portion to a customer with a link to the work on the relevant publisher’s website which will allow the customer to view the work including in circumstances where the work is behind a paywall and the customer has not separately subscribed for access.

150    By the agreement, Isentia is licensed to “create” a “searchable archive” of Licensed Works for access by Isentia’s customers of its Press Monitoring Service, and access to Portions by customers of its Online Monitoring Service, accessible in each case for up to 12 months.

151    The agreement licences Isentia to “use” the archive it has created to “display” Licensed Works or “parts” of such works to customers of its Press Monitoring Service, and to display Portions to customers of its Online Monitoring Service.

152    In order for Isentia to obtain PDF or other digital files of Licensed Works from publisher members of CopyCo, cl 3.2 makes plain that Isentia would need to “obtain appropriate access agreements with the relevant publishers”.

153    A “Licensed Work” means an Edition (as defined) and any other work which CA has been authorised to licence “other than an Excluded Work” and any work for which Isentia is licensed under another agreement (including agreements directly with publishers). A “Portion” means an extract from a single article or item published in a Licensed Work comprising any of the headline, citation and up to the greater of 50 words or 255 characters or the first sentence from the article or item. A “Scraped Copy” means a “reproduction” of a work in digital or other electronic machine-readable form obtained by “Scraping”. The term “Scraping” means the “systematic copying or extraction of electronic works by means of any automated or manual process”.

154    As to downstream licences, CA authorises Isentia to act as its agent in providing Isentia’s customers with a licence to make “downstream uses” of “Licensed Works” and “Portions” copied and communicated to them.

155    As to the Licence Fees, the total annual minimum licence fee was increased to (redacted) (excluding GST) made up of these components: a Press Monitoring Fee; an Online Monitoring Fee; and a Downstream Use Fee.

156    As to the Press Monitoring Fee (described as copying and communicating “full text print and digital print editions for [the] Press Monitoring Service”), the fees were these. For “press clips”, (redacted) multiplied by the number of clips supplied by Isentia to its customers annually (called the “variable fee clips”) if greater than the minimum number of press clips. For services described in the licence as the “subscription services” in relation to press monitoring, the fee was (redacted)

157    As to Online Monitoring (described as copying and communicating “portions from online editions for the Online Monitoring Service and downstream use by customers of content from online Editions”), a fee was payable of (redacted) (the “Minimum Fee Online”) or (redacted) of “Online Monitoring Revenue” (the “Variable Fee Online”) if greater than (redacted). The term “Online Monitoring Revenue” is defined in the licence to mean “per link revenue attributable to the Online Monitoring Service”. Recognising that the licence to scrape and communicate a Portion includes the right to communicate that Portion with a link to the work, the variable fee seems to be measured by the links communicated.

158    As to Downstream use (defined to mean “use by Isentia’s customers of content from print Editions”), the fee in relation to corporate customers was (redacted) (the “Minimum Fee Corporate Downstream”), or an amount based on the application of CA’s corporate rate card to Isentia’s customers (otherwise called the “Variable Corporate Downstream Fee”) if greater than the Minimum Fee Corporate Downstream. For government customers, the fee was (redacted) (the “Government Downstream Fee”) plus any amount received from government customers in excess (redacted) where such additional amount would not form part of the Minimum Total Annual Licence Fee: (redacted). The licence also provides that if CA licences an Isentia customer directly in respect of downstream use and the customer elects not to enter into a downstream licence with Isentia, the so-called Minimum Fee Corporate Downstream or the so-called Government Downstream Fee (whichever is applicable) is reduced by the amount of the customer’s downstream use.

159    The licence makes clear that if the sum of the components just described is greater than (redacted) the greater amount is the Licence Fee (plus the relevant excess, as just described), and if the same or less, the Licence Fee is (redacted) (plus the excess above (redacted) concerning rate card government customers, as just described).

160    (redacted)

161    The 2016 Isentia Licence is the agreement which prevailed between CA and Isentia immediately before the CA Licence proposal was put to Isentia. The circumstances in which the 2016 Isentia Licence came to be reached is a matter in controversy in these proceedings and we turn to that matter now.

Topic (2): The 2016 Isentia Licence and the circumstances in which it came to be negotiated and entered into

162    (redacted)

163    (redacted)

164    (redacted)

165    (redacted)

166    (redacted)

167    (redacted)

168    (redacted)

169    (redacted)

170    In June 2015, the former CEO of News Corp and Foxtel, Mr Kim Williams, became Chair of CA. In August, Mr Suckling was appointed CEO of CA having formerly been the Director of Policy, Corporate Affairs and Community Relations at News Corp. Mr McCaul was CA’s Director of Commercial Licensing.

171    (redacted)

172    (redacted)

173    (redacted)

174    (redacted)

175    (redacted)

176    (redacted)

177    (redacted)

178    (redacted)

179    (redacted)

180    (redacted)

181    (redacted) Isentia had other content licences including licences with: non-CopyCo publishers; broadcasters including substantially all television and radio broadcasters; licences with non-CA publisher members of print and online publications including The Guardian and the Australian Broadcasting Corporation; and content licences with social media networks.

182    (redacted)

183    (redacted)

184    (redacted)

185    (redacted)

186    (redacted)

187    (redacted)

188    (redacted)

189    (redacted)

190    (redacted)

191    (redacted)

192    (redacted)

193    (redacted)

194    (redacted)

195    (redacted)

196    (redacted)

197    (redacted)

198    (redacted)

199    Isentia again notes and emphasises Mr Suckling’s oral evidence that Mr McCaul’s response of 17 December 2015 reflected the position that CA had not itself undertaken any analysis of how that fee of approximately (redacted) for News Corp Mastheads was justified or supportable.

200    (redacted)

201    (redacted)

202    (redacted)

203    Again, Isentia notes Mr Suckling’s evidence that he did not know how the Nine/Fairfax number had been reached and nor did CA undertake any analysis of whether the figure was reasonable. (redacted)

204    (redacted). Again, Isentia notes Mr Suckling’s evidence on this topic to the effect that because he had been present at a CopyCo Board meeting with representatives of News Corp and Nine/Fairfax where discussion of the fact of increases sought from Isentia had occurred, to the best of his knowledge Nine/Fairfax knew that News Corp wanted an increase in their licence fees failing which News Corp was prepared to remove their mandate for their publications. However, Mr Suckling did not know whether Nine/Fairfax knew the quantum of the increase News Corp was seeking.

205    (redacted)

206    (redacted)

207    (redacted)

208    (redacted)

209    (redacted)

210    (redacted)

211    (redacted)

212    (redacted)

213    (redacted)

214    (redacted)

215    (redacted)

216    (redacted)

217    Isentia observes that Mr Croll’s account of the 2016 Isentia Licence negotiations was not challenged in cross-examination. Isentia also emphasises the evidence of Mr Croll as to the effect of the 2016 Isentia Licence. His oral evidence was this:

We lost a number of clients to Meltwater over the next 18 months. The price increases that we put into the market, there wasn’t that price elasticity in the market for clients to take on those increases. They hadn’t seen the value that we presented to them warranted a price increase. They looked for other providers in the marketplace at the time. And our churn increased significantly over that period of time, so we had to roll back the price increases. And it was a difficult competitive environment for us for the next period of time.

218    Isentia notes that the evidence of Mr Hickey for Meltwater was that Isentia’s 2016 licence fee was a significant price increase that caused them “irreparable harm in the market because they ended up with a [CA] agreement that was worse than ours”. (redacted) Isentia contends that while heightened competition from Meltwater and Streem and other factors played a role in these events, the circumstance that a (redacted) of Isentia’s revenue from all sources is expended on the licence with CA as a copyright cost, it would be unrealistic and thus unreasonable not to recognise the burden imposed upon Isentia by the CA licence fee and the impact it has had upon Isentia’s ability to operate a sustainable business.

Conclusions in relation to the negotiation of the 2016 Isentia Licence

219    The 2016 Isentia Licence does not assist the Tribunal in determining whether the terms as to the payment of charges contained in the CA Licence or the Alternative CA Licence or the Amended Alternative CA Licence are reasonable. The licence fee payable under the 2016 Isentia Licence to News Corp is the expression of the market power News Corp enjoyed in The Australian and The Weekend Australian to (redacted) unconstrained by rivalry and substitution possibilities. (redacted). Four other features of the chronology described above should be noted.

220    First, neither CA nor News Corp nor Nine Publishing explained the methodology it had applied to derive the number it sought, in any way that took account of any of the propositions advanced by Mr Croll.

221    Second, the negotiations with Isentia by CA on behalf of News Corp were conducted against the background of the threat by News Corp to withdraw access to the titles from the current arrangements.

222    Third, CA exercised no moderating voice of “reasonableness” in the negotiations but simply passed on the raw demands of News Corp and then Nine Publishing.

223    Fourth, CA exercised no voice of reason or reasonableness (redacted). Thus, a licence on the terms demanded by News Corp and also Nine Publishing had to be delivered.

Topic (3): Aspects of the statutory question to be determined by the Tribunal under s 157 of the Act

224    The scope of the statutory questions to be answered by the Tribunal begins and ends with the statutory text construed in the context of the Act taking into account the state of the jurisprudence aiding in the construction of the text.

225    Section 157(3) is set out earlier in these reasons at [41]. Section 157(3)(b) engages the question of whether the licence proposed by the licensor is subject to “the payment of charges” or to “conditions” that are “unreasonable”. Thus, the text is concerned with the relevant subject matter (charges and/or conditions) and whether the provisions are reasonable or unreasonable. Section 157(3)(a) is concerned with a licensor’s refusal or failure to grant the licence or to procure the grant of the licence, and that, in the circumstances, it is unreasonable that the licence should not be granted. Section 157(3)(a) (unlike s 157(3)(b)), expressly inquires whether “in the circumstances” it is unreasonable that the claimed licence should not be granted. Section 157(3)(b) is concerned with whether the particular charges or conditions (or both) are unreasonable (although the text of s 157(3)(b) does not use the phrase “in the circumstances”).

226    If the Tribunal is satisfied that the claims of the applicants under either limb of s 157(3) are “well-founded”, the Tribunal must either make an order specifying the charges and the conditions the Tribunal considers reasonable in the circumstances in relation to each of Isentia and Meltwater (s 157(6B)(a)) or order that each applicant be granted a licence in the terms proposed by, relevantly here, the applicant or the licensor: s 157(6B)(b).

227    So far as s 157(3)(b) is concerned, the Tribunal must determine whether any one or all of the three licences proposed by CA is or are subject to the payment of charges or non-price conditions that are unreasonable. The Tribunal must also consider whether CA has refused or failed to grant or to procure the grant of the licence each applicant claims it requires and whether in the circumstances it is unreasonable that the licence should not be granted.

228    We propose to consider the unreasonableness or otherwise of the impugned price and non-price provisions of each of the three CA proposed licences as we are required to do by s 157(3)(b) and also to consider whether it is unreasonable that the licence proposed by each applicant should not be granted. We propose to address the evidence going to each of those questions in the course of these decision reasons rather than sequentially. We note, however, that Isentia and Meltwater contend that they rely predominantly on s 157(3)(b) and thus we will focus on the challenge to the reasonableness of the terms as to the payment of charges and the non-price terms, but we also address the MMOs’ proposed licences in the circumstances prevailing between CA and each applicant.

229    We also note that although it was previously contended by CA that demonstrating a “well-founded” claim under limb (a) of s 157(3) confined the scope of a possible order under s 157(6B) and so too a claim made good under limb (b) confined the scope of a possible order under subsection (6B), it is now common ground that if the Tribunal is satisfied that either ground of s 157(3) is well-founded, the Tribunal may, in discharging the obligation under subsection (6B), make an order under either subparagraph (a) or (b) of subsection (6B) recognising that the Tribunal must, should the relevant state of satisfaction prevail, make an order under one of those subparagraphs.

The terms “unreasonable” and “reasonable”

230    The statutory text of s 157(3) and (6B) uses the terms “unreasonable” and “reasonable”. These terms are concepts of broad familiar use given content, however, in principle, by a number of considerations: the context in which the parties find themselves; the subject matter of the licence; the contended value attributed to rights granted by the licence; the content, burden and benefit cast upon the parties by the proposed terms as to the payment of charges and the non-price conditions; factors drawn from the history of previous dealings that may be relevant; and, no doubt, other matters, all informing an evaluative judgment of what is or is not reasonable including the proper approach to such questions reflected in the jurisprudence.

Background and role of market power

231    The applicants emphasise that the Copyright Bill (1968) for the Act was described by the then Attorney-General introducing the Bill, the Hon Nigel Bowen, as largely based on the work of the Copyright Law Review Committee (“CLRC” or “Committee”) chaired by Sir John Spicer. In the final CLRC Report, the Committee recommended that although one of the collecting societies, APRA, had proved to be a useful vehicle for both copyright owners of music and music users, it nevertheless controlled such a large amount of music that it could “fairly be described” as having “monopolistic control” over the public performance of musical works in Australia in which copyright subsisted. The Committee recognised a concern “expressed in many quarters” regarding “actual or potential abuse of this monopolistic power”. In the second reading speech for the Bill, the Attorney-General observed that in the musical field, the copyright owners had “so organised through licencing organisations” as to be “in a strong bargaining position” and “thus in a position to dictate the terms” on which music may be performed in public. The Attorney also recognised that the new Act, with the establishment of the role and jurisdiction of the Tribunal, would have an effect upon “substantial economic interests” built up under the existing law.

232    The Attorney was thus recognising that the conferral of a bundle of exclusive rights in relevant works coupled with the aggregation of the rights owners (as to music) for the exploitation of those rights could and did give rise to substantial economic interests. Aggregation was a particularly relevant matter in the case of musical works as it brought together in one place virtually all of the rights owners enjoying the rights in music capable of public performance in Australia. The critical role of aggregation was that it extinguished any possibility of supply side substitution and brought together a very wide group of rights owners who would otherwise have been rivals. As a result, a user seeking to exercise the public performance rights in preferred music could not just go to other music rights owners and obtain a licence to perform their works or a select number of musical works of particular copyright owners, should other particular copyright owners seek to extract terms thought to be unreasonable.

233    In circumstances where there is a narrow concentration of rights owners such as two critical publishers of print and online news, one or more of such entities may, without aggregation, be in a position to exercise a significant degree of market power derived from the exclusive rights they enjoy in the relevant works. If a licensee claims to require a licence to exercise rights in articles published in The Australian, no other print or online publication is a substitute because the demand is for content in articles published in that particular publication. The Parliament has conferred those exclusive rights on owners so as to provide incentives to originate works in which copyright is capable of subsisting. However, the conferral and subsistence of those rights in the context of the particular supply side and demand side dynamics of the transactions in which a person needs a licence and an owner has the capacity to impose or simply subject the licence to particular charges and conditions as a function of exclusivity, may give rise to charges and non-price conditions (adopted by the licensor, CA), which are properly characterised as the expression of market power. Such terms or conditions, so expressed, may be unreasonable.

234    It almost goes without saying that it is not the function of the Tribunal to determine, in a proxy or surrogate way, whether the terms as to charges and non-price conditions of a proposed licence are the expression of a misuse of market power in the sense of analysing conduct against the background of, for example, the statutory integers that would be engaged if the question was one of a contended contravention of s 46 of the Competition and Consumer Act 2010 (Cth). However, where copyright owners have elected to confer licence rights in their works on CA (as a licensor, as defined), the question of whether the terms as to the payment of charges or non-price conditions are “reasonable” or “unreasonable” is the statutory rubric by which the Tribunal is to form a view about the terms.

235    If the terms as to the payment of charges or as to conditions are properly characterised as the expression of the market power of one or more copyright owners effected through the agency of CA as a licensor for the purposes of s 157(3) of the Act, such terms, so expressed, may be “unreasonable” and the Tribunal is rightly engaged in examining that question. The circumstance that the relevant terms as framed reflect the emphatically held conceptions of a copyright owner (through the vehicle of CA), or one or more copyright owners, of a right to subject the licence to such terms because they own the copyright in the works, that circumstance is not and cannot be determinative of what is reasonable or unreasonable as to “the payment of charges” or as to “conditions” of the licence. The Tribunal must decide for itself what is “reasonable” or “unreasonable” and in doing so, the nature and value of the rights is, obviously enough, an important matter in that question. However, ownership of the copyright subsisting in the works and thus “exclusivity”, by itself, does not connote a right to render a licence “subject to the payment of charges, or to conditions, that are unreasonable”.

Incremental market power

236    CA contends that the Tribunal is only concerned with any “incremental market power” that CA may have by reason of the aggregation of the rights of multiple copyright holders.

237    To the extent that CA is in a position, by reason of aggregation, to subject a licence to the payment of charges or conditions that are unreasonable where it is said that those terms reflect an expression of CA’s market power so arising, the Tribunal must determine whether the impugned terms are unreasonable by reason of the exercise of that market power: that is, whether the claim is well-founded. Terms so expressed may be unreasonable. They are likely to be so in the absence of supply side rivalry where a supplier can act unconstrained by supply side substitution possibilities.

238    To the extent that CA finds itself subjecting a licence to terms going to the relevant subject matter dictated, for example, by a CopyCo publisher that has market power (such as News Corp and then Nine/Fairfax in the example of the 2016 Isentia Licence arrangements), and the terms are the expression of that market power (with CA acting as licensor for the purposes of s 157(3)), the statutory text is engaged and the Tribunal must decide whether the relevant terms so expressed are “unreasonable”. Again, they may be unreasonable in the absence of supply side substitution possibilities.

239    Thus, it is perfectly consistent with the statutory text that where a copyright owner engages CA to act in the capacity of a licensor (in terms of the definition in s 136(1) of the Act in respect of a licence on behalf of the owner or prospective owner of the copyright in the relevant work), the Tribunal may be called upon to determine whether a claim is well-founded to the extent that the relevant terms (charges and non-price conditions) of the licence are unreasonable as the expression of the exercise of market power by CA acting on behalf of the relevant owner. The text of s 157(3) and subparagraph (6B) is not subject to the qualification of incrementalism in the market power of the “licensor”.

How does the Tribunal go about deciding whether the relevant terms are reasonable or unreasonable?

240    The question then is how does the Tribunal go about deciding for itself whether the terms as to the payment of charges or non-price conditions are unreasonable? Generally, the Tribunal does so by examining whether there is a “going rate” which places a transactional “value” on that which is being supplied under the licence. If not, a “going rate”, there might be a particular licence transaction in the same field of rights which represents a contemporary comparable market bargain reflecting a transaction between a willing, but not anxious, licensor and a willing, but not anxious, licensee. The Streem transaction is said to be such a transaction. Alternatively, the Tribunal might seek by reference to the evidence overall to postulate a “notional bargain”, not expressly determined by an analogical particular transaction but a hypothetical or notional bargain which attributes the value a willing but not anxious licensor and a willing but not anxious licensee would ascribe to the rights being licensed. If all else fails, an “exercise in judicial estimation” (better described as an exercise in “judicial synthesis”) is called in aid: Re Applications by MCM Networking Pty Ltd (1989) 25 IPR 597 (“MCM Networking”), President Sheppard J, Members A Horton and Professor S Ricketson at 611; Reference by Phonographic Performance Company of Australia Ltd (2007) 73 IPR 162; [2007] ACopyT 1 (“PPCA 2007”), Deputy President Emmett J, Members Professor D Pearce and Dr R Smith, at [10] and [11].

241    Fundamentally, the Tribunal by these techniques seeks to benchmark the challenged provisions of the licence in the way described above by reference to at least some of those mechanisms but in the absence of identifiable, relevantly material, benchmarks, the Tribunal seeks to determine the reasonableness or otherwise of the impugned provisions by seeking to understand the methodology by which the licence fee (or other conditions) has (or have) been set, determined or calculated as appears in the licence instrument.

PPCA/Foxtel 2016

242    In Re Phonographic Performance Company of Australia Ltd Under Section 154(1) of the Copyright Act 1968 (Cth) (2016) 125 IPR 1; [2016] ACopyT 3 (“PPCA/Foxtel 2016”) (Deputy President Perram J and Member Professor McMillan), the Photographic Performance Company of Australia Ltd (“PPCA”) made a reference of a licence scheme to the Tribunal under s 154(1) of the Act for approval. The licence scheme concerned a scheme for the subscription television industry for the use of copyright sound recordings owned or controlled by the persons or entities represented by PPCA. The scheme, if approved, was to replace an existing agreement between the parties made in 2004. In its proposed terms, the licence scheme was to apply only to a subscription television provider supplying services exhibiting particular characteristics. As it transpired, there was only one supplier exhibiting those characteristics, Foxtel Management Pty Limited (“Foxtel”). Under the proposed licence scheme, PPCA sought to calculate the fee for use of the sound recordings by Foxtel based on Foxtel’s revenue rather than a payment calculated on a “per subscriber per month” (“PSPM”) basis. PPCA is a copyright collecting society which represents the interests of copyright licensors and registered artists including record companies and Australian recording artists (the owners or exclusive licensees for Australia of the copyright in commercially released sound recordings and associated cinematographic films) who have licensed their rights to PPCA. PPCA’s role involves granting licences authorising the exploitation of the sound recordings and music videos in its repertoire, including the public performance and forms of communication including broadcast. As a collecting society discharging that role, it receives and distributes licence fees for those uses to its licensors and registered artists.

243    The statutory jurisdiction exercised by the Tribunal in relation to such a reference is one of considering the licence scheme referred to it and making “such order, confirming or varying the scheme or substituting for the scheme another scheme proposed by one of the parties, as the Tribunal considers reasonable in the circumstances”: s 154(4).

244    Thus, the frame of reference is what is “reasonable in the circumstances”. That frame of reference required the Tribunal to consider the methodology adopted in the licence scheme for the payment of charges based on Foxtel’s revenue rather than the PSPM. At [35], the Tribunal took the view that the standard of what is “reasonable in the circumstances” required, in relation to the imposition of a pricing structure, that the remuneration be “reasonable or equitable” [emphasis added]. That is not, however, the language of the text. The Tribunal also said at [35] that in relation to non-price terms, the question was whether “the term is reasonable in the circumstances”.

245    As to the approach to the assessment of “equitable remuneration”, the Tribunal said this at [36]:

The approach of the Tribunal to the assessment of equitable remuneration in cases such as the present is fourfold. First, if a market price is available then that price will be imposed. Secondly, if no direct market price is available, then an attempt will be made to determine what bargain the parties might have reached in a hypothetical negotiation (on a willing but not over anxious basis). Thirdly, if this is not possible then the Tribunal will examine comparable transactions to see whether they can throw any light on price. Finally, if that cannot be done, the Tribunal will engage in a process of judicial estimation which will involve a synthesis of the relevant facts and circumstances into a rate which the Tribunal regards as reasonable or equitable in the circumstances: see as to these four matters, Reference by Phonographic Performance Company of Australia Ltd under s 154(1) of the Copyright Act 1968 (2007) 73 IPR 162; ACopyT 1.

[emphasis added]

246    These principles properly apply in determining what is reasonable for the purposes of s 157(3) of the Act. The above statement of principle from PPCA/Foxtel 2016 was affirmed in Phonographic Performance Company of Australia Limited v Copyright Tribunal of Australia (2019) 270 FCR 645 at [52]-[53], Besanko, Middleton and Burley JJ.

The reference to the decision in PPCA 2007

247    In PPCA/Foxtel 2016, the Tribunal relied upon the earlier observations of the Tribunal in PPCA 2007 where, at [10] and [11] the Tribunal said this:

[10]    In each reference under s 154, the tribunal must make a value judgment as to what it considers reasonable in the circumstances. It is not usually possible to calculate mathematically the correct licence fee in any particular case: Australasian Performing Right Association Ltd v Federation of Australian Radio Broadcasters Ltd (1999) 46 IPR 20 … at [11] and [19] (APRA v AFRB). Where approval of a scheme would lead to a substantial increase in fees, the increases can be phased in over a period of years rather than being introduced immediately. In the present case, the society proposes that there be a phasing in of the increase claimed by it.

[11]    In determining whether a proposed scheme, and the licence fee payable under it, are reasonable, a number of approaches might be adopted. The approaches include the following, which may overlap to a certain extent:

•    Market rate: the rate actually being charged for the same licence in the same market in similar circumstances.

•    Notional bargain rate: the rate on which the tribunal considers the parties would agree in a hypothetical negotiation, between a willing but not anxious licensor and a willing but not anxious licensee.

•    Comparable bargains: bargains not in the same market but sufficiently similar to such a notional bargain as to provide guidance to the tribunal.

•    Judicial estimation: the rate determined by the tribunal after taking into account a range of matters such as:

•    previous agreements or negotiations between the parties;

•    comparison with other jurisdictions;

•    comparison with rates set by other licensors, capacity to pay, value of the copyright material, the general public interest and the interests of consumers; and

•    administrative costs of a licensing body: see Audio-Visual Copyright Society Ltd v Foxtel Management Pty Ltd (No 4) (2006) 68 IPR 367 at [131] and [142].

248    CA identifies the following range of factors that may need to be taken into account in seeking to determine the reasonableness of the proposed licence fee. We accept that the identified factors reflect a field of factors which may need to be taken into account depending upon the circumstances of the case: MCM Networking 25 IPR 597 at 611; General Tire and Rubber Co v Firestone Tyre & Rubber Co Ltd [1975] 1 WLR 819; (“General Tire v Firestone Tyre”); Copyright Agency Ltd v Department of Education of New South Wales (1985) 4 IPR 5; 59 ALR 172 (“CAL v Education Department”); WEA Records Pty Ltd v Stereo FM Pty Ltd (1983) 48 ALR 91. The factors (other than those already mentioned above) are these:

[(a) to (d)] are matters already mentioned drawn from the authorities.

(e)    the nature and purpose of the business in question, and its place in or relationship to the market;

(f)    the nature and purpose of the copyright owners’ right to restrain the communication and reproduction of works;

(g)    whether the licence sought is exclusive or not (a non-exclusive licence leaves the door open to competition in the same market);

(h)    the nature of the use of copyright material;

(i)    the benefits of the use for the user, the copyright owner and the relevant industry;

(j)    [any] creative input of the licensee and the relative value of that contribution to the value of the works;

(k)    the value of neighbouring rights;

(l)    costs of production;

(m)    relevant legislative policy or provisions; and

(n)    the public interest. ...

249    CA contends that in these proceedings the question of the public interest has “particular resonance” as continuing access to high quality journalism from a number of news media publishing sources is said to be essential to the democracy.

Reasonableness does not connote exactitude

250    The evaluative judgment called for by s 154 of the Act was put this way in Re Phonographic Performance Company of Australia Ltd (under s 154 of Copyright Act 1968 (Cth)) (2016) 117 IPR 540; [2016] ACopyT 2 at [15], Deputy President Jagot J and Member Dr R Smith (which is also useful in relation to the Tribunal’s task under s 157(3)):

Consistent with our reasoning below in respect of selection bias, we do not consider that an approach that attempts to result in exact equivalence of payments under this scheme between the two possible payment methods or with the APRA – CRA Agreement is necessary or desirable. Such an approach involves an unjustifiable and inappropriate assumption that the process in which we are involved embodies values of accuracy, precision and exactitude. We consider it counter-productive to the interests of the parties to adopt any suggestion to this effect. The fact is that, of its nature, the task we must perform leaves open a wide range within which any selected value might be considered reasonable. If the parties and experts called by them do not accept the existence of this wide range then, as the course of this matter shows, the capacity for sensible debate about the range itself, and the factors that might assist in selecting from the range, is undermined. A commercial resolution outside the context of this Tribunal then becomes impossible. Instead, as occurred in this case, enormous expense is incurred in attempts to unearth every possible factor that might be relevant to value, no matter how insignificant. Equally of concern, no doubt also at great expense and with good intentions, experts are permitted to create economic models which, when all is said and done, merely translate a series of broad-brush evaluative assessments into a mathematical form. Such an approach tends to suggest that there is one answer to the problem, rather than a range of reasonable outcomes based on an evaluative assessment. In case it was suggested to the contrary, we wish to say that we do not consider it inappropriate that experts and parties expressly accept and adopt a range within which they believe the value should fall. To the contrary, we think that doing so would be beneficial. It would work against the artifice involved in the expression of singular values and the adverse consequences of such artifice – namely, the tendency to lose sight of the artifice and for a party to lock itself into a singular point value from which no deviation may be contemplated.

[emphasis added]

Unaffordability and incapacity to pay

251    In determining a reasonable licence fee, CA contends that if the rights the subject of the licence have the value contended for by CA, the circumstance that a licensee finds a fee deriving from that value “unaffordable” is no reason to reset the fee on the footing that those who CA represents ought to, in effect, subsidise the licensee. CA relies on the following statement of principle from PPCA 2007 at [200] which, in the context of [199], is in the following terms:

[199]    It is true that the proposed licence fee is very substantially higher than that presently charged by the society. However, in the absence of assistance from the nightclub respondents, the economic and financial consequences of the increase for nightclubs as a whole throughout Australia are a matter of pure speculation. If any inference is to be drawn from the state of the evidence, it is that evidence from the industry as a whole would not have supported the assertions by the nightclub respondents that the proposed licence fee would force large numbers of nightclubs to close.

[200]    The assertion, of course, is not to the point. If it be the fact that the right to play recorded music has the value claimed by the society, the fact that many nightclubs presently operating cannot afford to pay for that privilege is not a reason for them to be subsidised by those whom the society represents. Ultimately, market forces will operate. That is to say, one of the consequences of the introduction of the proposed licence fee may be to reduce significantly the number of nightclub operators who are prepared to pay the fee for the privilege of playing recorded music at their venues. Inefficient operators who are required to pay a market price for all the services they require in order to conduct their businesses may be forced out of business. That is the nature of a competitive market.

[emphasis added]

252    CA also relies upon an observation from MCM Networking to the effect that in determining what would be a reasonable fee in the circumstances, the licensee’s “capacity to pay” was not a relevant consideration. CA relies on the following passage at 25 IPR 597 at 631:

… MCM tendered a schedule in confidence which showed its financial results for the years it has been in operation. The purpose of the schedule was to show us that increasing rates of royalty could have a disastrous effect on the viability of MCM. We accept the evidence given about the figures by Mr McGinn and the figures themselves. But the fact is that MCM is in business. It is required to pay reasonable rates for all goods and services which it needs. It must pay a proper price or rent for premises, proper amounts for salaries and wages, appropriate amounts for equipment which it needs for its operations and so on. We see no reason why it should pay anything but a commercial rate for the music which it requires, which, after all, is at the heart of its operations. In other words, we are firmly of the view that the record companies ought not to have to carry MCM or, by receiving less than a fair rate of royalty, subsidise its operations. We are of the view that this is the approach which parties bargaining in the way we have supposed would approach the matter.

253    CA also relies on the Tribunal’s observation at [85] in PPCA/Foxtel 2016 that “generally speaking” Foxtel’s capacity to pay was not relevant to the fixing of remuneration.

254    CA also notes that in PPCA 2007 the Tribunal observed at [232] that in determining a reasonable fee payable by an entity characterised as a public instrumentality or charity, such bodies were not entitled to a licence “without paying a commercial price for the grant of it” and “[i]t is not the role of copyright owners to subsidise the activities of copyright users”.

255    CA, however, recognises that in the case of compulsory licences, the Tribunal has taken “capacity to pay” into account in determining a reasonable fee (Audio-Visual Copyright Society Ltd v New South Wales Department of School Education [1997] ACopyT 1; University of Newcastle v Audio-Visual Society Ltd [1999] ACopyT 2). Where the question is one of determining a price point which would be reached notionally between a willing but not anxious licensor and a willing but not anxious licensee (but taking into account for the purpose of the hypothesis “the actual position of the parties”), and being astute to not reject matters which “either side or both sides would necessarily and relevantly take [into] account when seeking agreement” (Re Phonographic Performance Company of Australia Limited (under s 154 of the Copyright Act 1968 (Cth)) (2015) 114 IPR 316; [2015] ACopyT 3 at [24] (“PPCA Commercial Radio 1”); General Tire v Firestone Tyre [1975] 1 WLR 819 at 833), the Tribunal has recognised that “if no prospective licensee can afford to pay the value claimed by the licensor so that an otherwise useful activity cannot be carried out at all, it tends to suggest that the rate claimed does not reflect a notional bargain into which the parties rationally could have entered”: PPCA Commercial Radio 1 at [23].

256    There seems little doubt that where there is a readily identifiable “going rate” or perhaps a readily “comparable bargain” that takes into account factors that bring that “relative concept” close to the circumstance in which the price setting task is to be undertaken, the circumstance that the rate is “unaffordable” for a particular licensee is not going to be relevant to the reasonableness of the fee. Where, however, the task falls outside the scope of determining a fee in those settings and engages a transaction where a licensor is seeking to reset the entire framework of the licence, address disruption due to loss of advertising revenue, increase the licence fee substantially as a percentage of revenue and bring to account in the calculus of all that is reasonable all of the factors that have been the subject of substantial evidence and lengthy submissions, the circumstance that the proposed reset may be substantially prejudicial at a foundational level to the business undertaking of each licensee, is plainly a relevant matter. The statements of general principle about “unaffordability” and “capacity to pay” have diminished relevance in that particular context, which is the relevant context in the proceedings at hand.

One further aspect of PPCA/Foxtel 2016

257    One other aspect of PPCA/Foxtel 2016 can be noted here about the difficulties presented in determining a so-called “market value” in circumstances where there is no supply side rivalry and thus no substitution possibilities. In the context of assessing “market value and the hypothetical bargain”, the Tribunal said this at [37]:

In the Tribunal’s opinion, there is no mechanism by which the market value of the rights that PPCA proposes to licence may be ascertained. This is principally because there is no market. In relation to sound recordings, PPCA’s licensors include the three major record labels, Universal, Warner and Sony, so that it is not practically possible to use a sound recording belonging to them without PPCA’s agreement. The Tribunal accepts that PPCA occupies the position of a monopolist in the market for the provision of such sound recordings. If it were not for the role of the Tribunal, PPCA would be able to raise its prices without suffering any adverse competitive consequences.

[emphasis added]

A workably competitive market

258    In determining whether the relevant provisions are reasonable, the Tribunal will take into account whether a contended market rate suffers from distortions brought about by market power in the price setting mechanism. The applicants contend that in determining whether a contended “market rate” as to charges (thus ascribing a market value to the rights) is reasonable, the rate must be one prevailing in a “workably competitive market” and thus rates nominated or set in circumstances where no supply side substitution occurs (characteristic of rivalry), are not likely to assist the Tribunal in deciding whether the rate so arising is “reasonable”.

The notion that the licence fee ought not be greater than a possible award of damages

259    Another feature of determining whether the proposed licence fee is reasonable, emphasised by the applicants (derived from authorities dealing with “equitable remuneration”), is the principle that the fee for exercising rights comprised in the copyright in the work should not be greater than the amount a court would assess as damages for the relevant act, if the act were an infringing act: see the discussion in WEA Records Pty Ltd v Stereo FM Pty Ltd (1983) 48 ALR 91 (“WEA Records”), Deputy President Lockhart J, Members RNJ Purvis QC and DK Malcolm QC, at 110; CAL v Education Department (1985) 4 IPR 5; 59 ALR 172 at 180-183.

260    These authorities recognise that a determination of equitable remuneration (or analogically that which is reasonable) “could not, of course, be greater than the amount of damages for infringement of copyright” (CAL v Education Department at 181 adopting WEA Records at 110), and then examine the various methods for making that assessment: sometimes, a loss of profit analysis is available where the “normal” profit (of a going concern rights owner) can thus be determined; sometimes, where licensing is the modality of the rights owner for monetising the rights, the prevailing royalty a user would have had to pay to act under the usual licences granted by the owner determine the measure of damages; sometimes, when neither of those approaches apply, an assessment of more general considerations is necessary recognising that the “ultimate process is one of judicial estimation of the available indications”: General Tire v Firestone Tyre and Rubber Co Ltd [1975] 1 WLR 819, Lord Wilberforce at 825-826.

261    All of these statements of principle have led in one form or other to the approach reflected at [245] and [247] of these reasons. These approaches are, in effect, techniques for measuring the value to the owner of the rights exercised, or to be exercised, by the licensee.

262    One feature of this approach that is apt to render the analysis of what is “reasonable” one sided (although not necessarily so as the relevant factors are properly synthesised) is that the orthodox monetary remedies for addressing an infringement of copyright are, at the election of the rights owner, damages or an account of the profits derived by the party exercising the right (who would otherwise be an infringer in the absence of a licence), but not both. However, when the statutory question is one of whether a proposed licence fee recited in a proposed licence is reasonable or unreasonable for the purposes of s 157(3) of the Act, the scope of the notion of what is reasonable would, as a matter of principle, take into account the value of the rights to the licensor (through CA standing in the shoes of the relevant owner) and the value of the exercise of the rights to the party claiming to require a licence. That bilateral consideration is the very point of inquiring into a price or fee which would represent something a willing but not anxious licensor and a willing but not anxious licensee would have agreed upon, as willingness to bargain (without anxiety) would take into account assessments made by each, of the value to each, of the rights the subject of the licence.

The notion that “consumption” of the article is the central matter

263    The applicants contend that a reasonable licence fee protects the publisher’s interest in receiving remuneration for the effort that has gone into creating newspaper articles which the publisher monetises principally through the sale of advertising and subscriptions. They say, in effect, that the act of copyright relevance which reflects the publisher’s interest in receiving remuneration is the consumption of an article (or, where relevant, a substantial part of an article) by an MMO’s customer and thus the act of communicating the article (or, where relevant, a substantial part) to the end customer is the act that ought to be remunerated by means of a licence fee and it is that focus which ought to determine whether the fee is reasonable or not.

The criticism of a granular deconstruction of the applicants’ services

264    The applicants also contend that an analysis which is focused upon deconstructing to a granular level each of the services offered by an MMO in order to isolate each and every technical process which may or may not result in copies of works occurring (or copies of a substantial part of a work occurring), made “along the way” to undertaking the relevant act of communication to the end user, misunderstands the substance of the licence relationship. They contend that each granular act ought not to be separately nominated for remuneration but rather the remuneration ought to take account of communication to the end customer as the dominant character and focus of the fee which appropriately accommodates other multiple steps along the path to communication to the customer.

265    One illustration of that principle is Copyright Agency Ltd v New South Wales (2013) 102 IPR 85; [2013] ACopyT 1, Acting President Perram J and Member C Riordan. In that case, the “LPI” Division of the State Department of Finance & Services supplied copies of registered survey plans to persons either by directly providing a physical copy over the counter or by providing them electronically from its online shop or by providing them online to intermediaries described as information brokers. The Tribunal found that each time the LPI Division made an electronic sale of a survey plan (particularly in the case of plans supplied to information brokers), “multiple acts of uploading, storage and sending” occurred. However, the Tribunal in determining the relevant rights fee observed that “there is no reason to treat these [multiple acts] as other than what they, in substance, are: that is the provision to a single user of a copy of a plan” [emphasis added]: Tribunal at [67]. As to the actual rate, the Tribunal determined that the same rights fee ought to apply whether the plan was supplied as a physical plan over the counter or whether supplied electronically (with the multiple acts previously described not ranking for separate remuneration).

266    CA takes an entirely different view about this matter. It contends that the correct approach must necessarily be to examine each and every one of the services offered by each applicant and subject those services to an analysis so as to determine the extent to which, in each service, each applicant exercises one or more acts of reproduction or communication, and to then bear all of that particular conduct engaging the rights of the owners in mind when deciding what is a reasonable fee for undertaking all of those acts.

The notion of “relevantly connected” revenue

267    As to percentages of revenue, the applicants say that an important consideration is that the rate (apart from the percentage itself, having regard to the evidence), is applied to revenue relevantly connected with the extent of use of the works, not just any or all sources of revenue of the applicant. The applicants seek to illustrate that point by reference to PPCA Commercial Radio 1 (Deputy President Jagot J and Member Dr R Smith). In that decision, the Tribunal considered a proposed licence scheme referred to the Tribunal under s 154 of the Act by PPCA. The scheme was concerned with the grant of licences to radio broadcasters to use sound recordings in simulcasts. The evidence demonstrated that on average approximately 10% of a radio station’s audience listened through a simulcast. Commercial Radio Australia Ltd (“CRA”) had also developed its own licence scheme and that scheme was also referred to the Tribunal. The Tribunal was asked to confirm PPCA’s proposed scheme which involved the payment of fees on a “per stream basis”. In the case of CRA, it proposed alternative schemes for licensing the same rights on the basis of a percentage of revenue with CRA acting as the agent for radio stations. At [84], the Tribunal accepted PPCA’s submission that it cannot sensibly be disputed that, all other factors being equal, “a rate that is closely tied to the actual extent of use of the copyright subject matter in the activity in question will be more reasonable and equitable than one that is not”. On the evidence, however, the Tribunal did not accept that all other factors were equal. The Tribunal also said this at [84]:

As discussed above, history and circumstance, which are important factors, weigh against the imposition on radio stations of a requirement that they be forced to implement a new system for payment of fees for use of sound recordings in simulcasts of their radio broadcasts given that they are already paying for the use of the sound recordings in their broadcasts on a percentage of revenue basis. We do not say this because of any concern about double counting. If there is double counting (discussed below), because radio stations pay on a total gross revenue (which includes revenue derived from simulcasting), an adjustment could be made to address this issue. We say this because, unlike the position in the US, there is a long history in Australia of radio stations paying for use of protected sound recordings in radio broadcasts on a percentage of revenue basis and, irrespective of this matter, they will continue to do so for the foreseeable time.

[emphasis added]

268    The applicants emphasise the following passages in the Tribunal’s decision:

[88]    We accept that, unlike the position in the US when per stream rates were adopted for the simulcasting of radio broadcasts, we do not have the freedom of a blank slate. In circumstances where radio has paid for the use of sound recordings in broadcasts on a percentage of revenue basis and will continue to do so for the foreseeable future, to impose a scheme which requires all radio stations to pay for the use of sound recordings on a per stream basis, in our view, would not be reasonable in the circumstances.

[89]    That said, if, as in the US, we were confronted with the need to determine a reasonable scheme for payment for the use of sound recordings in simulcasts where there was no arrangement already in place for payment for use of sound recordings in broadcasts, then the advantages of a per stream approach would be overwhelming and no consideration would need to be given to any role for a percentage of revenue approach.

[emphasis added]

269    Taking into account all those historical factors, the Tribunal created incentives for a fee to be paid on a volume-based per stream footing with radio stations having the choice of paying a fee on a percentage of revenue basis. The schemes were referred back to the parties and ultimately proposals were put to the Tribunal. In the subsequent proceedings, the Tribunal determined a quite modest percentage, according to the facts of the case, of 0.35% of gross revenue.

270    It is convenient to address at this point, as a matter of principle, the authorities relating to the calculation of a licence fee as a percentage of revenue (or in some cases as a percentage of expenditure).

271    Clearly enough, there is nothing unsound per se in calculating a licence fee (either entirely or in part) on the basis of a percentage of the revenue derived from an exercise of the relevant right subsisting in the copyright in the work. This is well illustrated where, for example, the owner of the copyright in a literary work (a book) receives remuneration as a percentage of the sale price of the book. The same approach is evident where an author (owner) of sheet music typically receives a percentage of sales of the work, and the owner of the copyright in a dramatic work typically receives a percentage of the box office receipts (ticket sales) for the public performance of the dramatic work. The same principle can apply where a broadcaster broadcasts the musical works of authors (represented by APRA) as a collecting society.

272    However, the example of a broadcaster illustrates how the question can become a little more complicated because not all music publicly broadcast remains protected by copyright and more importantly for present purposes, broadcasters provide a multiplicity of services some of which do not use any music at all and some use a little as mere background. Should the percentage apply to all revenue (or in the case of the ABC all appropriations representing its operational expenditures)? Should the rate be discounted to take account of programs (services) where no music is “used” (“talk” services, parliamentary broadcasts etc) and those services where very little use is made of music (introducing programs, closing programs, some theming). Alternatively, should revenue (in the case of the ABC, expenditure appropriations) be adjusted before applying the rate or should both approaches be adopted?

273    In Reference by Australasian Performing Right Association Ltd [APRA]; Re Australian Broadcasting Corporation [ABC] (1985) 5 IPR 449 (“Re APRA v ABC”), President Sheppard J; Member A Horton, some of these issues were addressed. The Tribunal accepted that the formula that had been in place of a lump sum did not reflect the true value of the APRA licence; that more often than not the remuneration to copyright owners is calculated upon the revenue the licensee/user receives for “the broadcast” or the “performance of the works”; that the adoption of a percentage of revenue (or expenditure) provides the parties with a formula which will continually reflect inflationary (or deflationary) trends; that in determining or fixing the percentage, the Tribunal attempts as far as reasonably possible, to suit the circumstances of the parties; that the considerations in doing so include the nature and extent of the licence, the need for the ABC to pay a commercial price for the right conferred, the circumstance that the commercial percentage should be regarded as a ceiling and the fact of the continuing trend within the broadcaster in programming to introduce “talk programs” and other non-music programs, needed to be taken into account: at 481-482.

274    Because of the special character of the ABC and the circumstance that its revenue is a function of parliamentary appropriations, the percentage was more conveniently applied to expenditure. As to that, the percentage was levied on “operating costs only” and not all expenditures.

275    Another general factor taken into account was the circumstance that the ABC’s costs included the cost of providing radio and television programs to remote parts of Australia. In the result, the Tribunal calculated the “operational costs” of the radio services and applied the percentage to those costs (which excluded the costs of parliamentary broadcasts, the costs of orchestras and concerts, merchandising costs and other costs not related to the broadcast of music), although the costs of news services and sporting services were not excluded from the calculation notwithstanding that these services did not use the licensed musical works.

276    As to television programming costs, those costs amounted to 40% of the total operating costs incurred in providing television services. The Tribunal regarded the appropriate percentage for use of the works in television broadcasts as 1.5%, and recognising that the programming costs were 40% of total costs, the appropriate rate was to be discounted to 0.9% (that is 60% of 1.5%). The Tribunal observed that “we think that the best way of making proper allowance for this deduction [the removal of the programming costs] is by taking it into account in the percentage which is fixed rather than in any deduction from the base upon which the percentage is charged”.

277    We mention the circumstances in APRA v ABC in a little detail simply as an illustration of the considerations in seeking to fix a reasonable licence fee where the licensee provides a multiplicity of services and where the issue of the fee payable to a copyright owner is not a simple linear one of work (book), sale, percentage of sale price.

278    Other examples are these.

279    In WEA Records, the Tribunal was required to determine the fee payable by the respondent for the right to broadcast published sound recordings (where the applicants were WEA and six other record companies and “PPCA”). The Tribunal concluded that there were obvious sound practical reasons why the assessment of the relevant amount was best expressed as a percentage of gross revenue: at 139. At that time, the Act contained a prohibition against the Tribunal requiring a commercial broadcaster to pay an amount exceeding 1% of the gross earnings of the broadcaster (a ceiling) in the relevant period. The Act, however, did not require the amount (the fee) to be calculated by reference to a percentage of gross revenue. Nevertheless, the Tribunal adopted that practical approach. The Act contained provisions concerned with the money value of gross earnings derived from broadcasting relevant “matter”. The rate applied was 0.45% having regard to “all relevant matters”. Although, of course, every case turns on its own considerations, WEA Records usefully illustrates some of the balancing factors. Two factors were said to be interdependencies and substitution possibilities.

280    As to the first, the Tribunal found that “airplay” of sound recordings was important to the record companies in promoting sales of records, and the supply of records and new releases was important to broadcasters in the conduct of their business. The Tribunal noted that the listening audience was related to the capacity to sell advertising attracted by the audience. The Tribunal concluded that in determining a fair and reasonable amount to be paid by the broadcaster, “it would be wrong, in our view, to ignore the benefit accruing to record companies by increased sales of sound recordings” as doing so “would confer an additional benefit on the relevant record company (the copyright owner) above that to which it is entitled from the statutory use of its property”: at 133; (described by the Tribunal as a “double benefit”, also at 133). The Tribunal concluded that the remuneration amount (the fee) “must be reduced on account of the benefit” accruing to the record companies by the broadcaster playing protected sound recordings. The Tribunal recognised that the extent of the reduction was “impossible to quantify … with any precision” and was to be taken into account as part of the “overall complex of relevant matters”: at 133.

281    The second factor considered by the Tribunal was the Tribunal’s acceptance as a “realistic consequence” that airplay acted as a substitute for the purchase of records due to the over-exposure on the radio of some sound recordings. That factor was also impossible to quantify in “precise terms” but would be weighed in the balance: at 134.

282    One of the recent authorities addressing this issue is PPCA/Foxtel 2016.

283    As already mentioned, PPCA had proposed a licence scheme for the subscription television industry for the use of sound recordings owned or controlled by PPCA represented entities. As it turned out, the proposed scheme only applied to Foxtel. The licence granted by PPCA was a “blanket licence” covering all sound recordings in its “vast repertoire” of over 40,000 licensor labels: at [7] and [8]. Previously, the licence fee had been calculated on a “per subscriber per month” (“PSPM”) basis. PPCA sought to change the pricing model to one based on revenues derived from Foxtel’s subscription television business. The Tribunal considered that a monthly charge based on the number of subscribers no longer reflected the revenues raised by Foxtel as there were “pay per view” services (among others) generating revenue from subscribers not included in the PSPM calculation (at [126]) and a remuneration structure based on the number of subscribers rather than the amounts they were actually paying for the activities undertaken did not reflect a “reasonable and equitable” pricing structure: at [126].

284    The Tribunal accepted that the PSPM model left out of account “other revenue streams” generated by Foxtel “by using [among other things] PPCA’s repertoire but for which it currently pays nothing” [emphasis added]. These other revenue streams included “expanded advertising revenues” and “ad hoc arrangements”: at [126]. Also, the Tribunal accepted that the “revenue complexion” of Foxtel’s business “had been changing and may continue to change” with the result that a revenue basis for licence payments was “far more likely to take account of the greater value that Foxtel can derive over time from its PPCA licence”: at [126].

285    All of these observations simply recognise that circumstances had changed such that Foxtel’s exercise of the licensed rights was producing options for subscribers resulting in revenue to Foxtel through the use of the works, not brought to account in the monthly charge per subscriber. However, as to Foxtel’s revenue streams that did not involve the exercise of any rights in sound recordings (such as “Foxtel Broadband”), Foxtel contended that PPCA ought not to share in any of its revenue from that service. PPCA seemed to accept that proposition as the Tribunal noted that the definition of “Gross Revenue” in the scheme document “only pick[ed] up” revenue related to the subscription television business. Other revenue streams outside that service were not available to PPCA for participation: at [128].

286    Another example of a service provided with little or no use of commercial sound recordings (which was discussed by the Tribunal in the context of a discussion about a decision of the Copyright Tribunal in the United Kingdom) was a “pay per view boxing match”. The Tribunal accepted “at least as a matter of impression”, that such a service would be unlikely to involve the use of sound recordings and that there seemed to be something “unfair about giving PPCA a cut of such revenue”. In the case of Foxtel, there were other pay per view services that did, however, engage a use of the licensed works: at [131].

287    A third example of activities which Foxtel contended ought to fall outside any revenue share with PPCA was Foxtel’s “entrepreneurial activities”. The contention was that PPCA ought not to share in the revenue if it does not share in Foxtel’s risk taking. The Tribunal thought that such an argument would have “greater purchase” if the proposed fee was levied on “profits”: no profits, no fee; at [132]. The Tribunal observed that the PSPM fee structure was itself a “revenue-based model” in that the fee was calculated on revenues from subscribers but failed to capture other forms of revenue “related to Foxtel’s subscription television business” and under either the revenue model or the PSPM model, PPCA was not exposed to losses or profits of Foxtel: at [132]. The Tribunal then repeats the point previously it had made that not to widen the revenue captured beyond that raised from subscribers would permit Foxtel to generate revenue “by using PPCA’s sound recordings without compensation”: at [132].

288    There is no suggestion in these observations of the Tribunal that broadening the revenue capture to include revenue from services using the licensed works, not otherwise captured from subscribers on a PSPM basis, provides a basis for capturing revenue from Foxtel’s “entrepreneurial services” not using the licensed works. In fact, the examples discussed by the Tribunal suggest otherwise.

Some qualitative factors

289    The further point the applicants make on this topic is that in determining a rate that is “closely tied to the actual extent of use of the copyright subject matter in the activity in question”, the Tribunal will not simply take into account a numerical calculation of how many times a work is communicated or reproduced but weigh in the assessment of the rate whether the whole of a work is reproduced or whether only a part of a work is used in the relevant activity, and if a part only, whether that part is a substantial part.

290    Ultimately, on that basis, the Tribunal is required to decide what is the actual extent of use; how does it arise; does it engage a reproduction of the whole or a substantial part; if not the whole (and a percentage of revenue fee is thought to be reasonable), what ought the rate be and how is it to be discounted (if at all) if the quality of use is a part not amounting to a use of a substantial part? Moreover, if the imperative (of reasonableness) is to ensure that the rate is applied to revenue derived from the “actual extent” of the use of the works (perhaps otherwise understood as “directly related revenue”), to what extent is the rate to be discounted (if at all) to recognise that revenue of a licensee includes revenue from other sources not directly related to use of the works, and especially where “bundling” occurs.

Statutory limits in the language of s 154(4) of the Act

291    One further aspect of PPCA 2015 Commercial Radio 1 should be presently noted.

292    The Tribunal concluded that none of the licence schemes put to it were “reasonable in the circumstances”. That conclusion presented the Tribunal with a difficulty in terms of the statutory language of s 154 of the Act. Section 154(4) as mentioned earlier conferred a power on the Tribunal to make such order “confirming or varying the scheme or substituting for the scheme another scheme proposed by one of the parties, as the Tribunal considers reasonable in the circumstances” [emphasis added]. The limiting language of s 154(4) is the language of “confirming”, “varying” and “substituting another scheme proposed by one of the parties”. The Tribunal took the view, applying earlier authorities concerning s 154 of the Act, that it did not have jurisdiction to “substitute a scheme of an entirely different kind” in light of that language and nor did it have the power to construct and thus determine a scheme it considered would be reasonable in the circumstances: PPCA 2015 Commercial Radio 1 at [35]-[37] and [332]. That, however, is not the language of s 157(6B) which provides that if the Tribunal is satisfied that the claim of an applicant under s 157(3) is well-founded, the Tribunal must make one or other of the orders contemplated by s 157(6B)(a) or (b), and subparagraph (a) confers power to make an order specifying, the charges, if any, and the conditions, that the Tribunal considers reasonable in the circumstances in relation to the applicant. The scope of the power is conferred in more broad terms than s 154(4) largely because s 157(3) and (6B) are concerned with an individual licence with a particular claimant for that licence rather than a “licence scheme” of more general application.

Some aspects of CA’s contentions: the relevant licences for the purposes of the proceeding

293    CA contends that the Tribunal can “conveniently proceed” on the basis that CA’s “principal offer” to the applicants is the Alternative CA Licence read “in the context of”, which we understand to mean, read “together with”, the Deed Polls offered by News Corp and Nine Publishing. However, the CA Licence is not abandoned. We have already noted the structural elements of the proposed Alternative CA Licence and, more particularly, the way in which the Amended Alternative CA Licence is put, related as it is, to the Pro Forma Deed of Settlement. Importantly, the fee proposal of the Alternative CA Licence is a revenue component of 23% of Revenue as defined. The Amended Alternative CA Licence, within the matrix within which it operates, is (redacted) of Revenue.

Digital engagement and decline in advertising revenue

294    CA contends that the decline in “press clippings” generally as a result of a decline in print publications coupled with behavioural changes in the way in which information is consumed in the community, has resulted in a “massive expansion” of online published content which is the “valuable content” of the publishers that the MMOs choose to monitor and communicate to their customers. CA says that notwithstanding the withdrawal of titles described earlier (at [109] to [122]), CA continues to licence “major masthead” news publications and a large number of “high-quality non-masthead” publications. CA also contends that a credible factor going to the reasonableness of the 23% of Revenue fee (redacted) is the circumstance that a decline in print publishing and a corresponding movement of advertisers away from print publishing has meant that revenue which previously, in effect, cross-subsidised the cost of production of news content has fallen away such that publishers must raise their prices for content in order to receive “full value” for the content and to “cover the high fixed costs of journalism and publication”. CA contends that the “value” of the content is necessarily “affected by” the prices at which publishers, as sellers, are willing to supply their content with the result that as “fundamental changes” have occurred in the publishing industry, those prices have gone up.

295    One factor in the methodology for determining the “value” of the content undoubtedly takes into account the price for which a willing but not anxious publisher is willing to licence its content but the determination of value also takes into account the price a willing but not anxious licensee will pay for that content. The applicants say that disruption to orthodox business models of the publishers requires flexibility and response but it is not “reasonable” for the purposes of s 157(3) for a publisher to set a licence fee for particular uses of content that seeks to recoup in some significant measure lost advertising revenue, because that measure of that loss is not a corresponding reasonable measure of the use of that content in the business circumstances of each applicant.

The centrality of ingestion

296    CA contests that notion and says that each applicant engages in an act of ingestion of online publisher content which involves copying the full text of works from the websites of publishers and storing those copies on their own databases (servers). CA says that the act of ingestion engages an act of reproduction which is “critical” to the ability of each applicant to monitor the online content, prepare extracts for communication to customers and supply those extracts (portions) to customers. CA contends that there are “numerous and constant acts of copying” involved in the step of ingestion, and the “centrality of ingestion” to the services provided by the applicants to their customers is an important element in valuing “accurately and on a fair commercial basis” the rights being exercised by Isentia and Meltwater. (redacted). The position is a little different for Isentia. We will come to that matter later in the discussion of that topic.

297    As to a licence fee set by reference to a percentage of the revenue of each applicant, CA contends that, in principle, there is nothing unreasonable about a revenue-based fee. CA contends that a fee based on a percentage of revenue is appropriate (reasonable) in the case of each of its proposed licences because the underlying content and the rights licensed to each applicant is central to each of the services each applicant provides and neither applicant could provide their media monitoring services to customers without a licence to copy and communicate copyright works through the activities of press clipping, ingestion and the supply of portions. CA says that the value of the licence must necessarily reflect those circumstances. CA says that the services such as the “analytics services” of each MMO “always use” copied content and often the provision of the services involves “further or other acts comprised in the copyright”. More fundamentally, CA says that the analytics services are “only made possible” by the initial act of copying (ingestion) which creates the media database that MMOs “then use” to provide such services. CA says that it is not seeking a share of revenue “unrelated to the rights it is licensing”. Rather, it says that the rights licensed by CA are the “essential foundation of, and used directly in” various services supplied by each applicant beyond basic media monitoring, including the analytics services of each applicant. CA says that those services could not be provided without the anterior actions of each applicant in copying copyright content by ingestion and when attributing value to the licence is undertaken, the analysis must necessarily account for the value it enables each applicant to realise which includes the revenues they earn from their analytics services. CA also contends that a fee based on a percentage of revenue reflects the “mutuality of interest” of publishers and MMOs in the production and distribution of high quality content reflecting the “interdependencies between publishers and MMOs”.

298    It needs to be kept in mind that the transaction in question is a licence to exercise rights comprised in the copyright on terms as to charges and conditions that are reasonable, not the determination of the commercial terms of a joint venture.

299    CA says that to the extent that either applicant derives revenue from services that do not engage the exercise of a reproduction or communication right, each applicant has the “capacity” to “unbundle” the services detaching those services that “depend” on the works from those that do not.

300    (redacted)

301    CA uses other emphatic language to describe the “centrality of ingestion” to the services of each applicant: for example, the copyright content is the “bedrock” for “all the MMOs “activities and output”; “all MMOs activities” use the initial copy ingested and reproduced; the value of the MMO’s business derives from access to the “universe of content”; the MMO generate revenue from their “pervasive and extensive use of copyright content”. These propositions and the factual contention they convey give rise to CA’s following proposition. CA contends that “[a]ll purposes licensed and uses undertaken with respect to the content by the MMOs, whether acts comprised in the copyright or not, are relevant to assessing the quantum of the licence fees” [emphasis added].

302    CA’s proposition is a “but for” test to be applied in determining a reasonable licence fee. In other words, “but for” the initial act of ingestion, neither applicant would be able to provide any of its services that engage with the licensed works (whether in analysing trends, writing commentaries or writing specific reports for customers about topics of currency in the media about them or otherwise), and even though some uses of the content do not involve any act comprised in the copyright of the publisher, it is nevertheless commercially reasonable that the licence ought to provide the publisher with a fee of either 23% or (redacted) of the “Revenue” derived by each applicant from all sources including those services (subject to a licensee unbundling the services if it is able to do so). That is said to be so, and thus reasonable, even though the publisher has not contributed to the actual value-added component of the MMO in writing the analysis, isolating the trends, writing the report or undertaking the interpretive tasks sought by the customer.

303    We will put these contentions in context when referring to the services offered by each applicant as one of the topics to be addressed in these decision reasons. We have already mentioned briefly that evidence of the Publishers and particularly the evidence of News Corp concerning the effect of disruption to the historical business model in which advertisers contributed significantly to revenue. Now, much of that advertising revenue is directed to the major platforms (Google and Facebook). (redacted) We note the revenue data over recent years and the period of the transition. We also note the challenge that this has presented. (redacted) We have also noted the challenges presented by the use of news alerts and news services by Facebook and Google. We also note the steps taken by the Commonwealth government to seek to encourage each major platform to contribute reliable annual revenue to the Publishers for the use of their content in those alerts and other news services. Later in these reasons, we note the recent agreements made with the platforms and the annual revenue contributions over the term of those arrangements. We will not set out that data here. We address it later.

Topic (4): The elements of the ACCC guidelines and aspects of the issues in the present proceeding

304    The ACCC has developed guidelines to assist in the determination of reasonable copyright remuneration in proceedings relating to voluntary licences and licence schemes before the Tribunal and to assist users and collecting societies in determining reasonable and equitable remuneration. The guidelines recognise that arrangements might exist where copyright owners have come together to be represented collectively in circumstances where they might otherwise have been in competition. The guidelines also recognise that the Tribunal is “intended to act as a constraint on the exercise of market power” in such circumstances by making decisions with respect to the reasonable terms and conditions of copyright licences, including licence fees. The focus of the pricing principles in the guidelines is upon constructing a hypothetical bargain between a licensor and a licensee “with equal bargaining power” as a “practical means” of determining a price that reasonably reflects an equal bargaining position having regard to the current regulatory environment and the practices of the Tribunal.

305    Part A of the guidelines addresses the economics of copyright and copyright licensing.

306    Part B of the guidelines addresses “pricing principles”.

307    Although the Tribunal has taken the entirety of the guidelines into account, we particularly note these matters drawn from the guidelines (together with comments in relation to aspects of these proceedings).

(1)    Generally, the best way to ensure that the resources of an economy are put to their most efficient use is through open competitive markets that maximise welfare. However, markets may fail to promote efficiency and welfare. Reducing the financial incentives for the creation of new works is not a desirable outcome where, for example, in unregulated markets the non-exclusionary nature of the good might reduce the financial incentives to a point where creators and producers cannot expect to secure an adequate return on investments: cl 2.1.

(2)    However, copyright can go some way towards addressing a “market failure” of this kind by conferring a bundle of economic (and legal) rights in the works of authors (the copyright material) which enables owners to exclude potential users and enhances the creator’s ability to receive remuneration and thus enhances the incentives for the production of content. The copyright system involves finding the right balance between the benefits of enhancing incentives for the creation of copyright material with the cost of restricting access to that material: cl 2.1.

(3)    Even though the ownership of the copyright in particular material confers a power to exclude others, much of the material in which copyright subsists will be of small value and thus the transaction costs of administering the rights, negotiating licences and taking enforcement action will be “high”. High transaction costs can cause market failure if they deter otherwise beneficial transactions. One solution to high transaction costs in licensing is to collectively manage licensing arrangements through collecting societies: cl 2.1.2.

(4)    So far as these applications to the Tribunal are concerned, CA, apart from representing a number of the CopyCo publishers, represents many non-CopyCo publishers (more than 3,500 publishers and over 17,000 publications which it is authorised to licence). The transaction costs of MMOs dealing individually with those content owners, and reciprocally those content owners seeking to deal individually with MMOs would rapidly exhaust the value of the transaction for each side. However, News Corp, Nine Publishing and Seven West Media are examples of corporations not so constrained by transaction costs in the choices they make about dealing with MMOs. (redacted) Thus, it can be seen that those publishers are in a position to control and determine whether or not they deal directly with the MMOs or whether they choose to licence some only of their works directly (such as the premium or high reputation or high demand titles as described earlier under the heading “Withdrawal of titles”), and licence other titles through CA. However, there are no orthodox substitution possibilities available to the MMOs in the sense of an opportunity to turn to the publishers as a rivalrous constraint on the conduct of CA should CA seek to extract an unreasonable rent for the grant of the rights. As the negotiations for the 2016 Isentia Licence reveal, CA acted as an intermediary to reinforce the monopoly rent seeking of the publishers rather than each acting as a constraint upon the other. Publishers such as News Corp and Nine Publishing can elect, and have elected, to supply independently of CA to the MMOs but those dealings are not a function of choices made by the MMOs as an exercise in substitutability.

(5)    Returning to the guidelines, the ACCC recognises that although the solution to high transaction costs can be found in collective licensing arrangements, the benefits of enabling copyright owners to act as one, through a collecting society, must be measured against the costs associated with the creation and possible exercise of market power: cl 3.1.1. The guidelines note that by “enabling competitors to act collectively, licence fees will potentially be higher, and other licensing terms and conditions more restrictive, than would be the case if competition between copyright owners was retained”: cl 3.1.1. The ACCC notes in the guidelines that in the current copyright regulatory framework, the competitive constraint offered by “direct licensing” on the prices and terms set by collecting societies in relation to “blanket licences” is likely to be limited by: “existing collective licensing arrangements” (which may prevent or limit direct licensing); and, the absence of effective mechanisms for “reducing blanket licence fees to take account of any direct dealing”.

(6)    These observations are focused upon many of the well-known collective licensing arrangements (such as those with APRA, PPCA and others). However, they also apply to other collective arrangements by which the owners of literary works have come together and conferred power whether through CopyCo or otherwise upon a collecting society (CA) to grant sub-licences in respect of their rights (and in the case of CA as earlier noted, there are many publishers and 17,000 publications). The guidelines suggest that the feasibility of direct licensing acting as a competitive constraint on the conduct of a collecting society is to be assessed essentially in three broad circumstances: (i)  licensees with predictable rights usage requirements who, if direct licensing is available and can be accessed efficiently, would only use direct licensing and not acquire a blanket licence; (ii)  licensees with semi-predictable rights usage requirements who may use direct licensing in certain circumstances; and (iii)  users who have unpredictable usage requirements and are unlikely to utilise a direct licensing mechanism. As to those categories, the guidelines contain this observation:

Where direct licensing arrangements are available they are a particular constraint on collecting societies in relation to the prices and terms of their licences offered to the first category of users, as the direct licence will be a close substitute for a blanket licence (provided there is a sufficiently large number of users within this category). However, direct licensing in relation to the second category of users would only be a direct competitive constraint on the prices and terms of licences set by collecting societies if the price paid by a user for a blanket licence was reduced to take account of any copyright material that the user accessed through direct licensing. On the other hand, direct licensing would generally not act as a significant constraint on the price and terms of licences set by collecting societies in relation to the third category of users except in limited circumstances.

The ACCC considers that where direct licensing is used and a blanket licence is still required for the remainder of the licensee’s usage, that user’s blanket licence fee should be reduced. …

[emphasis added]

(1)    Inherent in this discussion is the notion that if a person requires a licence to exercise, for example, the reproduction right or the public performance right in a work forming part of a group of works collectively licensed and the fee for the licence is $X, the fee ought to be reduced if the user has secured a direct licence to particular works which would otherwise have been licensed collectively. In other words, in the context of the blanket licence as discussed in the guidelines, there ought to be adjustments in the licence fee. As to the adjustment to a blanket licence fee, the guidelines contain this observation (at 3.2.2):

Even if appropriate opt-out or license back arrangements are in place, incentives to undertake direct licensing may be dampened if there is not an effective mechanism for adjusting a blanket licence fee to take into account the value of copyright material that is licensed directly. If such a mechanism is absent, users who engage in direct licensing, yet also need to obtain a blanket licence for the remainder of their usage requirements, may effectively be paying twice for the use of some material.

In order to ensure that direct licensing is promoted where it is efficient and feasible, the ACCC considers that blanket licence fees should be reduced to reflect the licensee’s lower willingness to pay for the remaining repertoire that excludes the works that have been directly licensed. …

(2)    In the present proceeding, the applicants contend that one vice of the CA proposal for a percentage of revenue fee structure is that the licence fee does not adjust when works in the CA repertoire are also the subject of a direct licence. The applicants also observe that the example of the negotiation of the 2016 Isentia Licence demonstrates that where two large publishers control the most significant part of the entire CA repertoire, the opportunity available to the publishers to withdraw from the CA licensing arrangements and revert to direct licensing exerted upwards pressure on the calculation of the CA licence fee. The applicants contend that when News Corp made it clear that it would withdraw particular titles from the CA arrangements unless the increase in licence fees was secured, CA demanded a large increase in the entire repertoire in order to ensure that distributions to News Corp from the licence fees would accommodate the demands of News Corp for an increase in distributions to it. In the result, Isentia accepted the licence as put to it. After the withdrawal of the particular titles described earlier, Isentia is confronting a demand for licence fees reflected in the three proposals (redacted)

(3)    As to Part B which is concerned with pricing principles, cl 4.1 of the guidelines makes the general point that the focus of pricing principles ought to be “countering market power”. At cl 4.2.1, the guidelines contain this observation:

The ACCC notes that the Copyright Tribunal has considered a range of approaches in making pricing determinations, including benchmarking, construction of a hypothetical bargain and judicial estimation (which takes into account a range of factors).

Of these broad approaches, the ACCC recommends the following methods as appropriate for pricing copyright materials, if feasible in the circumstances of the particular case:

•    benchmarking – using appropriate rates (which should preferably, as far as possible, be grounded in a more competitive market) as a source of information for determining remuneration

•    construction – constructing a hypothetical bargain between a willing, but not anxious, licensor and a willing, but not anxious, licensee by applying an economic model to construct an appropriate licence fee level.

Where possible, both approaches should be used as a cross-check against one another. If there are substantial differences between prices determined using the two approaches, then further investigation may be required.

[emphasis added]

(4)    As to benchmarking, the guidelines observe that appropriate benchmarks may include the existing rate for the licensing of the copyright material determined by previous negotiations or by previous determinations; rates or tariffs paid for use of the same copyright material in different uses; rates or tariffs paid for similar copyright material in other jurisdictions; and/or rates or tariffs paid in comparable, but more competitive markets: cl 5.1. The guidelines observe that in the proceedings regarding the licence fee for nightclubs, the Tribunal considered the market rate to be “the rate actually being charged for the same licence in the same market in similar circumstances” and the Tribunal regards that conception as what it describes as the “existing rate”. As to that rate, in the context of benchmarking, the guidelines make this observation and contain this note of caution at cl 5.2:

Existing rates for the particular copyright material in question may be an appropriate benchmark for determining reasonable remuneration. However, existing rates will be less appropriate if they reflect the exercise of market power. Market power may affect rates even if those rates are determined by arms-length negotiation between the collecting society and licensee(s).

One approach to considering the appropriateness of existing rates as a benchmark would be to consider the balance of bargaining power between the parties who negotiated the existing rates. If this is roughly equal, then the existing market rates determined by negotiation between the parties are less likely to reflect the exercise of market power by either party and may therefore be an appropriate benchmark. Generally speaking, if there is evidence that bargaining power is unequal, then existing market rates are more likely to reflect the exercise of market power by one of the negotiating parties and may not be an appropriate benchmark for assessing the reasonableness of proposed rates.

[emphasis added]

(5)    Having expressed that view, the guidelines identify the following factors at cl 5.2 that might be relevant when considering the use of existing rates as a benchmark for determining “reasonable remuneration”:

•    Changes in demand and/or supply conditions since the previous rates were determined. If demand and/or supply conditions have changed, then information about those changes should be combined with knowledge of the existing rate to make a new estimate.

•    On the supply side, consideration should be given to changes in the composition of the collecting society’s repertoire, the composition and magnitude of the collecting society’s administrative costs and the collecting society’s arrangements with copyright owners.

•    On the demand side, consideration should be given to factors which may have altered the value of the copyright material to users, including changes in the availability of substitutes for the copyright material licensed by the collecting society, the ability of licensees to obtain licences direct[ly] from copyright owners, and the way that copyright material is used by licensees.

(6)    Many of these considerations also are taken into account in constructing the hypothetical bargain as they go to the question of the basis upon which a willing but not anxious licensor would strike a price with a willing but not anxious licensee. As to the hypothetical bargain, the guidelines contain this observation at cl 6.1:

The hypothetical bargain approach refers to a hypothetical bargain between a willing, but not anxious, licensor and a willing, but not anxious, licensee. This description is symmetrical and implies that neither party has particular power over the other. In this sense, it reduces the effect of any market power held by the collecting society. It does so by assuming symmetry in power between the parties.

The use of a construction approach such as this allows economic principles to be applied in determining a licence fee.

Using this method, determining the licence fee consists of two steps:

1.    Calculate the economic surplus that is likely to be generated from the licensing transaction being negotiated compared with the outcomes that would arise should the licensing transaction not take place.

2.    Division of this surplus between the negotiating parties by setting a price.

[emphasis added]

(7)    As to calculating the surplus, the guidelines observe that the surplus, or value of a licence transaction, is the difference between the buyer’s willingness to pay (“WTP”) and the seller’s willingness to sell (“WTS”). The buyer’s WTP is the price at which a potential purchaser of a product is “indifferent between purchasing the product and not purchasing [it]”: cl 6.2. That point of indifference depends upon, among other things, the value of the next best alternative for the buyer. The seller’s WTS is a price reached at the seller’s point of indifference between selling and not selling. It represents the minimum price the seller would accept and still choose to supply: cl 6.2. The WTP price point might be revealed by existing market prices but where copyright material is the subject matter (conferring exclusive rights), “stated preference methods” usually provide “useful estimates of the distribution of the WTP” and those methods involve the use of survey evidence: cl 6.3. That course, rightly, has not been pursued in these proceedings having regard to first, the cost of conducting a survey and, second, the evidential difficulties of survey evidence. The evidence of experienced industry participants is to be preferred with, where appropriate, properly focused expert evidence.

(8)    The essential premise of the guidelines in the context of the discussion of constructing a hypothetical bargain is that the construct assumes an outcome (price) reflecting “symmetry”, not asymmetry in the bargain power of each side of the hypothetical bargain. The assumption of symmetry in bargaining power reflects an analogue of a workably competitive market where, in principle, each side is constrained by substitution possibilities. Assuming for the sake of the discussion that the supply of the rights is hypothetically contestable in the orthodox sense that if the licensor of the rights sought to “charge more and give less” the licensee could go elsewhere for those rights in that subject matter (which it cannot), the applicants say, in that circumstance, that the WTP is measured by an assessment of the “contribution” the licensed repertoire makes to the MMO’s products and services considered in the light of “all other inputs” going into those products and services. In other words, the price point is derived from taking into account all elements of the so-called value proposition. The guidelines observe that a licensee’s WTP for copyright material will typically reflect the contribution of the licensed material to the licensee’s profit. That being so, the derivation of the contribution to profit involves the contribution of all other inputs, a notion much along the lines of taking into account the totality of the inputs giving rise to the profit.

(9)    As to the WTS, the guidelines put the concept this way at cl 6.4:

Estimating WTS

The WTS concept relies on determining what the alternative to participating in the exchange is for the producer. For many products, the next best alternative for a seller is the price that can be received from selling to an alternative buyer. However, for copyright materials, selling to another buyer does not stop the copyright owner from selling to an extra buyer (due to the non-rivalrous nature of most copyright materials). This means that a common reasonable assumption in copyright is that the WTS is equal to the marginal cost of supply

(10)    The guidelines suggest that in general, for the supply of existing copyright material, the marginal cost is quite low and may be close to zero. For present purposes, we simply note these observations in the guidelines in the context in which they are expressed, on the topic of WTS (which Dr Eisenach describes as the “willingness to accept” (“WTA”)). CA notes that there are other factors affecting the price point of the WTS including alternatives in the form of subscriptions lost to the MMOs. There is also a debate about whether the relevant measure in all the circumstances is the average cost of supply. As to subscriptions, the applicants contest the proposition that customers who signed with them for their services (within the various mix of services) would seek subscriptions to a particular publisher’s content as a “substitute” to a media monitoring service.

Topic (5): The activities of and services offered by Isentia and Meltwater

The activities and services of Isentia

308    We set out a description of the services of Isentia at the end of this section. Apart from that description, we make these observations.

Ingestion

309    The first thing to note about the activities of Isentia, is concerned with the concept of “ingestion”. This concept and its content was the subject of some confusion in the oral submissions. The position is, however, this. In order to provide media monitoring and media intelligence services to customers by its various suite of services, Isentia monitors particular publications. (redacted) There is an issue about the scope of the publications so monitored. We address that matter under the heading “Licensed Work” later in these decision reasons. (redacted)

310    (redacted)

311    (redacted) However, Isentia says that not all tagged content is “communicated”. CA contends that all tagged content is communicated because it is “made available” on the Mediaportal platform for customers and that is enough. As to the activity of, for example, “media analytics” in the product called “Analytics”, the only media content examined for that service is the tagged retained content, not the “universe of content” licensed by CA. There is no dispute by the applicants that the act of ingestion of the content of monitored publications is an initial act of copying or reproduction.

312    CA accepts that there may be “at the margins” some publications of CA’s list of members that are not monitored by Isentia but CA contends for the general position that there is monitoring of the whole of the content of the publisher’s works represented by CA (relevantly for the purposes of these proceedings), and that there is a “wholesale copying” of that material onto the Mediaportal platform. However, see the discussion concerning the scope of licensed works later in these decision reasons.

313    CA emphasises its contention that as to the service of Analytics, the applicants are attempting to set up a “false dichotomy” between a service of media research and analysis by reference to retained tagged content on the one hand, and acts comprised in the copyright on the other hand, on the footing that the research and analysis service does not involve an “act” falling within the copyright. Whether there is such a false dichotomy depends upon whether there are acts undertaken in providing the research and analysis function which constitute a further act of reproduction (or communication of a work or a substantial part of a work) apart from initial ingestion.

314    The applicants accept that tagged content as described earlier may be communicated to the customer but emphasise that not all tagged content is communicated. The applicants accept that the act of ingestion requires a licence because it involves an act of reproduction of the content but they contend that the act of ingestion is, in truth, in effect, a technological step along the way to a possible communication of an article or part of an article to a customer. The act of “communication” is said to be the very point of monitoring content. The applicants contend that the fee for the act of communication ought to include, as part of a reasonable licence fee for communication, the anterior enabling technical step of ingestion. They say that having paid a reasonable licence fee for the right to communicate content to the customer (which ought to take into account the act of ingestion), any further utilisation of the ingested content tagged and saved as described ought not to attract another fee (in effect, they say, a second licence fee) for an act of undertaking research and analysis of the content.

315    They say that such an approach is merely double-dipping, especially where conducting research and analysis of content is not itself an act comprised in the copyright.

316    CA contends (apart from its analysis of technical steps mentioned shortly which it says all have to be compensated appropriately) that but for ingestion, none of the services could be provided in any event.

317    Before turning to Isentia’s particular services, the following further general matters should be noted.

The critical distinction: full print article v online service

318    Under the licence proposed by Isentia (and Meltwater), the presently operative distinction is sought to be maintained between the provision of access to a full print article whether provided as an electronic (digital) PDF copy or otherwise provided, on the one hand, and the provision of an online service on the other hand, on the footing that the two activities of the licensee are “very different” and indeed, the applicants say that preserving the distinction is not only material but “vital” to a true understanding of what is actually done by licensees when it comes to determining a reasonable licence fee in all the circumstances.

319    In the case of a “full print article” (which confusingly is described as a “press clip service”) provided by Isentia to the customer, the customer receives the full article (that is, an entire reproduction of the whole article generally by PDF supply). As to that reproduction of the entire full print article in PDF form (or otherwise), Isentia pays what it describes as the “full rate” which is currently (redacted) per full print article supplied.

320    (redacted) Where the article sits behind a paywall on that site, in order for the customer to see the full article after having been taken to the publisher’s website through the link, the customer is in the hands of the publisher. Access will usually be made available if the customer is a subscriber of the publisher. Alternatively, the publisher may simply elect to grant access for a number of reasons including that there are good commercial reasons for the traffic to be taken to the website.

321    (redacted)

322    Whereas in the case of a full print article, Isentia communicates or is entitled to communicate the full article (and pays the full per article rate), the extent of online content communicated to the customer is limited to the elements of the extract (portion as defined) with the full text available from the publisher’s website.

323    The applicants contend that the distinction between these two acts is qualitatively significant in fact and the distinction ought to be recognised in the licence fee. It would be otherwise, they say, if Isentia simply communicated the online content as full text to the customer as if it were exercising a right to provide a full text article.

324    Isentia says that the portion communicated to the customer with the link is very similar to the portions freely available on the website of some publishers and notably in respect of The Australian and The Sydney Morning Herald and as provided in a Google search.

Tagging

325    (redacted)

326    (redacted)

Rights of use: full press article compared with an online service

327    One other aspect of the distinction between a full press article and a portion of an online article is important in the context of the services of Isentia. When Isentia communicates a full press article, it pays the full prevailing press clip fee which it calls the “full rate”. Isentia says that this fee is a “large fee” for a right to full text use. Moreover, as to that use (for which it says it “pays handsomely”), it enjoys the right to choose how much of the text it displays in a report to the customer (perhaps a little, a lot or the whole) and if it chooses to display half the article as part of that service, it is entitled to do so because it is not restricted to the limitations applicable to a portion for online content.

328    Accordingly, there is said to be a danger in comparing an extract of online content confined by the portion limits with an extract of, or from, a full press article which is not so confined. The danger is that use of a full press article in a way permitted, cannot be called in aid as an illustration of an extract which is said to be a portion of an online article beyond the definitional limits of a portion or said to be an illustration of a substantial part of an online article so as to import the proposition that portions at large represent a “substantial part” of the licensed works. In the context of the Daily Briefings, CA relies upon examples at para 100(a) to (h) of the written submissions as emblematic examples of portions that contain sufficient text to constitute a substantial part of the corresponding article. Those examples, however, confuse material properly used within the full press article authority with extracts subject to the limitations applicable to portions of online items where the extract used is subject to the definitional limits earlier described.

Portions

329    As to portions, these two matters can be conveniently noted here.

330    First, News Corp Australia has a cohort of experienced decision-makers of longstanding industry experience. On 5 June 2020, News Corp lodged a submission with the ACCC on the topic of the proposed Mandatory News Media Bargaining Code in which it said this on the topic of Use of Copyright Material at para 6.1:

Australian copyright laws provide important safeguards for publishers and is crucial for protecting news media businesses’ brand and investments in the creation of original content. In essence, such laws prevent unauthorised use, reproduction or dissemination of material by someone who is not authorised to do so. While Australian copyright laws do not usually extend to news headlines and short snippets, they typically protect news articles and the expression of that original content in different formats. These laws recognise the significant human capital that is expended upfront in the creation of original content, and therefore provide important economic incentives for creators to keep producing original content.

[emphasis added]

331    Of course, in this passage, News Corp is stating its corporate view as a general proposition overall about news headlines and short snippets and not stating a position concerning particular examples. However, looking at the matter overall, the informed understanding of a participant such as News Corp is obviously not irrelevant. News Corp’s decision-makers have no doubt thought long and hard about this topic and the quote reflects their view.

332    In the Digital Platforms Inquiry Final Report, the ACCC said this at para 5.4.2 on the topic of News headlines and snippets:

Digital platforms often reproduce headlines and snippets of content from original news media articles created by journalists and media businesses (see discussion on digital platforms’ use of snippets in section 5.3.1). Generally, digital platforms’ use of article headlines is unlikely to infringe copyright protections in Australia. This is because many headlines are concise statements of facts and therefore headlines alone are unlikely to be copyright protected.

Digital platforms reproducing a snippet of a copyright-protected news article does not infringe copyright protections if the snippet does not reproduce a substantial part of the article. Only courts may determine whether a snippet reproduces enough of a copyrighted work to constitute copyright infringement, which means rightsholders must engage in litigation to seek a court’s decision whether infringement has occurred in each instance where a snippet is reproduced.

Aspects of CA’s contentions in relation to the services provided by Isentia

333    As to Isentia’s services, CA emphasises these matters.

334    As to the ingestion of content, (redacted). As to this step, Mr Gerstmyer did not know whether there is one copy of ingested and tagged content on the platform viewed by all customers or whether further copies are made for each customer folder. Mr Gerstmyer accepted that Isentia makes all content on its platform available in two ways (using the notion of “making available” in a non-definitional sense for present purposes; see the discussion at [806] and following): either by “actually push[ing] out” a portion or by enabling the content to reside on the platform with the “potentiality” of being seen by the customer. CA says that Mr Gerstmyer likened it to “Spotify” where all of the songs in the library of songs are available on demand.

335    Mr Gerstmyer’s evidence, however, is quite precise and the reference to Spotify was to make a point of distinction not similarity.

336    The evidence he actually gave, in context, was this. His understanding was that content was ingested, matched to the “client’s brief” or “keywords”, and “tagged” as relevant. The client might say, as to the scope of the content, “I only want national newspapers” or “only metros” or “certain types of content”. Mr Gerstmyer said that he was not aware of any client who said that they wanted “what’s relevant in the universe [of content]” or “always tell me what’s relevant in the universe”. As to broadcasts, for example, a client might say “I want news bulletins but I don’t want any sports reports”. He said that there is no-one who wants everything. Mr Gerstmyer accepted the hypothesis that if there was someone who wanted “everything” (that is, wanted tagged to that client every ingested article), he “suppose[d] theoretically” such a service would be possible although he thought that the cost of the content and communication would be “prohibitive”. He observed that such an hypothesis was “a different use case to a media monitoring organisation” and that the example put to him “spill[ed] more over to the library research type services” where a database going back over 25 years is interrogated. When pressed about whether, as a “technical matter”, all scraped (ingested) content on the platform could respond to a customer’s brief (that is, could all content be tagged if the customer wanted to receive all the content referring to it or responding to its keywords), Mr Gerstmyer accepted that such a service was “theoretically possible” but such a service would be like saying it is possible to download or listen to every single song in the Spotify catalogue “but in practice, and all reality, I can’t see that ever being a likelihood”: T, p 102, lns 16-46; T, p 103, lns 1-2.

337    Mr Gerstmyer was asked about the technical processes engaged in converting raw ingested text which responds to a client’s keywords into what he “supposed” could be described as a “cleaner, more useful format and enriched in that sense”. Mr Gertsmyer said that he was only familiar with these processes at a high level and not a technical level, but understood that the ingested headline was allocated to one field and the text to another. In the case of the Daily Briefing service, a portion of an ingested and tagged article (responsive to the keywords) is sent to the customer in an “email-layout” format. As to the technical intermediate steps, Mr Gerstmyer was asked whether the process of enrichment included a process of making a subsequent copy of the text from the scraped “raw” or “messy” (T, p 119, ln 21) content and using it for various purposes. Mr Gerstmyer did not know whether any subsequent or temporary copies were made in “machine processing”. He gave this evidence at T, p 120, lns 3-8:

My understanding is there is a copy in the database and that’s the copy used for our enrichment – for our information relevance. Where there would be other copying … the daily briefing for example – if someone gets an email with a portion in it, there is obviously a copy of that portion in the email as well as in the database.

338    Mr Gerstmyer said that he did not know the processes surrounding enrichment or temporary copies being made but supposed that there could be “temporary copies just in machine processing … held in memory or something like that”: T, p 120, lns 33-34.

339    CA’s end proposition is that for every article ingested there is a reproduction by ingestion (by scraping). Once ingested content is tagged as responsive to the client’s search terms, portions are extracted and made available online on the Mediaportal platform. Portions are also communicated in email format as Daily Briefings and otherwise communicated to the client. Sometimes, the whole article is also communicated. CA says that these steps do not take account of any temporary copies that might be made.

340    As to the service described as Insight Reports, CA contends that the evidence given by Mr Horell and Mr Gerstmyer that the reports (which summarise the sentiment of media coverage referable to the focus sought by the customer) do not reproduce a substantial part of a licensed work (or communicate a substantial part), is not correct having regard to two nominated examples. The first example is a report entitled (redacted). As to Insight Reports generally, Mr Horell gave evidence that (redacted). Mr Horell gave evidence that it is not correct to say that the key findings page highlights portions of content: T, p 197, lns 21 and following.

341    (redacted)

342    Mr Horell explained that the (redacted) is an example of a report where Isentia has sought to “showcase” or “sample” Isentia’s “media insights capabilities” and that other reports such as the reports for (redacted) “are probably more reflective of what a client receives”. The proposition was put to Mr Horell that when the analyst is presenting the key findings in a report, it may be appropriate depending on the customer and the analyst, to include quotes of publisher’s content in the report. Mr Horell said this at T, p 198, lns 11-14:

I think that would be extremely rare, and used in the case of instances like this [the (redacted)] where it seeks to showcase [comments]. I think what is more representative of what normally sits within a key findings [page] is what you see on p 11 [an assessment of the impact of the statement, sources of support, likely future trends].

[emphasis added]

343    Mr Horell said that he was not aware of any direction given to the Insights team not to quote “slabs” of text from articles but said this at T, p 198, lns 21-24:

I do read a lot of Insights Reports and I – yeah, would restate that the [vast] majority sit more like the (redacted) ones that you see after tab 3, as opposed to (redacted)], which is an anomaly of sorts.

[emphasis added]

344    The Isentia (redacted) Media Analysis Report of April 2020 is Tab 3 of the Horell cross-examination bundle. (redacted) Mr Horell accepted that the extent to which an analyst might refer to an article whether by headline or whether the analyst might “go a bit further and add text from an article” is a matter as between the customer and the analyst and every customer’s report content is different “but there certainly editorial guidelines that are administered by our insights directors”. Mr Horell accepted that if the customer saw a reference to an article in a report, the customer would be able to go to the Mediaportal platform, use the dropdown menu and access the article.

345    As to Insights Reports, we accept the evidence of Mr Horell that the (redacted) Report is characteristic of reports produced for clients by Isentia and that reports of that kind in that form do not reproduce either the whole or a substantial part of articles of the publishers. We accept that media content is analysed in the way described earlier and “used” to produce the highly interpretive and analytical findings set out in the reports but none of that activity of analysis and organisation of a report setting out the findings, trends, analysis by text, graphs, bar charts and pie charts and other analysis are acts within the copyright of the publisher of the articles. Nor is there a communication of a substantial part of an article to the customer in the formulation of reports of the kind illustrated by the (redacted) Report.

A description of each of the services

346    Turning now to the specifics of the services offered by Isentia, these matters should be noted.

347    As mentioned earlier, a key part of the services Isentia provides, occurs through Mediaportal, which is Isentia’s proprietary software that provides customers with a personalised cloud-based workspace. It can be accessed by a customer online or via a related Mediaportal app. The platform aggregates relevant media content and data for customers, and provides tools for customers to understand and review content quickly and efficiently.

348    When a customer engages Isentia to provide its media intelligence services via Mediaportal, the customer is assigned an Isentia Account Manager and account management team to provide personal service. The Isentia team, amongst other things, develops a “client brief”. This involves determining a set of search terms and specific criteria to track the media content of relevance to the customer. These search terms and criteria then inform the process of “tagging” described above. The use of “briefs” limits the provision of duplicate and irrelevant material to the client. The purpose of this service is for Isentia to review a large amount of content and then make available only that which is relevant in a folder for a customer to access. As mentioned above, any online information “scraped” from the internet that does not meet one or more client briefs is discarded.

349    To access tagged documents for a customer’s brief, the customer can view and search the relevant collection of documents in folders on the Coverage page of the Mediaportal platform. This includes the online media that is tagged in accordance with the process above, as well print media, broadcast media and social items. Within the folders, the headline of the content is displayed with the option to expand the item to show the first few sentences of the article, a brief summary of the material in the case of broadcast media, or the social media post. Metadata generated by Isentia during the ingestion process is also displayed when a document is expanded.

350    For press clip content, customers are given the option to read the entire article by selecting “read more” or “view original” (the PDF version). Customers can then share the content shown in the folders with other individuals within the customer’s organisation.

351    However, for online media, customers are only given the option to “view original” which redirects the customer to view the full article on the publisher’s website. As discussed above, whether a customer can bypass a paywall depends on the customer’s specific arrangements.

352    Mediaportal also integrates a number of additional tools and functionalities for customers to use which are referred to as “modules”.

353    The first of these modules is Connect which is a journalist and media contact database and media release distribution service. This includes a regularly updated database of journalists and key media contacts in Australia and internationally. The module also enables customers to create and distribute media releases. These media releases are then integrated with other Mediaportal modules to enable the customer to track the media coverage associated with that media release using separate tracking folders. The purpose of this service is for Isentia to provide a media-focussed address book for its customers that can be directly searched, as well as enabling the creation, distribution and monitoring of media releases generated by the customer.

354    The next of these modules is Insights which involves typically quarterly or half-yearly reports prepared by Isentia employees for the customer that capture and summarise the sentiment of media coverage. The form of Insights reports has already been discussed above, but it is sufficient to note here that the purpose of these reports is for Isentia to inform customers of information about the media coverage they have received, rather than specific media items.

355    In a similar vein but focusing on social media content, the next module is Reputation analysis. This service provides customers with a score to reflect public sentiment about the customer, informed by three key drivers: strategy, culture and delivery. This score is based on a combination of analysis of social media posts (what people say) and survey data (what people think). It also involves a corresponding report that summarises this information against the three key drivers. It does not involve any online or print media content, but it does include social media items, as shown in the example exhibited to Mr Gerstmyer’s affidavit of 3 July 2020 at pp 1 20 of TGG-5. The purpose of this service is to inform the customer of what members of the public think about their organisation, rather than the content of online or print news.

356    The next module is the Daily Briefing report. This report is designed to provide customers with a snapshot of the most relevant daily coverage and topics, based on content selected by Isentia’s Daily Briefing Editors and according to a customer’s specific requirements. The reports come in the form of an email that includes headlines and short summaries or extracts from the content and a “read more” link. For online content and social media content the “read more” link takes the client to the publisher’s website (or relevant social media platform) where a customer can then view the full text. For print media and broadcast media, the “read more” link takes the client to Mediaportal where the client can view the PDF copy or play the relevant video. The Daily Briefing report may also include an executive summary prepared by an Isentia employee summarising significant news stories relevant to the customer. This service requires Isentia to review the content it monitors and communicate those that are relevant in the form of summaries it prepares and links to read (or watch) more to the customer. This is demonstrated in the example exhibited to Mr Gerstmyer’s affidavit of 3 July 2020 at pp 21-25 of TGG-5.

357    The next of these modules is the Analytics function. This service is designed to allow the customer to prepare its own charts and graphs using information about the media content on Mediaportal. It does not involve using or reading the actual monitored content, but rather the data extracted by Isentia about the content (such as data about ‘volume’, audience reach). The charts and graphs, once created, are displayed on a customer’s Dashboard. Where displayed on the Dashboard, customers can click through to the relevant media items. The charts and graphs can also be exported or downloaded as a report in the form of an email, PDF or hyperlink. The exported reports do not include access or click through links to the media items. An example of an Analytics report created through this service is exhibited to Mr Horell’s affidavit of 15 June 2020 at pp 76-88 of RMH-2. The report includes data such as number of mentions or commonly used words over a set period of time. This service is strictly focussed on communicating data about monitored content and does not contain any headlines, extracts, summaries or the like from the actual content.

358    (redacted)

359    A related module is the Report Builder function. This service allows customers to select certain content to create a unique report specific to their business’ needs. Customers can choose how much of the online and print content they display and could therefore conceivably be limited to just a headline, or include a more extensive extract. The reports can take the form of a number of formats and be shared via email or link within their organisation. The purpose of this service is to allow customers to extract information stored on the Isentia database into a report that can then be shared internally within its organisation.

360    Another of the modules is the Alerts service. This enables customers to customise when and how they are notified of relevant media coverage. Customers are given the option to customise how the media items are included in the alerts – this can be headlines, summaries or full text. It also allows, depending on the subscription a customer has, an option to for the alerts to enable customers to view videos of broadcast coverage and order transcripts. The purpose of this service is to inform customers of new monitored content that is relevant to that customer and, depending on how the customer has configured the service, to provide them with a way to read the content in full.

361    There are two additional types of alerts Isentia provides, or will soon provide, to its customers. The first of these is Smart Alerts, which is a feature of the Mediaportal mobile app. Smart Alerts provide immediate notifications regarding important news relevant to the customer’s organisation. It does not involve communicating the actual articles, but does direct customers to view the relevant articles on Mediaportal. The second of these is Peak Alerts, which is still in its development phase. It is intended that the Peak Alerts will notify customers when a particular story or issue is trending across all media types. The purpose is to allow customers to track when a crisis broke, when the organisation responded, and to then assess whether coverage is increasing or settling down as a result of the organisation’s response. The notifications will appear on the Mediaportal platform and, like with the existing Alerts service, customers can choose to receive email notifications in relation to their Peak Alerts. Unlike the existing Alerts service, however, the emails will direct customers to view the content on the Mediaportal platform, rather than providing the option to review it through the email.

362    Separate to the Mediaportal platform, Isentia also provides Slice, which is its own web-based platform that is offered as a low-cost, basic media monitoring service. The service is a “do-it-yourself” style product. It involves the customer setting up the relevant folders and entering search terms to follow. The platform then uses the information entered by the customer (rather than an Isentia team as with Mediaportal) to gather relevant content (press, radio, television, online media). It is limited to Australian and New Zealand media sources. Customers receive an email when new content is available in their folders, but need to log on to the platform to actually view the new content. Customers can also place orders via this platform for transcripts, audio or video files for any broadcast item they have received for an additional cost. This service involves Isentia routinely reviewing the web for relevant results to customer-generated search terms and compiling them for the customer in a folder they can view on the website.

363    The third platform that Isentia provides is Unified Media Alert, which is its most basic legacy platform. The service involves Isentia sending one daily email alert containing portions of the news content for the day. This is a legacy service and not offered to new clients as Mediaportal can provide the same service.

A description of the activities and services of Meltwater

364    Meltwater operates in a similar fashion to Isentia. Like Isentia, Meltwater delivers most of its services via a secure online software platform referred to as either Fairhair or the MI Platform. Like Mediaportal, it can be accessed by a customer with a subscription from a web browser or mobile app. Customers are able to go onto the Fairhair platform and view and search the relevant results displayed – including a headline, a short passage of text from the article, and an option to read more. For online content, the ‘read more’ option takes the client to the website, whereas press content takes the client to an image of the article and makes available a PDF version for download.

365    In terms of what articles are presented to customers, this is similar to Isentia in that only the relevant articles are actually displayed to a customer when they access the Fairhair platform. (redacted)

366    (redacted)

367    (redacted)

368    (redacted)

369    Meltwater also offers various reporting services.

370    (redacted)

371    (redacted)

372    (redacted)

373    (redacted)

374    (redacted)

375    (redacted)

376    (redacted)

377    (redacted)

378    (redacted)

379    (redacted)

380    (redacted)

381    (redacted)

382    (redacted)

383    (redacted)

384    (redacted)

Licensed Work

385    As mentioned earlier, there is a dispute about the scope of works licensed under the CA proposals. The applicants claim that they require a licence to provide their service but they do not require a licence over the entire CA repertoire. They criticise the reasonableness of each licence proposed by CA on the footing that CA is seeking to force them to take a licence of a total field of works as part of the justification for a significant licence fee and they say that many of the publications are not monitored and simply not relevant to the services they seek to provide. The definition of Licensed Work is mentioned earlier but for convenience we set out the definition as it appears in the Alternative Licence and the Amended Alternative Licence:

Licensed Work means each Article published in an Edition or Online Publication and any other work that Copyright Agency has been authorised to license to MMOX under this agreement, including Domestic Licensed Works and Foreign Licensed Works, and includes Articles in the publications of the publishers listed at (<https://www.copyright.com.au/mmo-publisher-list/>) as updated from time to time and Major Mastheads listed in Schedule 3, but excluding an Excluded Work.

386    There are three separate lists that appear at the webpage to which the reader is taken by the hyperlink in the above definition. One list is described as Domestic Licensed Works and the other two are broadly described as Foreign Licensed Works with the further sub-categories of NLA Licensed Works and CFC Licensed Works. To download the lists the person interrogating the website must enter a name, a company and an email address.

387    In addressing the definition of Licensed Works in his fifth affidavit, Mr Suckling refers to what he describes as a “non-exhaustive list” of publications CA is authorised to license that includes over 17,000 publications (Mr Suckling’s fifth affidavit of 25 January 2021 at para 224(b)). Mr Suckling indicates at para 225 of that affidavit that “the same repertoire of works are licensed under the terms of the CA Licence and the Alternative CA Licence”.

388    Isentia tendered a number of documents evidencing a list of publishers that are to be covered by each of CA’s proposed licence arrangements that provide little relevance to its business activities. These documents are at Tabs 40 and 41 of Isentia’s Tender Bundle. These tabs are extracts from tables prepared by Dr Eisenach in his first expert report (29 September 2020). The extracts concern some, but by no means all, publishers (publications) to which the proposed CA Licence, Alternative CA Licence and Amended Alternative CA Licence would apply. These tables (and exchanges about them at T, p 772, ln 12 – T, p 781, ln 5) do not, however, provide a sense of the scope of all the publications that would be covered by the definition of Licensed Work. The field of publications comprehended by the definition is very extensive indeed.

389    Isentia makes the not unreasonable point that CA has never provided it with a comprehensive list of publications which CA says it is authorised to licence. Isentia says that it is apparent that CA seeks to subject the proposed licences to a charge related to a licence for an entire “repertoire” which includes works that the applicants do not wish to licence. In that sense, the “reasonableness” of the proposed charge is said to be supported by the suite of publications covered by the licence. By a letter dated 22 December 2020, Baker McKenzie, the solicitors for Meltwater, sought to clarify CA’s position in relation to identifying the field of publications brought within the proposed Pro Forma (redacted). In that letter, Baker McKenzie said this:

(redacted)

390    (redacted) We are not in a position to make judgments about each of the illustrations in the schedule. However, it is not unreasonable on the part of the applicants to call upon CA to identify the publications falling within the licence which in turn are said to support the basis for the fee for rights exercised in relation to those works. Moreover, it is not unreasonable for the person claiming to require a licence for the purposes of s 157(3) to say to CA that it requires a licence for particular works but not necessarily the entire block blanket field of publications for which CA enjoys a field of rights.

Topic (6): The question of whether the Streem FSMM Licence in the context of the arrangements between CA and Streem made on 9 October 2020 is a comparable market bargain or market rate

391    The logical starting point in seeking to understand how the CA/Streem FSMM Licence (redacted) came to pass and whether the Streem Licence reflects a contemporary “market bargain” between a willing but not anxious licensor and a willing but not anxious licensee, is the context which gave rise to the negotiations ultimately leading to the licence, the Deed of Settlement and the Streem proceeding and the entry into the Deed Polls.

392    That context begins with the position adopted by Streem in its proceeding against CA.

393    We have already noted that on 29 May 2018, CA had provided Streem with the proposed CA Licence. Streem’s position in that proceeding was that the proposed CA Licence contained terms as to charges and other non-price terms which were unreasonable.

394    Streem contended that since the commencement of its licence relationship with CA, Streem had enjoyed, as part of that licence relationship, a licence to copy and communicate and scrape the online content of The Australian and The Weekend Australian. On 5 February 2020, CA notified Streem that from 20 March 2020, CA would cease to have the right to sub-licence those two publications of News Corp. Streem’s position in its proceedings was that those two publications were important sources of media content to large corporations and government departments. (redacted) On 5 February 2020, CA notified Streem that CA had lost the right to sub-licence particular publications of Nine/Fairfax and the list of publications in that notification are the same publications withdrawn from Isentia and Meltwater by Nine Publishing as described at [111] of these decision reasons. The withdrawal of those publications took effect from 17 January 2021. Thus, Streem was facing the loss of rights from CA concerning each of those publications (respectively) absent some other arrangement having been entered into.

395    In the Streem proceeding, Streem contended that the fee payable under the proposed CA Licence was unreasonable. (redacted) The evidence of the contended unreasonableness of the price and non-price terms of the CA Licence proposal put to Streem on 29 May 2018 is set out in the affidavit of Mr Keith Forbes, Streem’s Commercial Director and Chief Operating Officer, affirmed on 26 June 2020. The criticisms contained in that affidavit were put to Mr Elgar Welch, Streem’s Chief Executive Officer, in cross-examination.

396    (redacted)

397    (redacted)

398    (redacted)

399    (redacted)

400    Mr Welch was cross-examined about aspects of the position Streem had taken leading up to its change of position on 9 October 2020. The applicants observe that an affidavit affirmed by Mr Welch on 28 January 2021 had been filed in the proceedings by CA. The affidavit is very brief and provides little explanation of the circumstances leading to the sequence of arrangements made between CA and Streem on 9 October 2020. Mr Welch’s affidavit exhibits the Streem FSMM Licence, the Deed of Settlement and the two Deed Polls. He also exhibits copies of emails sent to Streem’s Board of Directors on 8 and 9 October 2020. He describes the 8 October 2020 email by which he briefed the Board on the proposed arrangements as a “comprehensive account of the factors relevant to that recommendation”. He says that on 9 October 2020, the Board approved Streem’s entry into the FSMM Licence and the other documents. However, Mr Welch says nothing further in his very brief affidavit about any aspect of the arrangements with CA which (redacted)

401    (redacted)

402    (redacted)

403    (redacted)

404    (redacted)

405    (redacted)

406    (redacted)

407    (redacted)

408    (redacted)

409    (redacted)

410    (redacted)

411    (redacted)

412    (redacted)

413    The applicants note that Mr Johnson’s email and the attached draft licence agreement were produced after Mr Suckling (CA) and Mr Welch had been cross-examined. Thus, they were not able to take those witnesses to the documents. This point is made many times in relation to other documents on this topic produced after those witnesses had given evidence or after the completion of the hearing.

414    (redacted)

415    (redacted)

416    (redacted)

417    (redacted)

418    (redacted)

419    (redacted)

420    (redacted)

421    (redacted)

422    (redacted)

423    (redacted)

424    (redacted)

425    (redacted)

426    (redacted)

427    Again, the applicants complain that this email was produced after Mr Suckling and Mr Welch had given their evidence and thus those witnesses were not able to be cross-examined about the content of the documents.

428    (redacted)

429    Mr Johnson forwarded this email to Mr Suckling. Again, the applicants make the point that these documents were produced after Mr Suckling and Mr Welch had been cross-examined.

430    (redacted)

431    (redacted) Mr Suckling accepted that identifying whether CA had revenue and usage data for the nominated companies made clear to Streem that they were already customers of either Isentia or Meltwater. Mr Suckling also accepted that providing Streem with this data was a breach of the confidentiality obligations in the interim licence and that the previous denials of the fact were not correct and unfortunately were a mistake.

432    (redacted)

433    (redacted)

434    (redacted)

435    (redacted)

436    (redacted)

437    (redacted)

438    (redacted)

439    (redacted)

440    (redacted)

441    (redacted)

442    (redacted)

443    (redacted)

444    (redacted)

445    (redacted)

446    (redacted)

447    (redacted)

448    Again, the applicants complain that these documents were produced after the hearing of the evidence and Mr Suckling and Mr Welch were not able to be asked about them.

449    (redacted)

450    (redacted)

451    (redacted)

452    (redacted)

453    Again, this email was produced after the conclusion of the hearing.

454    (redacted)

455    (redacted)

456    (redacted)

457    (redacted)

458    (redacted)

459    (redacted)

460    (redacted)

461    (redacted)

462    (redacted)

463    (redacted)

464    (redacted)

465    (redacted)

466    (redacted)

467    (redacted)

468    (redacted)

469    (redacted)

470    (redacted)

471    (redacted)

472    (redacted)

473    (redacted)

474    (redacted)

475    (redacted)

476    (redacted)

477    (redacted)

478    (redacted)

479    (redacted)

480    (redacted)

481    (redacted)

482    (redacted)

483    (redacted)

484    (redacted)

485    (redacted)

486    (redacted)

487    (redacted)

488    (redacted)

489    (redacted)

490    (redacted)

491    (redacted)

492    (redacted)

493    (redacted)

494    (redacted)

495    (redacted)

496    (redacted) Having regard to the exchanges described above in some detail, it is clear enough that CA, Nine Publishing and News Corp conducted themselves on the basis that the period of the licence between CA and Streem, and the relationship between the licence and the period of the Deed Polls were inextricably linked and were matters about which CA, Nine Publishing and News Corp all needed to be and were of a common mind.

497    (redacted)

498    (redacted)

499    (redacted)

500    (redacted)

501    (redacted)

502    (redacted)

503    (redacted)

504    (redacted)

505    (redacted)

506    (redacted)

507    (redacted)

508    The Tribunal is invited by CA to proceed on the basis that the case really being put by it is one of a licence offered to Isentia and Meltwater on the footing of the Alternative CA Licence. A further alternative is that each applicant is offered the Amended Alternative CA Licence if each applicant (either or both) settles its respective proceeding on the basis of the Pro Forma Settlement Deed leading to a Pro Forma Licence (redacted) as amended by the Deed, coupled with a negotiation with News Corp and Nine Publishing to resolve the final basis upon which Deed Polls could be offered to each applicant.

509    As to the Alternative CA Licence (redacted) offered as a stand-alone document taken out of the context of the Streem Settlement Deed and the Deed Polls, the licence provides for a revenue share percentage of 23%. It has no provision giving the applicants the comfort of a guarantee of access to the content of Nine Publishing or News Corp.

510    (redacted)

511    In other words, to select the Streem FSMM Licence out of the context within which it necessarily came to pass and hold it up before the Tribunal ((redacted)) as a stand-alone example of a market transaction (said, in effect, to be a determinative market transaction for the purposes of s 157(3) of the Act) between parties characterised as a willing but not anxious licensor of that instrument and a willing but not anxious licensee of that instrument, as if the licence is not to be entirely understood and comprehended in the context of all of the other factors centred upon all of the elements of the resolution of the Tribunal proceedings and the other critical elements of the arrangement (especially those in relation to the Deed Polls), is entirely unhelpful.

512    (redacted)

513    (redacted)

514    (redacted)

515    (redacted)

516    We will now identify some of the particular difficulties which suggest that the Streem FSMM Licence is not a comparable market bargain:

(1)    The Streem FSMM Licence is not an example of a discreet market rate or market bargain because it only exists at all due to the interdependencies with the Settlement Deed and the Deed Polls.

(2)    To the extent that the revenue share percentage of 23% in the Alternative CA Licence (redacted) might be (redacted) the availability of that rate assumes a settlement of each proceeding before the Tribunal (although a settlement of one is not dependent on settlement of the other) on the terms of the Pro Forma Settlement Deed coupled with the opportunity for each applicant to engage in further negotiations with Nine Publishing and News Corp so as to work out the final terms of Deed Polls that might, upon the resolution of any relevant matter, then be available. That settlement simply did not come to pass and absent settlement on those terms as put by CA, there is no (redacted) and there are no Deed Polls.

(3)    The applicants rightly contend that as to the availability of the (redacted) and the opportunity to seek to reach finality with Nine Publishing and News Corp on the precise terms of Deed Polls, the entire proposal turns on settlement of the relevant proceeding essentially on a “take it or leave it” basis as delivered by CA. The only truly unqualified proposal apart from the initial CA Licence proposal (which, in substance, is not pressed although not formally abandoned), is the Alternative CA Proposal. As to the Amended Alternative CA Proposal, it leaves extant an argument about adjustments during the period of the interim licence and is contingent upon settlement of the proceeding on the terms of the Pro Forma Deed.

(4)    The availability of the Amended Alternative CA Licence rises no higher than an offer to settle each proceeding which was not accepted. Even in its own terms, it does not propose either as part of the licence or outside the licence (by Deed Polls from the two primary publishers) a clear, precise, unqualified Deed Poll by each publisher. There always remained the matter of further discussions to resolve any issues relevant to entry into those Deed Polls specific to the publishers on the one hand and Isentia on the other, and the resolution of relevant issues as between the publishers and Meltwater. (redacted)

(5)    Moreover, as to the Deed Polls, CA (the s 157(3) licensor) does not offer the guarantees conferred by the Deed Polls. It expressly says that it cannot do so. The Deed Polls must be given by third parties standing outside the Tribunal proceedings. As to the Deed Polls, CA says (by its 11 November 2020 document) that it understands that if Isentia and Meltwater accept the terms of either proposal and similar assurances are sought from News Corp and Nine Publishing as obtained by Streem, then the publishers would grant the Deed Polls on “substantially the same terms”. CA does not offer the Alternative CA Licence coupled with a commitment procured by it from each head licensor in the form of News Corp and Nine Publishing. The guarantees secured by Deed Polls are not offered as an element of the orders the Tribunal might make so far as they relate to the Alternative CA Licence, let alone any licence that might depart from the terms of the proposed Alternative CA Licence.

(6)    The applicants correctly emphasise the role of the Deed Polls in the Streem negotiations and the ultimate arrangements with Streem. (redacted)

(7)    (redacted) In these proceedings, the applicants seek orders that “major mastheads” not be withdrawn from the licence as ordered by the Tribunal, during the term of the licence. The Tribunal has previously determined that it has the power to so order. CA contests that position. (redacted) It seems a little odd to the Tribunal that licence terms, both as to price and non-price elements but particularly as to fees and charges, could be set in consideration of a licence to exercise rights over “major mastheads” but immediately after the licence is struck, one or more major Mastheads might be withdrawn by its relevant publisher fundamentally altering the bilateral commercial balance in the value of the rights and the licence itself. (redacted)

(8)    (redacted)

(9)    As to the particular titles which have now been withdrawn by News Corp and Nine Publishing as described earlier, (redacted)

(10)    (redacted)

(11)    (redacted)

(12)    The applicants contend that another reason why the Streem FSMM Licence ((redacted)) is not a market bargain emblematic of application to Isentia and Meltwater is that the circumstances of each of the three entities is very different. The applicants say that the percentage of revenue fee has very different consequences for each entity because of the different product offerings of each entity. (redacted) Applying that perception of the principle to the evidence of the services offered by Isentia and Meltwater and weighing in the examination of each service the extent to which each service engages, in whole or in part, or not at all, with the licensed works, a fee that captures as a minimum a 23% rate applied to all revenue derived from all services on the footing of a “but for” approach fails, unreasonably, to recognise that the relevant rights engaging a licence fee are acts of reproduction and communication to the public. A fee for exercising either or both of those rights must necessarily take into account the scale and character of the use of the works and where a range of value-added services are offered (where there is little or no use of the works amounting to reproduction or communication), the fee must take that matter into account either by discounting the rate as an “evaluative judgement” (recognising that minds might differ but also recognising that nevertheless an “evaluative judgement” must be made), or by taking the matter into account in some other way.

(13)    A further point of distinction between Streem on the one hand and Meltwater and Isentia on the other hand is that Streem is a relatively early market entrant which, as Dr Eisenach observes, (redacted)

(14)    (redacted)

(15)    The Deed of Settlement, central to the arrangements, provided (redacted). This history, its resolution by the release and the emphasis given to the matter (redacted) is an important feature of the arrangements overall that contextually deprives the Streem FSMM Licence of the character of a comparable bargain.

(16)    (redacted) The applicants correctly say that as a comparable transaction, none of those factors that caused the FSMM Licence to come into existence apply to the Alternative CA Licence.

(17)    (redacted) Streem was in part induced to enter into the arrangements so as to secure the advantage of (redacted) a factor that does takes the arrangements and particularly the Alternative CA Licence beyond a rational characterisation as a “comparable market bargain”.

517    Having regard to all of these considerations, we are not satisfied that the Streem Alternative CA Licence or the Amended Alternative CA Licence represent instruments which can be described as a comparable market bargain for the purposes of these proceedings. The character and content and detail of the arrangements make it clear that the instruments are not a comparable market bargain. They are specific to their own unique set of circumstances in the context of the arrangements overall. It is true that the parties engaged in a sequence of negotiations which began in the first half of 2020 but a careful examination of the exchanges shows that things fundamentally changed on 24 September 2020 in the context of the intensity of CA’s preparation of its evidence for the Tribunal proceedings in which the claims of all three applicants were to be determined. The text of the emails makes the context perfectly clear. It is also true that Streem and CA were represented by lawyers. These factors have been emphasised by CA in seeking to support the proposition that the Alternative CA Licence and the Amended Alternative CA Licence represent comparative market bargains between a willing but not anxious licensor and a willing but not anxious licensee. We reject that contention. We have examined all of the evidence on this topic in considerable detail to satisfy ourselves that the characterisation contended for by CA is not correct.

518    As to the views of Dr Eisenach, it must be remembered that there is a fundamental matter of fact to be examined before an opinion can be formed about whether either the Alternative CA Licence or the Amended Alternative CA Licence can be regarded as a comparable market bargain. We have examined the foundation facts going to that matter. Dr Eisenach has not. Dr Eisenach was not intimately informed about these foundation facts in coming to his opinion. In the light of the foundation facts, we are not satisfied that we can attach weight to Dr Eisenach’s view about the two transactions.

519    The Alternative CA Licence departs so substantially from any notion of a comparable market bargain influenced and informed as it is by all of the factors we have described, that we cannot regard it as a comparable market bargain. To the extent that the Amended Alternative CA Licence is put forward as a licence proposal rather than a settlement offer, it too falls within the matrix of facts and circumstances described above taking it outside the notion of a comparable market bargain.

520    We now turn to the question of the benchmarking exercise conducted by Dr Eisenach.

Topic (7): The extent to which Dr Eisenach’s benchmarking exercise and robustness check is of assistance to the Tribunal in deciding the questions in issue

521    In order to understand whether the analysis undertaken by Dr Eisenach assists the Tribunal on the question of the reasonableness of the charges in the CA Proposed Licence, it is necessary to first understand exactly how Dr Eisenach approached the benchmarking exercise.

522    At Section B of Part VI of his report of 29 September 2020 (“ER 1”, [203] and following), Dr Eisenach explains that his benchmarking analysis is designed to assess the reasonableness of the licence fees in the CA Proposed Licence (and also the MMO Proposed Licence) based on “market-based comparable bargains”. The reference to the CA Proposed Licence is a reference to the CA Licence proposed to Isentia on 21 May 2018 and Meltwater on 18 May 2018. Dr Eisenach understood that the licence fee would be the greater of the sum of a fixed fee and a variable percentage of “Revenue” and a “Minimum Guarantee” based on the MMO’s “Revenue” for the 12 months prior to the “Commencement Date”. The licence fee would be based on the amount of each MMO’s revenues earned “directly or indirectly” from the use of “CA Licensed Content” excluding revenues earned from particular services to the extent that those services were acquired by customers as “stand-alone separately priced service[s]” which do not use CA Licensed Content. Dr Eisenach then sets out the definition of “Revenue” and identifies the scope of the “Excluded Services”. Dr Eisenach describes the “revenue amount” upon which MMO licence fees would be based (after taking into account revenues from excluded services) as the “CA Content Attributable Revenue” (which includes fees collected by the MMO in respect of “downstream licences”): ER 1, [131]-[135]. Dr Eisenach then sets out the CA Proposed Licence Fees by reference to Annexure A to the licence which is as follows:

523    Dr Eisenach also sets out his understanding of the scope of the rights granted under the CA Proposed Licence.

524    In the benchmarking analysis, Dr Eisenach seeks to estimate a “reasonable range” for the fair market value of the “Major Mastheads” to Isentia and Meltwater based on “direct licences” between (redacted) for The Australian. Dr Eisenach says that he relies on these licences to value the Major Mastheads because they are of “disproportionate value” to the MMOs and their customers as compared with the “remainder of the CA repertoire”, and The Australian is “among the Major Mastheads”: ER 1, [204]. Dr Eisenach then seeks to estimate a reasonable range for the value of CA’s remaining (non-Major Masthead) repertoire based on (redacted) represent a reasonable proxy for the value of CA’s non-Major Masthead repertoire”: ER 1, [205]. Combining the estimated valuation ranges for the CA Major Masthead repertoire and the non-Major Masthead repertoire is said to yield an estimated reasonable range for the fair market value of the “CA Licensed Content” for each MMO: ER 1, [205]. Dr Eisenach concludes that the licence fees embodied in the CA Proposed Licence fall within or below the reasonable range so benchmarked: ER 1, [205]. Dr Eisenach also contends that the licence fees contended for by the MMOs by their proposed licences fall “well below” the reasonable range: ER 1, [205]. For the moment, we propose to focus on Dr Eisenach’s benchmarking of the CA Proposed Licence.

525    In the third section of his benchmarking exercise, Dr Eisenach presents a “robustness check” which operates on the following footing. The Major Masthead analysis and the non-Major Masthead analysis each utilise “measures of readership” to “adjust” for the “relative values” of the “benchmark content” as compared with the CA Licensed Content, and like the Major Masthead analysis, the robustness check also relies on The Australian as a benchmark but relies upon “clip volume data” to “adjust” for “relative values”: ER 1, [206]. Dr Eisenach concludes that the robustness check affirms his benchmarking analysis.

526    Next, Dr Eisenach uses (redacted) to assess the value of the CA Licensed Content in “online monitoring” in order to assess the reasonableness of the online monitoring fee element of the MMO Proposed Licence. Finally, Dr Eisenach considers other licences made available to him.

527    As to the Major Masthead analysis, we note the following matters.

(1)    Dr Eisenach notes that on 20 March 2020, News withdrew The Australian and The Weekend Australian from the CA Licence arrangements and that (redacted). These are the licences Dr Eisenach uses as “comparable bargains” to estimate the value of the Major Mastheads in the CA repertoire (and thus their contended value to Isentia and Meltwater): ER 1, [209] and [210].

(2)    Dr Eisenach says that the (redacted) are comparable bargains because, first, they are “sufficiently comparable in scope and coverage” (which means sufficiently comparable for benchmarking the CA Proposed Licence of Major Mastheads notwithstanding that both titles were withdrawn from the CA Proposed Licence on 20 March 2020) and second, the licences in each case are “market-based bargains in the sense that both parties were willing and fully informed”: ER 1, [211]. Each CA Proposed Licence was proposed to Isentia and Meltwater in May 2018. The term was two years (with no particular commencement date). Had it been accepted on say 1 July 2018, it would have endured until 30 June 2020 and presumably the licence of The Australian and The Weekend Australian would have remained within the content grant. Dr Eisenach is no doubt conducting his analysis on that basis. As at the date of the hearing, of course, to the extent that CA continues to offer and support the CA Proposed Licence, it seems difficult to see how selecting titles as benchmarks which the owner has chosen to remove from the CA grant operate to benchmark those titles that do remain within the grant. However, we examine that matter later in these decision reasons. Dr Eisenach expands upon the two factors mentioned above (at ER 1, [211]), further at ER 1, [212] to [223].

(3)    At ER, [224], Dr Eisenach notes that, first, the (redacted)

(4)    (redacted)

(5)    (redacted)

(6)    Next, Dr Eisenach seeks to make adjustments for the “relative value” of The Australian compared with the 12 other Major Mastheads and does so by relying on “audience measures” compiled by Enhanced Media Metrics Australia (“EMMA”) in surveys undertaken by it (accepted by Mr Eisman (Nine Publishing) as the industry standard for audience data for the news media industry) and published in the Australian Total Audience Report (“ATAR”). In relying on audience measures as the basis for the adjustment, Dr Eisenach observes that it is reasonable to assume that the value of The Australian relative to other Major Mastheads in the CA repertoire is “proportional to audience size (or ‘readership’)”. Dr Eisenach regards that assumption as reasonable as the value of a publication to an MMO’s customers is “highly correlated” with readership. (redacted)

(7)    The data relied upon by Dr Eisenach to account for differences in audience size among the Major Mastheads is “monthly print audience”, “monthly digital audience” and “monthly total audience”: ER 1, [228]. Dr Eisenach calculates the ratio of each Major Masthead’s readership relative to the readership of The Australian using those three measures and then multiplies those ratios by the adjusted benchmark price for The Australian: ER 1, [228]. To take an example, The Australian has a print readership of 1,759,000 monthly readers, a digital readership of 3,044,000 monthly readers and a total monthly readership of 4,413,000 readers. The Advertiser (one of the other 13 Major Mastheads) has a print readership of 725,000 monthly readers or 41.2% of the print readership of The Australian, a digital readership of 1,242,000 or 40.8% of the digital readership of The Australian and a total audience of 1,632,000 monthly readers or 37.0% of the total readership of The Australian. (redacted)

(8)    A summary of the analysis set out in Dr Eisenach’s Tables E-4 (for Isentia) and E-5 (for Meltwater) taking account of the E-6 sources reveals a contended range of fair market values for the Major Mastheads for Isentia and Meltwater (relevantly for present purposes, but also the values for Streem). (redacted) We will return to some specific aspects of the schedules later in these decision reasons.

(9)    The benchmarking method adopted by Dr Eisenach for attributing value to the Major Mastheads rests upon use of the (redacted) for The Australian and The Weekend Australian (taken as one), on the footing that those licence agreements are comparable bargains (adjusted for the particular disconformities already described) and that each is market-based in the sense that both parties are willing (but not anxious) and informed: ER 1, [211]. Dr Eisenach says that he estimates the value of the Major Mastheads because as a class they reflect disproportionate value as compared with the balance of the CA repertoire and The Australian is among the Major Mastheads. Even if the two factors identified at ER 1, [211] are true as expanded at ER 1, [212]-[223] (about which there may be some doubt), those circumstances do not identify why the particular single title, The Australian, is the emblematic anchor point against which all other values are to be measured.

528    As to the non-Major Masthead repertoire, we note the following matters:

(1)    (redacted)

(2)    (redacted)

(3)    (redacted)

(4)    Again, Dr Eisenach adjusts for the circumstance that the CA Proposed Licence covers a “larger volume of content” than (redacted). The adjustment mechanism is “measures of readership” gathered (redacted) and the June 2020 ATAR Report concerning 48 publications including 13 Major Mastheads and 35 other publications including news mastheads, magazines and multiplatform publications: ER 1, [237].

(5)    (redacted)

(6)    Next, Dr Eisenach estimates the fair range of market values of the CA licensed content for Major Mastheads and non-Major Mastheads (as described earlier) in the aggregate as set out at Table 15. (redacted) Having isolated the range of contended fair market values of each publication for each MMO benchmarked in the way described, Dr Eisenach then compares the maximum estimated fee payable under the CA Proposed Licence by each MMO with the value range for each MMO. (redacted)

529    Having undertaken those exercises, Dr Eisenach then seeks to check the robustness of the outcome of his benchmarking method. To do that, Dr Eisenach adopted an “alternative measure of relative value” based on the “number of clips copied and communicated for each publication”: ER 1, [248]. Dr Eisenach contends that the results of the robustness check are consistent with the benchmarking and supports his benchmarking analysis. As to the robustness analysis, we note these matters:

(1)    To undertake the clip analysis, Dr Eisenach used the clip data concerning the number of clips copied and communicated by Isentia (and Streem) for each publication as an alternative to the ATAR audience data so as to estimate the relative values of each publication in the CA repertoire. Dr Eisenach relies solely on the (redacted) for The Australian as a benchmark. That is, there is no differentiation in value between Major Mastheads and non-Major Mastheads in this analysis because a clip is assumed to have value per se based on the fact that it meets the customer’s search criteria regardless of where it is published.

(2)    The results of the clip analysis in Tables E-13 and E-14 suggest that the fair market value of the content licensed under the CA Proposed Licence is that set out in Table 16 at ER 1, [250]. (redacted)

530    The further exercise undertaken by Dr Eisenach is to seek to determine a value for online monitoring services based on the value of an agreement (redacted)

(1)    (redacted)

(2)    Dr Eisenach considers that the (redacted) is nevertheless useful as a benchmark for the purposes of establishing the value of the online monitoring services in assessing the reasonableness of the online monitoring fees under the CA Proposed Licence to MMOs and for assessing how the value of news content has changed over time.

(3)    Another difference is that the CA Proposed Licence covers a much larger volume of content than the (redacted). Data on readership of (redacted) is not available through EMMA. Dr Eisenach notes that Mr Gerstmyer says that (redacted). Dr Eisenach notes “data” provided by Mr Gerstmyer (affidavit, 3 July 2020) that (redacted) Dr Eisenach says that he used this data to “adjust” the value of the (redacted) to establish the value of the CA Proposed Licence: ER 1, [255].

(4)    (redacted) and so Dr Eisenach seeks to estimate their fair market value for the (redacted) using the licence fee paid by (redacted) adjusted downward based on the fees paid under the (redacted)

(5)    The next step in this part of the analysis is to determine the “relative reach” of the (redacted) Dr Eisenach does that by dividing the number of (redacted)

(6)    The next step in estimating the value of the online monitoring component of the CA Proposed Licence is this. (redacted)

(7)    Dr Eisenach draws two conclusions from this analysis of the value to each MMO of online monitoring services of CA licensed content based on “relativities” extrapolated from fees paid to (redacted). The first is that the right to monitor and exercise rights in relation to CA online licensed content provides each of Isentia and Meltwater with “significant value” even in the absence of paywall access and second, the MMO proposed licences “dramatically undervalues” the use of CA licensed content in the provision of online monitoring services.

531    Isentia and Meltwater make the following criticisms of Dr Eisenach’s benchmarking analysis:

(1)    The applicants contend that the analysis is based upon unsound assumptions and that an examination of the results of the analysis demonstrates absurdities and anomalies that call the reliability of the entire analysis into question. The first difficulty is said to be that the value of the Major Mastheads is calculated by reference to (redacted) for The Australian and The Weekend Australian (once those titles were withdrawn from the licence arrangements with CA) and by reference to print and digital audience reach. This is said to incorrectly assume that “the value of a licence of each of the Major Mastheads to the MMOs is directly proportional to that masthead’s print and digital audience size” [emphasis added]. They say that Dr Eisenach fails to explain how he came to choose The Australian as the anchor point for that part of the benchmarking analysis.

(2)    The second difficulty is said to be that a value for the remainder of the CA licensed repertoire is calculated by reference to (redacted) which again is said to incorrectly assume that the value is directly proportional to the print and digital audience size for each non-Major Masthead.

(3)    The third difficulty is said to be that the robustness check using clip data also incorrectly assumes that The Australian is representative (that is, a relevant benchmark) of the value of each Major Masthead.

(4)    The fourth difficulty is said to be that Dr Eisenach’s analysis, in seeking to determine the value of online monitoring services using clip data relating to (redacted), incorrectly assumes that the (redacted) is representative of CA’s online repertoire as a whole such that the value to Isentia and Meltwater of a licence of each of CA’s online publications is to be determined by relativities extrapolated from the number of (redacted) and, judged by that metric, said to be equivalent in value to (redacted). Again, as to the benchmarking anchor points of the (redacted)for non-Major Mastheads, and the (redacted)for online publications, the applicants say that Dr Eisenach has failed to explain how the (redacted)came to be selected as the benchmarks notwithstanding that having selected the benchmarks, Dr Eisenach then seeks to explain certain disconformities and similarities between the selected transactions under which particular licences arise for the selected publications, on the one hand and the CA licensed publications, on the other hand. The applicants contend that Dr Eisenach has failed to explain how the particular publications and the selected transactions are representative of value of the publications he then values by reference to them.

(5)    (redacted) Isentia also says that (redacted) is targeted at public sector employees. Isentia contends that (redacted) are publications which influence politics and business disproportionately to their audience share. Their audience size is a function of the fact that “only paying subscribers have access to their content” and thus they are not representative of CA’s repertoire as a whole and reliance on them as benchmarks of wider value is apt to throw sand in the eyes of the analysis and inflate the value of the CA repertoire in a benchmarking analysis using audience data. For that reason alone, the applicants say that the benchmarking analysis is “fundamentally flawed”.

(6)    As to The Australian specifically, the applicants note Mr Gray’s evidence that The Australian “[c]ertainly … is one of the leading newspapers in Australia”. The applicants say that because The Australian (leaving aside paper sales to the public) is, relevantly for present purposes, a subscription only publication behind a “hard” paywall, its “relative audience size” is simply not reflective of its “importance” to the Australian “political and business community”. The applicants say that because of the importance of the publication in that context, an MMO, qualitatively, in order to provide a credible media intelligence service must put itself in a position to be able to access and monitor the content of The Australian. The applicants say that a benchmarking analysis that translates a fee paid for access to a qualitatively significant publication such as The Australian (because of its contribution to and influence upon national discourse on matters of politics, business and related cultural issues) as a benchmark of value attribution for all other Major Mastheads in the repertoire on the basis of proportional relativities derived from audience numbers “is simply not valid”.

(7)    The applicants say that, similarly, (redacted) are publications with “outsize” importance to the political and business community relative to their audience size. Each is a (redacted) influential in political and business circles and in the case of the (redacted). The applicants’ point is that value attribution on the basis that these publications are representative (true benchmarks) of the CA licensed repertoire, relative to audience size, is both unsound and very likely to over-value the CA repertoire. Dr Eisenach recognises that audience numbers (readership), without taking into account qualitative distinctions in the respective publications, “would not, by itself, be adequate”. Dr Eisenach has sought to address qualitative distinctions by identifying two broad clusters: Major Mastheads and non-Major Mastheads. The applicants contend that isolating these two broad clusters does not answer the criticism that the particular publications selected as representative benchmarks of all other publications within the two respective clusters is not truly representative of the CA repertoire (or other components of each cluster) for which the publications are used to attribute value.

532    The force of these criticisms became apparent in the course of the cross-examination of Dr Eisenach. As already noted by reference to the summary of values attributed to the mastheads in the Major Masthead cluster (ER 1, [230]), Table E-4 sets out the value assigned to each Major Masthead for (redacted)

533    (redacted)

534    (redacted)

535    (redacted)

536    We accept that newspapers (print and digital) predominantly associated with State-based catchments have a value to entities seeking to provide media monitoring services which includes access to and monitoring of the content of those newspapers and we also accept that a relevant factor in seeking to form a view about the value of a particular title is the number of people who “take” that publication either in print or by digital subscription. The difficulty we have is the attribution of value on a relative basis to each publication by reference to the benchmark publication of The Australian, by audience numbers by reference to the price paid in the circumstances of the withdrawal by News of its licensing arrangements for The Australian with CopyCo and CA.

537    News Corp elected to take The Australian and The Weekend Australian out of the CopyCo/CA orthodox licensing arrangements. It sees itself and its national leading Australian newspaper as outside of the ordinary and outside the ordinary arrangements for its licensing. It is seen by News Corp and others as a “leading” newspapers in Australia. Isentia had to have it. (redacted) Audience reach, print, digital and total, is no doubt one of the important factors in attributing value to a title. There are obviously other factors. It seems odd, however, to select as a benchmark of each and every Major Masthead the subject of the CA Proposed Licence, a publication that News Corp, by March 2020, had decided to remove from CA’s stable of licensed content. At the moment in time when the CA Proposed Licence (2018) was put to each applicant, The Australian was presumably thought to enjoy its status as a leading Australian newspaper. We know from the earlier licence negotiations in 2015 and 2016 the importance News attached to the value of The Australian for the purposes of the Isentia negotiations leading to the 2016 Isentia Licence. We have also seen the way News secured its special value as demanded of CA in those negotiations. It seems equally odd to then benchmark the value of every other Major Masthead in CA’s repertoire by reference to the amount paid by Isentia to secure access to the content of The Australian once News took its publication away from the orthodoxy of the CopyCo/CA licensing protocol. One thing that is perfectly clear in the light of the conduct of News in seeking to extract its conception of value for access to its content, is that News sees the notion that The Australian is just relatively, and proportionally, one of a number of Major Mastheads, as heterodoxy. Rather, it is seen as special and was seen as having special value especially in the 2015/2016 negotiations.

538    The applicants put these matters a little more emphatically by observing that The Australian, as an important national daily paper for record, is a publication an MMO must have and therefore must licence in order to be taken seriously or credibly in providing a media monitoring service. They say there are no “substitutes” for it and characterising the annual (redacted) as emblematic of a price point reflecting a willing but not anxious licensor and a willing but not anxious licensee, is itself a distortion because the analytical utility of such a price point must necessarily arise in circumstances where there is a workably competitive market that operates to condition or moderate the critical factors of “willingness” and “anxiousness”.

539    Moreover, the robustness check does not affirm the analytical integrity of the benchmarking exercise but rather suggests real caution in embracing it. The results of the robustness check (that is, the “alternative measure of relative value” based on clip data: ER 1, [248]) are set out at Table E-13 (ER 1, p 215) for Major Mastheads and Table E-14 (ER 1, p 216) for non-Major Mastheads. (redacted). The applicants contend that these values for The Courier Mail and The Advertiser as a relative derivation of value based on clip data produces a value output relative to The Australian that is simply “not plausible” and “unreasonable”. They test the value by asking: would someone pay News (or CA, as The Courier Mail was left by News Corp within the CA licensing protocol) (redacted) for a licence of the content of that publication if a licence of the content of The Australian is (redacted) On the clip data robustness check, those two publications are, for all practical purposes, of the same value.

540    Table E-14 sets out the estimated fair market value of non-Major Mastheads to Isentia (and Streem) (using the robustness check). It suggests a value for The Townsville Bulletin (a newspaper serving the Townsville northern catchment) to Isentia of (redacted) of the value of a licence of the content of The Australian) and to Streem a value of (redacted). At Table E-9 (see ER 1, [240] and [241]), the analysis of a reasonable range of fair market values of non-Major Mastheads for Isentia suggests that the value range for The Townsville Bulletin is (redacted) (print audience), (redacted) (digital audience) and (redacted) (total audience). The highest value (total audience) is (redacted) of the value at E-14 under the robustness check.

541    One of the non-Major Mastheads is a publication called “Realestate”. Table E-8 suggests a value range (for Isentia) for that publication of (redacted) (print audience), (redacted) (digital audience) and (redacted) (total audience). The Table E-14 robustness check suggests an estimated value to Isentia of (redacted) and an estimated value to Streem of (redacted).

542    Dr Eisenach makes the point that the virtue and utility of the robustness check lies in seeking to establish the overall reliability of the benchmarking exercise even though some results for some publications may suggest anomalies in the value. However, the applicants contend that the wide variations in the values demonstrates that “the analysis is not reliable in any way as a measure of value”. (redacted) The applicants say that the method deployed by Dr Eisenach in attributing value to each of the publications in the table attributes high value (high postulated licence fees) for many publications of little (if any) value to the MMOs (and, in particular, Isentia). The applicants have identified a range of lifestyle, general interest or real estate publications of no particular interest to an entity monitoring news and business media publications which have been assigned total audience values (redacted). The publications (with particular emphasis on the publications in bold) are these:

543    (redacted) The applicants characterise these valuation outcomes as simply “absurd” calling into question the utility and reliability of the valuation methodology.

544    The applicants also emphasise the table at E-12 of ER 1 which sets out, relevantly for present purposes, Isentia’s total clips and the clip ratio relative to The Australian for many of the publications in the above table valued by reference to total audience value. The number of clips and the clip ratio relative to The Australian is set out below:

545    The anomalies in the attributed total audience value (to Isentia) (Table E-8) for many of the non-Major Masthead publications and (redacted) (Table E-12) can be seen in the following table: