NZME Ltd v Commerce Commission
[2017] NZHC 3186
•18 December 2017
IN THE HIGH COURT OF NEW ZEALAND WELLINGTON REGISTRY
I TE KŌTI MATUA O AOTEAROA TE WHANGANUI-Ā-TARA ROHE
CIV 2017-485-445 [2017] NZHC 3186
BETWEEN NZME LIMITED
First Appellant
AND
FAIRFAX MEDIA LIMITED Second Appellant
AND
FAIRFAX NEW ZEALAND LIMITED Third Appellant
AND
COMMERCE COMMISSION Respondent
Hearing: 16−20 October, 24–27 October 2017 Counsel:
D J Goddard QC, A S Butler, S C Keene, T J Pilkington, M W McMenamin and C M Marks for the Appellants
J A Farmer QC, J D Every-Palmer QC, F J Cuncannon,
P I C Comrie-Thomson and G Spittle for the RespondentJudgment:
18 December 2017
JUDGMENT OF DOBSON J AND PROFESSOR RICHARDSON [Public, redacted version]
CONFIDENTIALITY:
The Court made orders on 31 July 2017 to protect the confidentiality of matters that were provided to
the Commerce Commission in confidence in the course of its investigation. Those orders are now made permanent, with the effect that no confidential material will be publicly available either in the content of the public version of this judgment, or by search of the Court file.
The unredacted version of this judgment has been distributed on a restricted basis to counsel and nominated representatives of the parties on the date appearing above. After receiving submissions from counsel on the extent to which confidential matters need to be redacted, the redacted version of the judgment was re-issued at 4.00 pm on 19 December 2017.
In a small number of paragraphs, text containing confidential material has been redacted. In a larger number of cases, where it is sufficient to support the content of this judgment by cross-reference, references have been made to confidential content in the non-publicly available version of the
NZME LIMITED v FAIRFAX MEDIA LIMITED [2017] NZHC 3186 [18 December 2017]
Commission’s determination. The scope of entitlement to access that document therefore dictates the scope of entitlement to have regard to that cross-referenced material in this judgment.
Contents
Introduction ............................................................................................................. [1] Outline of the appellants’ businesses................................................................... [3] The Commission’s determination ...................................................................... [14] Nature of the appeal ........................................................................................... [33] Grounds of appeal .............................................................................................. [39]
The clearance decision .......................................................................................... [41] Market definition................................................................................................ [41] The counterfactual ............................................................................................. [44] Functional dimension ........................................................................................ [47] Distribution of news and information? ............................................................. [52] Treatment of two-sided markets......................................................................... [60] Effect of zero price? ........................................................................................... [66]
Online national news reader market ................................................................... [70] Advertising constraints on quality ..................................................................... [72] Other business constraints on quality ............................................................... [87] Barriers to entry in the production of news ...................................................... [97] Multi-sourcing/multi-homing .......................................................................... [109] Prospect of a paywall........................................................................................ [118]
Sunday newspapers ............................................................................................. [125] The reader market ............................................................................................ [133] The advertising market .................................................................................... [139]
Community newspapers ..................................................................................... [149] The reader market ............................................................................................ [150] The advertising market .................................................................................... [160]
Conclusions on clearance appeal ....................................................................... [175]
The authorisation decision ................................................................................. [177] Relevance of loss of media plurality ................................................................ [177] Scope of Commission’s jurisdiction................................................................. [178]
Analysis............................................................................................................. [190]
(i) Legislative history ................................................................................. [190]
(ii) The position elsewhere .......................................................................... [196] (iii) Commission’s prior conduct and scope of expertise ............................. [202] (iv) Judicial consideration ........................................................................... [215]
Weight given to loss of plurality....................................................................... [232]
The plurality debate.......................................................................................... [236] Relative importance of media plurality......................................................... [242] Adequacy of internal plurality ...................................................................... [268]
Other unquantified benefits and detriments ................................................... [277] Quantified benefits and detriments.................................................................. [290] The overall assessment..................................................................................... [298]
Conclusion on authorisation appeal .................................................................. [306]
Process complaints .............................................................................................. [310] Conduct of post-conference interviews............................................................ [313] Terms and timing of instructions to BDO ....................................................... [337]
(i) Terms of instructions ............................................................................. [343]
(ii) Limited time to respond ......................................................................... [356]
Nature of any relief .......................................................................................... [362] Costs ..................................................................................................................... [365] Summary .............................................................................................................. [370]
Introduction
[1] The first and third appellants (jointly the appellants, separately NZME and Fairfax) both have substantial media businesses in New Zealand. In May 2016 they applied to the respondent (the Commission) for either a clearance or an authorisation to complete a merger of their New Zealand businesses. On 2 May 2017 the Commission declined both applications and the appellants have appealed from that determination.1
[2] In addition to substantive challenges to the correctness of the Commission’s determination, the appellants have raised criticisms of the process adopted by the Commission. Those criticisms might have been dealt with separately as an
application for judicial review. However, over the Commission’s objection, I
1 NZME Ltd and Fairfax New Zealand Ltd [2017] NZCC 8 [CCFD].
allowed the appellants to include them as a component of the appeal. We deal separately with the issues raised by those process criticisms towards the end of this judgment. The outcome on them has not had any bearing on arguments raised in the substantive appeal.
Outline of the appellants’ businesses
[3] NZME is a media and entertainment business with newspaper interests producing: print publications, namely the New Zealand Herald, the Herald on Sunday and the Weekend Herald; digital publications, including nzherald.co.nz; radio broadcasting businesses, including Newstalk ZB, ZM and Radio Hauraki; and other e-commerce services. NZME also has ownership interests in other newspaper and publishing companies. Its businesses are located in the North Island and include six daily newspapers, two paid weekly papers, 11 online versions of newspaper websites, two lifestyle websites, 10 radio station websites, 16 other websites, six magazines, nine radio stations, and 23 community newspapers.
[4] NZME was, until June 2016, known as Wilson and Horton Ltd and its current business resulted from a demerger from APN News & Media Ltd. NZME is listed on the NZX.
[5] Fairfax is a New Zealand subsidiary of the Australian entity, Fairfax Media Ltd, the second appellant. Fairfax produces numerous print publications, o perates websites and tablet and smartphone apps for stuff.co.nz, and has additional media interests by virtue of shareholdings in publishers of other newspapers. It also operates a website providing a private neighbourhood forum for neighbours to talk and share online. Fairfax’s principal newspapers include The Dominion Post, The Press and the Sunday Star-Times. In total, it publishes nine daily newspapers of which four are in the North Island and five in the South Island, three paid weekly papers, seven websites, 62 community publications spread throughout the country, and 10 magazines.
[6] The proposed transaction would involve NZME acquiring all of the shares in Fairfax. In exchange NZME would pay NZ$55 million in cash and would issue shares equal to a 41 per cent shareholding in NZME to an Australian subsidiary of
Fairfax Media Ltd. The Commission treated the appellants as each other’s strongest competitor in most of their businesses’ principal spheres of activity.
[7] In recent years the rapid growth of dissemination of news and information in digital form has caused a radical transformation of the way in which these activities occur. Media businesses such as the appellants have supplemented their traditional print products with digital forms of making news and information available. Both NZME and Fairfax have adopted a “digital first” strategy so that the content prepared for publication is first posted on the digital sites maintained by each of them (nzherald.co.nz for NZME and stuff.co.nz for Fairfax). A sub-set of the items posted since the last print publication are selected for hardcopy publication in the daily newspapers for each, including their masthead papers, namely the New Zealand Herald for NZME, and The Dominion Post and The Press for Fairfax.
[8] This transformation in the mode of dissemination has radically altered the sources of revenue for the businesses. Their circulation of printed products has dropped consistently, taking with it the level of subscription income and receipts from casual purchasers of the printed newspapers. Reduced circulation means that publishers cannot continue to charge at the same rate for advertisements in printed newspapers.
[9] At the same time revenue generated from advertisements on the appellants’ digital sites has increased with the growth in users going to those sites as a source of news. Neither of the appellants charge viewers to access their websites or apps, so the sole source of revenue from the digital form of their publications is the payments by advertisers for having their advertisements appear on those platforms. The amounts charged for digital advertising are dictated by the number of visits to the site or app, so that the appellants are incentivised to optimise the quality of their digital format. That supposedly means that they display the most popular content which encourages the largest readership of their digital forms of publication.
[10] One feature of the marketing of digital advertising in New Zealand is a joint venture between NZME, Fairfax, TVNZ and MediaWorks through KPEX Ltd.2
2 KPEX is an abbreviation of Kiwi Premium Advertising Exchange.
KPEX operates as a clearing house to sell unbooked digital advertising inventory that is available in any of the joint venturers’ digital platforms. The matching of advertisers to available digital advertising slots is conducted by algorithms that match the needs in advertising capacity without regard to an individual publisher’s requirements. Currently each of the four shareholders hold one quarter of the shares. [
].
[11] Media companies such as the appellants do not have the digital media platform to themselves. First, the internet gives readers access to an extensive range of worldwide major newspapers that have adopted similar strategies to the appellants in publishing their newspapers or equivalent content in digital form. Some of the world’s best known newspapers charge readers wishing to access their content,3 some afford access to some part of their content free of charge, but then require payment for full access,4 and others afford entirely free access to the whole of their publication in digital form.5 This extraordinarily expanded access to sources of international news means that there is no market within New Zealand confined to New Zealand producers of international news. The focus of any competition analysis is therefore on the production of New Zealand news, opinions, and other information relating to New Zealand.
[12] Second, the internet and digital technologies have facilitated the growth of businesses that collate and redistribute news produced by a range of sources. Internationally, this role is dominated by entities such as Facebook and Google. Their ability to very speedily collate and make available news and information with the facility to tailor content to the algorithmically predicted areas of interest of the viewer has transformed the manner in which news and current affairs information is accessed for those familiar with (and potentially reliant on) communications in
digital form.
3 For example The Times and The Wall Street Journal.
4 For example The New York Times and The Washington Post.
5 For example The Guardian, which does invite donations.
[13] The recent experience in New Zealand reflects similar changes in comparable economies. There has been a great deal of analysis and academic writing about how these financial arrangements can be rebalanced to enable digital forms of media publications previously produced only as printed newspapers to be maintained sustainably. There is a degree of pessimism, at least in some quarters, that a satisfactory outcome will be achieved.
The Commission’s determination
[14] The appellants sought a clearance from the Commission under s 66 of the Commerce Act 1986 (the Act), on the basis that their proposed merger would not have or would not be likely to have the effect of substantially lessening competition (an SLC) in a market. Their alternative application was for authorisation of the transaction under s 67 of the Act. The Commission is empowered under that section to grant an authorisation where, despite not having been satisfied that the acquisition will not have or would not be likely to have an SLC in a market, the acquisition would nonetheless result in such a benefit to the public that it should be permitted.6
[15] The Commission consulted widely in its consideration of the application. On
8 November 2016 it issued a Draft Determination which signalled the preliminary view that the proposed merger would be likely to substantially lessen competition in certain relevant markets, and that it was not likely to generate sufficient benefits to warrant authorisation. Following release of that draft the Commission convened a public conference to hear submissions for two days in early December 2016. After the conference further submissions and evidence were received and considered.
[16] When considering whether a proposed merger is likely to lead to an SLC, the Commission analyses the scope of the market or markets in which the participants’ commercial activities occur. It defines the relevant market or markets as a tool in helping to identify the state of competition in that market as it projects that occurring
in both the factual and counterfactual scenarios.
6 Commerce Act 1986, s 67(3)(b). That section is set out at [181] below.
[17] The Commission found that the appellants’ businesses operated in two-sided markets. The first side is a market between each of their businesses as producers of news and related content, and the readers who receive that content. Traditionally, that market represented the sale of newspapers to consumers. The second side is a market between the publishers of the media product and advertisers who pay to display advertisements to promote their goods and services to the readers of the media businesses’ product. In essence, the advertisers are paying for the attention of the readers to what is produced in the appellants’ publications.
[18] The task of defining the scope of the markets in which the appellants conduct their business is complicated by the distinct relationships between the appellants as producers of news and the consumers who read their publications on the one hand, and the appellants and the advertisers who pay for advertising space in their publications on the other. These two distinct sets of commercial relationships are nonetheless closely interdependent because the quality and reputation of the news produced by the appellants will dictate the extent of readership that they attract, and the demand for and value of the advertising space in their products will depend on the number of readers of their publications. Accordingly the Commission treated the appellants as operating in two-sided markets.
[19] For reasons that we address in considering the appellants’ criticisms of the determination, the Commission defined separate markets on both the reader and advertising sides of the platform for national news in the appellants’ daily papers (separately in their printed and online forms), for the Sunday newspapers and for community newspapers.
[20] In projecting the way in which the businesses would operate in the next two years both with and without the merger, the Commission considered it appropriate to assess the conduct of the participants on two scenarios: first, that within the relevant time frame the businesses (whether remaining separate or merged) would continue to operate more or less the existing mix of digital and print publications; and second, that the businesses (whether remaining separate or merged) would scale back their print publications to focus increasingly on production of digital news. In some
aspects of its analysis, the Commission made distinctions between the outcomes in
these alternative “digital and print” or “digital and limited print” scenarios.
[21] Because of geographical separation of the areas in which their printed daily newspapers circulate, the Commission found there was no likelihood of an SLC in the reader aspect of that market (that is, between producers of newspapers and buyers of them). However, the Commission did identify the likelihood of an SLC in the production of online national news.
[22] The Commission considered the market for Sunday newspapers as a separate one. Fairfax publishes the Sunday Star-Times and the Sunday News. NZME publishes the Herald on Sunday. Only Fairfax Sunday newspapers are circulated in the South Island, but there is direct competition between the Sunday newspapers published by both appellants in the North Island. The Commission determined that there would likely be an SLC in both the reader and advertiser markets for Sunday newspapers in the North Island.
[23] Both appellants have substantial businesses in producing community newspapers. These are typically published weekly and distributed free within defined local boundaries. NZME produces 23 such publications and Fairfax 62 of them. Their publications compete in 10 areas in the North Island.7 The Commission found that there would likely be an SLC in both the reader and advertiser sides of community newspaper businesses in the 10 areas in which the appellants compete.
[24] Because of the findings of the likelihood of SLCs in those markets, the Commission declined to grant a clearance for the proposed merger. The Commission went on to consider whether it should nonetheless grant an authorisation for the merger.
[25] This task involved an analysis of the projected benefits and detriments of the merger in both quantifiable and unquantifiable forms. Because of the significant
7 Whangārei, Hamilton, Rotorua, Taupō, Napier, Hastings, Stratford, Palmerston North, Horowhenua, and Kāpiti.
uncertainties in projecting the circumstances that would pertain in the first years of the merged entity, the range of quantified benefits/detriments outcomes was wide:8
Table 17: Estimated net quantifiable impact – ‘digital and print’ scenario
Time frame
High detriment/low benefits
Low detriment/high benefits
5 years
$41 million
$204 million
Table 20: Estimated net quantifiable impact – ‘digital plus limited print’ scenario
Time frame
High detriment/low benefits
Low detriment/high benefits
5 years
$55 million
$196 million
[26] These quantifiable benefits arose from projected savings related to overhead costs such as marketing, IT, premises and management costs, as well as employment and other operational cost savings that were likely to follow from the merger.
[27] The Commission acknowledged that there could also be unquantifiable benefits if the merger prolonged the viable life of various print publications and the overall level of editorial resourcing that could result from a strengthening of the businesses’ financial position. The Commission took the view that this advantage was likely to be relatively transitory.
[28] The Commission also assessed unquantifiable detriments that it saw as arising from the merger. This focussed primarily on the loss of media plurality which provides a diversity of viewpoints and editorial approaches under the existing competitive ownership structure. The Commission found that plurality of the news media is essential to maintain a well-functioning democracy because it facilitates the availability and exchange of a divergence of views. Given that a merger would result in a single organisation controlling nearly 90 per cent of all print media, New Zealand’s two largest news websites and one of New Zealand’s two largest
commercial radio companies, the Commission was concerned that the merger would
8 CCFD, above n 1, at [1326] and [1329].
result in a level of media concentration unprecedented in a well -established liberal democracy.9
[29] The Commission was not persuaded that commitments by the appellants to maintaining divergent editorial policies so as to maintain internal plurality within the merged businesses would be adequate to maintain the external plurality that presently exists. The loss of plurality was found likely to be significant and potentially irreplaceable. The Commission treated the relative importance of maintenance of plurality as so significant that its conclusion on the point was not a finely balanced one.
[30] The Commission also found that the merger would lead to a reduction in the quality of news content as a result of loss of competition. This was classified as a non-price allocative efficiency loss. The Commission treated the extent of this reduction in quality as likely to be significant.10
[31] The Commission did not consider that the need to attract audiences in order to attract advertising would constrain the merged entity from cost saving initiatives that would detrimentally affect the quality of the merged entity’s product. In addition the Commission saw the appellants as playing a particular role in setting the agenda for news produced by other publishers. The Commission was concerned that a reduction in quality of the appellants’ output would also have a flow on effect in reducing the quality of other sources of news.
[32] It concluded that these detriments, although unable to be quantified specifically, were significant, and clearly outweighed the significant quantifiable benefits that were projected to arise from the merger. The Commission accordingly
declined to authorise the merger.
9 At [1728].
10 At [1672].
Nature of the appeal
[33] Section 91 of the Act gives unsuccessful applicants to the Commission a right of appeal against the Commission’s determination. This is a general right of appeal, to be conducted by way of re-hearing.
[34] The standard to be applied by the Court on appeal is that described by the Supreme Court in Austin, Nichols & Co Inc v Stitchting Lodestar.11 This requires the Court to form its own views on the merits of the case. The appellants bear an onus of establishing that the Commission’s determination is incorrect and, if that point is reached, the Court has jurisdiction to modify or reverse the determination or any part of it. The Court has the powers that could have been exercised by the Commission in relation to the matter that is appealed.12 If considered appropriate, the Court can send the matter back to the Commission for reconsideration in light of an identified error.
[35] The Court is not obliged to give any particular level of deference to the Commission’s views. The Court can have regard to the range of expertise appropriately attributed to the original decision maker, and any advantages the original decision maker had in assessing the evidence and submissions as presented to it.
[36] In this case counsel for the Commission emphasised that the Commission conducts an inquisitorial process for which it is expertly equipped by the range of skills possessed by Commission members and the investigative staff assisting them. The Commission has a reservoir of institutional expertise in investigating and analysing the sequence of issues that arose in dealing with both applications. It was submitted for the Commission that these advantages warrant a usual degree of deference being shown to its decision as that of a specialist body.
[37] It was submitted for the appellants that the Commission had neither experience nor expertise to evaluate the critical issue determ ined against them on
their authorisation application, namely the relative importance of maintaining
11 Austin, Nichols & Co Inc v Stitchting Lodestar [2007] NZSC 103, [2008] 2 NZLR 141.
12 Commerce Act 1986, s 93.
plurality in the mainstream provision of news. Accordingly, no deference should be accorded to the view it came to on that issue.13
[38] We agree with the appellants that the Commission did not have expertise in evaluating the relative importance of plurality in the media. On other issues the material evidence was largely documentary so the Commission did not have an advantage in observing it first-hand. Ultimately our decisions on the important issues raised by the appeal are not affected by any measure of the level of deference appropriately given to the views of the Commission.
Grounds of appeal
[39] The appellants challenged the definition of the markets in which the activities occurred, as well as the Commission’s findings on the prospect of each of the possible SLCs. These were components of a wider challenge that the Commission ought to have given a clearance to the transaction. On numerous parts of the appeal there were also more detailed levels of criticisms of the approach adopted by the Commission. We deal with the criticisms of the Commission’s analysis and decision on the clearance application in the sequence that appears in the contents at the outset of the judgment.
[40] In the event that the appellants are not successful in having their clearance application upheld on appeal, they also argued that the Commission’s assessment of benefits and detriments of the merger was wrong in sufficient respects that the Court should grant an authorisation. They contend the sum of the benefits outweighed the sum of relevant detriments. The appellants’ primary ground of challenge was that the Commission had taken into account the likely detriment from a loss of media plurality which they argue is outside the Commission’s jurisdiction. If that challenge is not upheld, the appellants argue that the Commission’s evaluation of detriments when weighed against identified benefits arising from the merger was wrong, so that an authorisation should be granted. Again, the sequence of the issues addressed
appears in the contents at the outset of the judgment.
13 This argument was secondary to a prior objection that the terms of the Act do not provide any jurisdiction for the Commission to have regard to broad societal concerns so that the Commission exceeded its jurisdiction by having regard to anything beyond economic considerations.
The clearance decision
Market definition
[41] The Commission identified six markets in which the appellants operated that it found were relevant to the competition analysis. These markets comprised the advertising and reader sides for online national news, Sunday (print) newspapers, and community newspapers (in the 10 areas within the North Island where the appellants’ publications overlap). The appellants mounted a number of criticisms of the Commission’s definition of the scope of these markets.
[42] Numerous judicial decisions have emphasised that market definitions are but a tool used in various competition law contexts to provide a framework for analysis of the relevant competition law concern, in this case constraints on competition. An early description of the function of market definition, reflected in numerous more recent decisions, is that of French J in Singapore Airlines Ltd:14
In competition law [market definition] has a descriptive and a purposive role. It involves fact-finding together with evaluative and purposive selection. … In its statutory setting the market designation imposes, on the activities which it encompasses, limits set by the law for the protection of competition. It involves a choice of the relevant range of activity by reference to economic and commercial realities and the policy of the statute. To the extent that it must serve statutory policy, the identification will be evaluative and purposive as well as descriptive.
[43] The parameters of any market as defined are not necessarily determinative of the analysis of competitive constraints on the conduct of participants in that market. Where the market is appropriately defined in narrow terms, constraints from outside the market must still be taken into account if they operate on the state of competition.15
The counterfactual
[44] A necessary component of the Commission’s analysis is to identify one or more counterfactual scenarios that would apply if the merger did not proceed. Treating the
14 Singapore Airlines Ltd v Taprobane Tours WA Pty Ltd (1991) 33 FCR 158 (FCA) at 174. See also
Woolworths Ltd v Commerce Commission (2008) 8 NZBLC 102,128 (HC) at [270(d)].
15 Brambles New Zealand Ltd v Commerce Commission (2003) 10 TCLR 868 (HC) at [137].
scenario that would pertain if the merger is allowed as the factual, for the purposes of comparison the Commission must settle on a counterfactual of the scenario that would pertain without it. The Commission is not obliged to define the counterfactual scenario that it considers to be the most likely if the merger does not proceed and, provided it is within the range of scenarios that are likely, the Commission can adopt the counterfactual that would give rise to the most competition concerns.16
[45] Identifying likely counterfactuals in this case is challenging because the state of news media markets is changing, in some respects dramatically and at considerable speed. Assuming continuation of the status quo is therefore unlikely. The Commission adopted a counterfactual in which both applicants would be likely to increasingly focus on developing their online news businesses, and their print products would be likely to continue to diminish in number and comprehensiveness over time.17 The Commission took the view that the future for the appellants’ print publications was likely to be similar in both the factual scenario where the merger proceeded, and in the counterfactual where their businesses remained under separate ownership.18
[46] The appellants abandoned criticisms that the Commission had erred in the definition of the counterfactual that it adopted. We have adopted the Commission’s counterfactual for the purposes of assessing the appellants’ criticisms. In its authorisation decision, the Commission adopted two possible scenarios under which the merged business would either continue more or less its present scale of print publication businesses, or alternatively would maintain a more limited range of print publications.
Functional dimension
[47] One of the criteria recognised by the Commission in its Mergers and
Acquisitions Guidelines when testing the prospects of substitution for the goods or services in issue is the level in the supply chain at which the parties operate.19 On this
16 CCFD, above n 1, at [63]; Commerce Commission v Woolworths Ltd [2008] NZCA 276, (2008)
12 TCLR 194 at 207; and Godfrey Hirst NZ Ltd v Commerce Commission [2016] NZHC 1262, [2016] 3 NZLR 645 [Godfrey Hirst 2 (HC)] at [70].
17 At [X15].
18 At [183].
19 Commerce Commission Mergers and Acquisitions Guidelines (July 2013) at [3.14.3].
aspect of the market definition, the Commission defined “two separate functional markets for New Zealand news” being the upstream production of New Zealand news content and the downstream distribution of that content across two-sided platforms.20
The appellants criticised the Commission for what they argued was an inaccurate characterisation of a functional market.
[48] The appellants criticised the Commission’s approach on the ground that the separate activity involved in producing news content could not constitute a market on its own until there was a supply that was intended to generate a return on the cost of production.
[49] Holding the Commission to what the appellants treated as a separate market for the production of news, they argued that there was no competition issue identified at the functional level of news production. It would then follow that all competition concerns should focus on the distribution or supply of news content, where the merged entity would arguably face significant competition so that no prospect of an SLC could arise.
[50] We do not accept that the Commission did analyse the appellants’ activities as involving participation in two separate functional markets. Other comments in this part of the determination treated the appellants as vertically integrated firms that operate at both the upstream and downstream levels.21 Having identified the two functional dimensions, the Commission commented:22
The primary focus of our competition analysis is in the upstream production of New Zealand news content and how this is accessed by consumers. However, we recognise that the desire to attract consumers at the distribution level influences the quality of content that is produced to some degree.
[51] We agree with the appellants that production alone does not constitute a market standing on its own until the product is traded either to wholesalers or retail consumers. However on an evaluation of the Commission’s reasoning, we do not consider that it
proceeded on the basis of separate markets. The functional distinction was ultimately
20 CCFD, above n 1, at [581].
21 At [576].
22 At [582].
not relevant because the Commission, correctly in our view, treated the appellants as being vertically integrated and being the biggest players in both components of the wider functional market including upstream production and downstream distribution.
Distribution of news and information?
[52] Before the Commission the appellants argued that the relevant market should be provision of news and information services. The addition of the latter category was rejected by the Commission.23 It treated New Zealand news content as a separate product, for which the appellants competed with a limited number of others who also produced it. Irrespective of how definitional difficulties in identifying the scope of information services are resolved, we agree with the Commission that adding such services would erroneously broaden the market. For the purposes of the appellants’ applications, the market is appropriately defined as production and dissemination of news.
[53] A related aspect of the appellants’ criticism of markets as defined by the Commission being confined to the production and dissemination of news is the line between hard news on topics of national and local interest on the one hand, and items in the nature of gossip on the other. Mr Goddard QC, counsel for the appellants cited comments by a media representative interviewed by the Commission to the effect that while public discussion focused on the merits of high quality journalism, the public tend to be more interested in gossip.24
[54] We do not accept that any blurring of the line between news and information contributed to by consumers’ misrepresentation of the topics that maintain their interest could justify an expansion of the market for production and distribution of news.
[55] The appellants also argued that collators and distributors of news like
Facebook and Google are treated as sources of news by many readers. From their
23 At [536.2].
24 Transcript of interview CC.FS.4098 at 15.
perspective such distributors are close substitutes for accessing news via the online versions of the producers of news such as the appellants.
[56] The Commission maintained several grounds for distinguishing news content made available by Facebook and similar sites. Those sites channel news that has been produced by others, with the content readers are presented with being haphazard by comparison to the ordered content of sites maintained by online producers. The content offered by such collators is rated by popularity rather than by reliability, timeliness, or its importance. The Commission submitted that readers wanting serious journalism will go to a website or use a news media organisation’s app.
Mr Every-Palmer QC, counsel for the respondent cited statistics showing that readers who arrive at the online Stuff and Herald websites via Facebook spend much less time on them than those readers who go directly to those sites.
[57] Mr Goddard submitted orally that a New Zealand Law Commission report published in 2013 adopted a more forward-looking and realistic approach to the scope of news media in New Zealand than that adopted by the Commission in the determination.25 The Law Commission report addressed the need for, and possible scope of, regulation of “new media”. Mr Goddard suggested that the Law Commission’s approach included collators, commentators and distributors of news as a component of the news media in New Zealand. Arguably the Commission ought similarly to have included them within its market definition.
[58] The Law Commission did, however, maintain the distinction between those generating news and those disseminating the news produced by others:
[4.56] First, and most significantly, we concluded that this proliferation of publishers is enriching public debate and has the potential to strengthen democracy by increasing participation in public affairs, widening the sources of information available to the public and providing a greater diversity of opinion. It is also providing a new form of accountability for the mainstream news media as bloggers and others critique aspects of the mainstream media’s coverage of political and other events.
[4.57] However, we also noted a number of caveats: despite the massive proliferation of publishing online, only a small percentage of this new publishing activity is focused primarily on the generation and dissemination
25 Law Commission The News Media Meets “New Media”: Rights, Responsibilities and Regulation in the Digital Age (NZLC R128, 2013) (citations omitted).
of original, local, news and current affairs. For example, we identified only a small number of professional, internet-native entities for whom this was the primary focus: these included sites such as Scoop, NewsWire, BusinessDesk, allaboutauckland.com and interest.co.nz.
We consider that the Law Commission’s appreciation of this distinction is consistent with the approach adopted by the Commission in the present determination.
[59] We consider that the distinction between producers of news and collators and redistributors of news produced by others remains a relevant one and the Commission’s approach was appropriate. Although it oversimplifies the point,
Mr Every-Palmer’s distinction that those accessing material via Facebook are accidental rather than deliberate newsreaders has some merit.
Treatment of two-sided markets
[60] The appellants also criticised the way in which the Commission treated the interrelationship between the two sides of each of the markets that operated on a platform – that is, the provision by the appellants of news reports to readers, and the provision by the appellants of publication of advertisements for advertisers.
[61] The arguments on both sides tended to exaggerate the difference between them on the relevance of two-sided markets, and the relative importance of this component of the analysis in correctly defining the nature of the markets in which the appellants presently compete.
[62] Counsel referred us to a range of published articles on the features of different two-sided markets. Those articles included various hypotheses on how such markets should be treated in competition analyses. Apart from media businesses, another tripartite activity often considered in the research is the business of credit card companies. They depend, in interrelated ways, on the number of merchants prepared to accept payment from purchasers by use of their credit card, which involves a cost to the merchant in procuring payment from the credit card company. The relative popularity of the credit card company with consumers will depend, at least in part, on how widely merchants will accept their credit card as a form of payment and its
popularity with merchants will depend, similarly, on how widespread is its use by consumers.
[63] One of the articles we were referred to drew a distinction between the position on the one hand of credit card companies, and on the other hand media businesses where the firm transacts separately with purchasers of the news (in whatever form), and with advertisers. Unlike the second direct transaction between a merchant and a cardholder, the present relationships do not require any relevant transaction between the advertisers and the consumers of the news. The point was expressed in the following terms:26
However, in a case involving newspapers, a product might be in the relevant market on the advertiser’s side but not on the readers’ side. For instance, suppose that people do not regard TV and newspapers as substitutes because they read newspapers on the metro going to work and watch TV at home in the evening. However, if advertisers are interested in reaching each person only once during a day, they will tend to regard TV and newspapers as substitutes. TV would then be in the same relevant market as newspapers on the advertiser’s side but not on the readers’ side. The analysis of a merger case involving newspapers should then consider that TV exerts competitive pressure on newspapers in the market for advertising, but not in the market for content.
We submit that the crucial element distinguishing a newspaper market from a payment cards market is that, in the latter, a transaction is present between so- called end-users – that is, between customers on the two sides of the market.
[64] We agree with that analysis of the two-sided markets in which the appellants operate. We also agree with the approach adopted by the Commission. It considered each side of the platform individually, but also took into account the constraints arising from the activity on the other side of the platform. As another respected writer on anti- trust economics has observed, the importance of the interaction between the two sides is an empirical question.27
[65] An aspect of the treatment of two-sided markets was the appellants’ criticism that the Commission did not adequately take into account the extent of strong
interdependence or “feedback loops” in assessing the competitive constraints that the
26 Lapo Filistrucchi and others “Market Definition on Two-Side Markets: Theory and Practice” (2014) 10 Journal of Competition, Law & Economics 293 at 301 (citations omitted).
27 David S Evans “The Antitrust Economics of Multi-Sided Platform Markets” (2003) 20 Yale J Reg
325 at 358.
appellants argued for. There is scope for debate on the weight to be given to the nature and effect of the interdependence, but we consider the Commission’s appreciation of interdependence was at a point within the correct range on this topic. A preference for any different level of interdependence which would fall within that range would not change the outcome on either market definition or the nature of competitive constraints.
Effect of zero price?
[66] The appellants’ readers or viewers (we henceforth refer to them as “readers”, irrespective of the form of publication they are accessing) accessing their online news sites do not pay to do so. Adopting an approach used in 2009 and 2012 OECD reports on two-sided markets,28 the appellants submitted that analysis of competitive constraints must encompass both sides of the market because the pricing and production decisions are so closely interdependent and competitiveness in the entire market should reflect the sum of all prices. Given the Commission’s conclusion that there was no SLC on the advertisers’ side, that conclusion ought to apply to both sides of the two-sided market. Where the price paid by the advertisers is only added to by the “zero” price the reader pays, arguably the competitiveness should therefore be determined by the competitiveness of the advertiser side of the platform.
[67] We do not agree that this analysis can determine the prospect of an SLC on the reader side of the online news market. It is unnecessary to review in detail the economic analysis on the impact of a zero price when that factor at least sends a warning signal that the reader accepting a “free” service will likely be making a contribution in different form to direct monetary consideration passing to the producer of the online news. As the author of a recent article on the topic observed:29
Human attention, both valuable and limited in supply, is a resource. … It has become commonplace, especially in the media and technology industries, to speak of an “attention economy” and of competition in “attention markets”. In the business slang, firms in the attention economy “compete for eyeballs”;
28 OECD Policy Roundtables: Two-Sided Markets DAF/COMP(2009)20 (2009); and OECD Policy
Roundtables Market Definition DAF/COMP/(2012)19 (2012).
29 Tim Wu “Attention Markets and the Law” (unpublished paper, 2017) at 2 (citations omitted). This paper has since been updated and can be found at papers.ssrn.com: Tim Wu “Blind Spot: The Attention Economy and the Law” (unpublished paper, 2017).
and in another widely used metaphor, whoever gains attention “sucks oxygen”
from everyone else.
[68] In this two-sided market, that price is the attention “paid” by readers to the website, necessarily including the advertisements that appear there for which the producer of the news is paid by the advertisers. The cost to readers of giving their attention to a news website is recognised as a form of consideration passing from them, despite the absence of a monetary cost. The number of visits to a publisher’s website is monitored and directly affects the cost of an advertisement, which is calculated at a certain cost per given number of visits to the site. In that sense the reader does provide a form of consideration in going to the site and thereby counting as a view that contributes to the advertising revenue earned.
[69] We are not persuaded that the zero price feature of the online news market needs to be treated differently from the way in which it was analysed by the Commission.
Online national news reader market
[70] The Commission rejected the appellants’ wider market definition that would include different types of media so that television, radio and printed newspapers would be combined with online news products. In doing so the Commission rejected the approach adopted by its Australian counterpart, the ACCC, in a recent informal review of a merger of media assets in Western Australia.30 The ACCC did not form a view on whether printed newspapers and online products were within the same market, preferring to focus on the degree of constraint each provided the other. The Commission recognised that other platforms may provide some constraint, but found that it was appropriate to define separate print and online markets.31 The different approach to that of the ACCC is appropriate given the materially different scale of the merger participants’ activities in the areas in which they were operating.
[71] In assessing the reader market for online news, the Commission found that the merger would likely result in a reduction in quality which would be achievable because
30 ACCC informal review: Seven West Media Limited – proposed acquisition of the Sunday Times publication and website (15 September 2016).
31 CCFD, above n 1, a [536.3].
of an SLC. The appellants challenged this conclusion on four grounds. First, it resulted from an incorrect view on market definition. We have dealt with this criticism in the matters we have already traversed.32 Second, the Commission did not give proper weight to evidence about the pressure from advertisers on the other side of the two-sided market for the producer to maintain quality. Third, the Commission failed to give adequate weight to the extent of constraint on quality that was provided by the range of other online news providers. The second and third grounds require a review of the Commission’s analysis on barriers to entry in the production of news, and the relevance of multi-sourcing/ multi-homing. Fourth, the appellants contend the Commission wrongly assessed the prospects of a paywall being introduced by the post-merger business.
Advertising constraints on quality
[72] The appellants submitted that the pressure from advertisers to maintain quality would operate to prevent any SLC. The concept of quality of online news encompasses a number of dimensions. The dimensions considered by the Commission included:
• coverage of important and relevant news topics;
• coverage of a variety of perspectives, opinions and viewpoints on common news topics; •
•
in-depth analysis and investigation; and
timely and accurate reporting.33
[73]
In
assessing presentation and delivery as distinct from content, the
Commission also recognised these dimensions:
32 See [47]–[69] above.
33 CCFD, above n 1, at [825].
• the design and format of the online platform; and
• inventive ways of presenting individual stories.34
[74] Mr Goddard emphasised there was an economic imperative that he submitted would arise post-merger for the business to maintain and enhance the quality of its online news. It was only by doing so that the business would optimise the circulation and the number of visits to its websites, thereby optimising advertising revenue.
[75] The contrary hypothesis is that the immediate imperative for the merged entity would be to reduce costs. Condensing journalistic and editorial resources would be so compelling an area for cost reduction that it would be difficult to justify completing the merger without doing so in a material manner. The financial imperative to reduce
costs [
] would almost certainly be put to management.
[76] Of these competing options, we consider it more likely that the merged entity would adopt the second strategy. We consider it an inevitable part of the rationale for the merger that efficiencies pursued would include scaling back journalistic and editorial resources, as well as managerial resources. One necessary consequence of doing so would be a reduction in quality. We are therefore satisfied that there would be a material reduction in the scope of topics reported upon, and the diversity of views expressed within them. That would constitute a reduction in quality.
[77] We do not consider it a sufficient answer that the merged business would be economically incentivised to provide coverage on what readers want. Dissatisfaction with a reduction in the scope of what is published is unlikely to be conveyed either promptly or effectively enough to influence management decisions on what may appear to be duplicated resources, but which readers would prefer were retained. So too with the level of more arcane topics that are likely not to be covered post-merger. We agree with the Commission that the merged entity could reduce the quality of its
output in numerous respects that would not be observed by consumers in any concerted
34 At [826].
way, so that the reduction in quality could be effected without producing any compelling demand for restoration of the qualities that existed in the competitive era.
[78] We also agree with the Commission that competition between news producers stimulates coverage by each of a wider range of topics than would be covered in a non- competitive environment. If one publication covers a news story, then its rival will be incentivised to cover that story as well, albeit with a different perspective.
[79] We are mindful that senior editorial staff currently employed by the appellants provided assurances of their intentions to maintain variety and diversity of news content. We have no reason to doubt the integrity of the editorial personnel and respect their commitment. The commercial reality is that the extent to which they can achieve variety and diversity of news content will be dictated by management decisions.
[80] We were not referred to any evidence that might enable an arithmetic calculation of the impact of a reduction in quality causing a reduction in the number of visits to an appellant’s online site. The competing views are therefore intuitive.
[81] Mr Goddard explained that advertisers usually pay for their advertisement to be shown when readers open a given online page, per specified number of visits to the page. Once the number of visits an advertiser has paid for has occurred, that advertisement will disappear and the next advertiser to have paid will have its advertisement appear.
[82] The decision confronting management of the merged entity can be illustrated by a simple numerical example. We assume, for the purposes of it, that each of two firms, pre-merger, has the same number of readers, reader habits and cost structures. Say that each firm incurs $100,000 in costs to provide news coverage on health matters and that the visits to their sites to read health news coverage generates $110,000 in advertising revenues for each publisher, provided they maintain a measure of diversity and difference in the health topics covered; were no such diversity to be provided consumer demand would be such that only $200,000 of revenues would be created. Each firm then has a clear competitive incentive to provide coverage differentiated from that of its rival, earning profits of $10,000 versus the $0 profit they would earn
from providing identical perspectives. The merged entity could then earn $20,000 in profit by replicating the competitive solution, but could earn $100,000 of profit by, instead, shutting down one of its health news coverage resources, losing $20,000 in revenues but saving $100,000 in costs.
[83] This fictitious example is entirely hypothetical and its exact conclusion would be modified by the presence of meaningful third party competition, but it illustrates that the structure of competition can materially affect a firm’s supply decisions; the mantra that ‘a firm must supply what the market wants’ is overly simplistic. We consider that in numerous choices that would confront the merged entity on rationalisation of news production resources, it is more likely than not that the financially beneficial outcome would be to rationalise the production resource rather than incur a measure of duplicated costs for the sake of maintaining modest or marginal increases in advertising revenue.
[84] The appellants enjoy a strongly dominant position in the New Zealand online news market. On the extent of editorial and journalist resources the merged entity would employ over 300 more editorial staff than the three next biggest mainstream media organisations combined.35 A survey in the period between February 2016 and January 2017 showed that New Zealanders spent an average 5.5 million hours per month accessing stuff.co.nz and nzherald.co.nz. This is more than seven times the number of hours spent on the three next largest New Zealand news websites, amounting to some 700,000 hours per month for visits to newshub.co.nz,36 radionz.co.nz and tvnz.co.nz.37
[85] Other statistics discussed below demonstrate that they each attract many times more visits than the next most popular sites, and their combined number of visits compounds their dominance in that market.38 On the basis of the evidence, we consider that they have earned positive reputations because of their longstanding
strengths in production of printed daily newspapers with positive reputations for
35 CCFD, above n 1, at [656].
36 “Newshub” is the collective brand under which the news platforms of MediaWorks operate.
Newshub and MediaWorks are used interchangeably in this judgment.
37 CCFD, above n 1, at [678], fig 5.
38 See [102]–[105] below.
quality and reliability transferring across to the online form of their products. Given that position, it appears likely that management of the merged entity would be confident that their business would withstand a drop in quality caused by rationalisation of resources, without a reduction in the level of online advertising revenue they generate to an extent that exceeded the available cost savings.
[86] The appellants disputed that the merged entity could reduce the quality of its online news product without its readers noticing a drop in quality or using their websites less. If that were so, arguably the appellants would not risk a drop in readership because they would instead be motivated to optimise the numbers accessing their website when it has so direct a bearing on their source of revenue. This risk would be recognised as so likely to follow that the merged entity would not take it. However, we consider it foreseeable that such adverse changes in quality could occur without readers of their online sites being aware that it had occurred. We do not accept that the merged firm would maintain all standards. There are greater financial imperatives to cut costs.
Other business constraints on quality
[87] As to the third ground, the appellants argued that competitive constraints were imposed on the quality of the appellants’ online products by a range of other businesses including TVNZ, RNZ,39 and Newshub. Each of those news producers now have an online presence, accessible in the same way as the appellants’ digital products.
[88] The appellants also argued that the online news market included specialist publications, such as the National Business Review (NBR), that produce relatively small amounts of news in niche areas (in the NBR’s case, business reporting). The appellants’ market definition would also include collators and distributors of national news such as Facebook, Google, Twitter and syndicators such as Spinoff, Newsie and AAP News Wire.
[89] The appellants argued that the Commission had erred in excluding this last group of news distributors from the market for the provision of online national news.
39 Formerly Radio New Zealand.
Alternatively, if the market was defined narrowly as the Commission did, then it failed to give appropriate weight to the constraint on anticompetitive behaviour that would be exerted by all of these other online news providers and distributors from outside such a market. Mr Goddard labelled collators who present news produced by others to digital audiences defined by algorithms as “disrupters”. He predicted that their power in doing so would expand their influence to an extent that constrained the appellants, despite the disrupters not producing news content of their own.
[90] We are satisfied that the market for online production of national news should exclude collators and distributors such as Facebook and Google, and other operators of websites that package and in some cases comment on news produced by others. Observed patterns of behaviour suggest that readers are likely to access content from sites operated both by producers and by collators. However those providing the content for online news on such sites have carried out distinct functions, with production of original news being a recognisable and distinct feature. In seeking out reliable original news, visitors to collators’ sites are likely to discriminate in their level of attention, placing greater credence and therefore spending more time on items from reputable producers of news.
[91] In their competing online formats, the other mainstream producers of national news, namely TVNZ, Newshub and RNZ, do provide a measure of constraint on the potential for the merged entity to reduce the quality of its online product. The appellants submitted that the Commission failed to deal with evidence that those established competitors would have the incentive and capacity to fill any gap in the quality of online national news, by providing enhanced diversity or quality in their own reporting.
[92] Within this category, we agree with the Commission that there are grounds for doubting the capacity of both TVNZ and RNZ to materially expand the scope of their online national news coverage. TVNZ’s chief executive had recently announced a reduction in newsroom staff. Other information, provided confidentially by [ ] added to the grounds for the Commission’s finding that it would be unlikely to seek to
expand if a gap was left by the contraction of the scope, or reduction in quality, of the
merged firm’s online websites.40
[93] [
], we consider that the prospects of material expansion
in the online news coverage by RNZ are limited by the extent to which it is able to procure greater Government funding. In mid 2017, there was a commitment to a measure of increased funding, but it was not calculated to fund any substantial expansion of its capacity to produce online national news.
[94] TVNZ, RNZ and Newshub have all entered the online news market with established brand awareness but, despite that, the data before the Commission on market shares showed that they were getting relatively low online market shares.41
[95] Adopting the statutory test, we consider it unlikely that the existence of sites managed by collators and distributors can operate as a meaningful constraint on the quality of the online news product produced by the appellants, or the merged entity post-merger.
[96] The prospects of competition from new entrants are more limited. We are satisfied that a reputation for reliability and quality of news cannot be gained in a short period of time and the requirement for a positive reputation constitutes a significant barrier to entry. It is less relevant that costs of increased distribution of original online product may be insignificant.42
Barriers to entry in the production of news
[97] A routine component of the analysis of whether merger participants would enjoy market power is an analysis of the extent of barriers to entry into the market in which they operate. Although this raises many of the considerations we have just dealt with, the focus on this topic in argument bears some repetition when viewed through
this lens. The appellants submitted that numerous online news platforms now provide
40 CCFD, above n 1, at [743]–[744].
41 See [102]–[105] below.
42 Interview with Newsroom principals CC.FS.0848.0002 at 19.
more than adequate competition for the online forms of the appellants’ newspapers. They characterised the extent of that activity as increasing rapidly. Arguably, these new ventures require little by way of set-up capital, and a flexible extent of resources to maintain a website that, depending on its quality, can be a valid alternative to the appellants as a source of New Zealand news.
[98] We have dealt at [89] to [95] above with the appellants’ arguments that relevant competition includes collators and distributors of news.
[99] Collators and distributors of online news do assist in promoting plurality by inviting readers to access numerous sources of news. However we remain of the view that a valid distinction should still be drawn between the production and distribution of news. Fifteen recyclers of the product of two producers of news are still only making two views available. We therefore do not test barriers to entry by considering the position of entities such as Facebook and Google.43
[100] There are significantly fewer alternative producers of news. The five identified in the Commission’s analysis were the two appellants, plus MediaWorks, TVNZ and RNZ. An analysis of consumption of news by media type was compiled by the Commission for its determination. The table contained numerous statistics that remain confidential, and the more detailed points arising from it cannot be made without referring to those details.44
[101] The table confirms that the appellants’ daily circulation of its printed newspapers is overwhelmingly dominant in that part of the market. Despite certain impressions of a wholesale change of reading habits to online sources, the Commission noted elsewhere that a sizeable group of readers continue to consume daily newspapers with total circulation of all daily newspapers being approximately
420,000 per day, which is equivalent to a quarter of all households in the country.45
43 We identify with the final observation in the Levy and Foster report obtained by the Commission that the two functions remain distinct when assessing alternative sources of adequate media plurality, post a merger of the appellants’ businesses. David Levy and Robin Foster Impact of the Proposed NZME/Fairfax merger on media plurality in New Zealand: expert review of the Commerce Commission’s Draft Determination Document (16 November 2016, CC.FS.0649.0001). See [255] below.
44 CCFD, above n 1, at [1522], table 23.
45 At [604].
The appellants’ online forms of publication are similarly overwhelmingly dominant in the numbers of browser page views per month, and app page views per month. Because the measurement of resort to television and radio (views monthly versus listeners weekly) is not easily compared with the level of purchase of print forms of newspapers and extent of access to online sites, precise ranking of importance is difficult.
[102] The Commission also cited statistics gathered by Fairfax in a survey it undertook of the habits of New Zealand news consumers in 2016.46 The results of that survey indicated that, in a typical week in New Zealand in 2016, online sources accounted for approximately 54 per cent of the news consumed, television 21 per cent, traditional newspapers 13 per cent, radio nine per cent and other sources three per cent. The 54 per cent consuming news online was made up of 25 per cent from New Zealand news sites, nine per cent from international news sites, 15 per cent from Facebook, three per cent from other social media and two per cent from blogs or other commentaries.
[103] Both these measurements of news consumption reinforce the relative importance of news producers, and relegate the significance of collator/distributors. The second category are dependent on the first for their product, and the reach achieved is very modest by comparison.
[104] The Commission conducted interviews with some of the collator/distributors, and also with new businesses seeking to produce their own New Zealand news. Smaller online publishers of New Zealand news do not consider themselves competitors of the appellants, whereas they see the appellants as currently competing intensely with each other.47 Confidential information and opinions conveyed to the Commission justified the view that set-up costs are not the most significant barrier to entry. Nonetheless they do encounter difficulties securing investment capital. A greater barrier is developing trust as a brand of news provider. That trust is
fundamentally important to developing a readership, and cannot be achieved in a short
46 At [1520] and fig 12.
47 At [646] and [648].
or even medium term timeframe.48 In addition the financial viability of production and distribution of online news remains uncertain. Even well-established news producers in other formats are struggling to establish a viable business structure for their online presence.49
[105] In the year from the fourth quarter of 2015 to the third quarter of 2016 research commissioned for Fairfax revealed that the average combined reach of the appellants’ websites represented about 60 per cent of New Zealand’s population aged 10 years and over per month.50 In contrast the average monthly reach of the next largest, newshub.co.nz is 22.5 per cent with the equivalent figures for TVNZ’s news website at 14.3 per cent, and RNZ’s at 12.1 per cent.51 All of these measures reflect a position of substantial dominance for the merged entity, and on all measures they are inarguably each other’s strongest competitor.
[106] It is clear that in the production of national news the scale of established resources is important to the extent of coverage that can be achieved. Each of the appellants has substantially more extensive resources in terms of numbers and national coverage, than any of the other producers of news.52
[107] For these reasons, we treat the barriers to entry for new producers of New Zealand news to be significant. Although the barriers are lower for distributors, they are nonetheless material. We do not accept the appellants’ proposition that any loss of “voices” caused by a contraction of the number of views expressed by their businesses post a merger would be readily filled by “other voices”. Certainly, in specialised niches, small players may attract a following for views and commentaries on politics and other New Zealand news subjects, but those initiatives will not reach statistically significant portions of the consumers of New Zealand news.
[108] Whether viewed as a two to one merger within the part of news production originating with traditional newspapers, or a five to four contraction of the wider
48 At [725].
49 At [64], [643], [644], [719], 721], [722] and [732.1] (redacted in public version).
50 This comprised an aggregated 2.4 million unique visitors per month, with the individual figure for
Fairfax being 2.1 million, and NZME being 1.8 million.
51 At [673] and fig 2.
52 See for example, the statistics at [655] and [656], and confidential numbers at [657] and [741] .
online news production market including Newshub, TVNZ and RNZ, the contraction in such markets would be significant, and there would be significant barriers to entry by others.
Multi-sourcing/multi-homing
[109] A feature of the data about those accessing online sites for national news is the extent to which those doing so go to multiple sites or sources of online news. Economics literature on media markets describes this as “multi-homing” or “multi- sourcing”. Data compiled by a market research firm that was in evidence before the Commission showed that very high proportions of those who visited each of the appellants’ online sites at least once a month also visited the other site.53
[110] The appellants cited the extent of multi-homing demonstrated by these statistics as evidencing substitutability between all of the sites which would arguably apply as a constraint on the merged entity if it attempted to use any perceived market power to reduce quality. That data also showed that not insignificant numbers of those visiting these two major sites also visited other sites, at least once a month. For instance, 39.3 per cent of stuff.co.nz visitors visited newshub.co.nz, and 14.1 per cent visited radionz.co.nz. Of nzherald.co.nz visitors, 43.9 per cent also visited newshub.co.nz, and 13.8 per cent visited radionz.co.nz.
[111] Mr Goddard criticised the approach adopted by the Commission in assessing the evolving state of the online news market. He submitted that it was an error for the Commission to have adopted a static analysis of a freeze frame as matters stood at the time of its determination. He went further and characterised the analysis as backward looking when the Commission arguably ought to have projected forward to assess the conditions that were most likely to apply given continuation of present trends. On
Mr Goddard’s approach, the trend of increased resort to multi-sourcing and the growth in popularity of a range of online news sites should have led to a finding by the Commission that such sites would comprise materially stronger competition and
therefore be a stronger competitive constraint than is suggested by the historical data.
53 Spreadsheet CC.FS.0214.0001: 87.5 per cent of those accessing stuff.co.nz also accessed nzherald.co.nz and 73.4 per cent of those accessing nzherald.co.nz also accessed stuff.co.nz.
[112] The appellants criticised the Commission’s view that multi-sourcing indicated that different sites engendered different and complementary demand.54 The appellants submitted that this finding set the Commission apart from other competition authorities that have identified multi-homing as indicative of a lack of structural lock- in (that is, fixed patterns of resort by readers to a particular form of news media), and as negating any ability to exercise market power post-merger.
[113] Statistics recording at least a single monthly visit to a site provide a basic picture. We agree with the Commission that the pattern of visiting online sites is more likely than not to reflect the markedly more regular pattern of visits to the dominant sites, with visits likely to be for a greater length of time than the numerically smaller number of visits to the lesser visited sites.
[114] We do not accept that the practice of multi-homing can elevate the less visited sites into a position as meaningful competitors of the appellants. The statistics on monthly visits, viewed in light of the other evidence before the Commission of the difficulty of establishing a reputation for reliability, and the reduced scope of coverage provided by many of the less frequently visited sites, lessens their stature. Treating them as a competitive constraint cannot be justified merely because of evidence that numbers of readers practice “multi-homing”.
[115] The appellants’ arguments depended on an assumption that recent trends would continue to an extent that the appellants’ present dominance in terms of the number of visits to their online news sites would be substantially eroded in the foreseeable future. We accept that there is a prospect of the trend as described by Mr Goddard occurring. Circumstances confronting competition authorities in other jurisdictions may have progressed further with this trend than is apparent on the evidence in New Zealand. However a feature of disruptive technologies in the digital era is the unpredictability of the manner in which trends in behaviour adapting to new technology can change.
[116] Numerous uncertainties exist, and the prospect of others is quite likely. For instance there appears no consensus on how pervasive adblockers will become, or the
54 CCFD, above n 1, at [589.2].
impact of them on the manner of charging for digital advertising space.55 The projected ubiquity of Facebook may not develop consistently with recent history. Resistance to the intrusive extent of algorithmic analysis of viewers’ areas of interest and concerns over the extent to which it facilitates “fake news” raise the prospect that its development may not continue in an uncontrolled manner.56 Because of such uncertainties, we do not consider it appropriate to assess the extent of competitive constraints on an increase of the recent trend in matters such as multi-homing.
[117] Considering the proposed transaction as a five to four merger in the market for online New Zealand news, the patterns of usage suggest the appellants are the dominant participants in that market. Their position in the market is helped by the strength of their reputation. We consider there is scope for a reduction in quality to be imposed by the merged entity irrespective of the practice of multi-homing.
Prospect of a paywall
[118] If the operators of a digital news site wish to charge its readers to access the content they publish, it is done by imposing a paywall which requires readers to pay before obtaining access. Such arrangements can include “soft” paywalls where the obligation to pay is only triggered after usage above a certain level or to access certain restricted content.57 The viability of paywalls is another aspect of the evolution of digital news media that has attracted substantial debate and analysis. Before the Commission, and in the argument on the appeal, consideration of the prospects of a paywall involved reference to commercially sensitive information from the appellants. We are able to explain our own analysis of this point at a level of abstraction that
avoids any detailed disclosure.
55 An adblocker is an app (or a number of apps) developed to allow the user of a digital device to view websites in a way that blocks advertisements the operator of the site is projecting when a particular page is viewed. It was suggested that this can prevent that page view from counting towards the number of readers for which an advertiser pays.
56 Even an elementary internet search on the subject reveals a large volume of writings critical of their unbridled power. For example, Jonathan Taplin Move Fast and Break Things: How
Facebook, Google and Amazon Cornered Culture and Undermined Democracy (Little, Brown and
Co, Boston, 2017); and Christina Larson “Facebook Can’t Cope With the World It’s Created” (17
November 2017) FP < example The Washington Post, The New York Times, and The Australian. The Guardian is an example of a news media website continuing to offer all its content free to readers.
[332] We are satisfied that the appellants were afforded sufficient opportunity to adequately respond to the matters raised in the post-conference interviews.
[333] A specific criticism about the one-sided choices the Commission made in identifying those that it would interview post-conference is its failure to re-interview representatives of the Press Council. The appellants’ case was that the self-regulatory standards enforced by the Press Council would be effective in maintaining plurality. Some of the post-conference interviewees doubted the effectiveness of the Press Council to achieve this, and the appellants argued that a balanced process required the Commission to seek a rejoinder from the Press Council before accepting the opinions of those who doubted its ability to prevent a material loss of plurality.
[334] The Commission found that the Press Council (and internal codes of conduct) may ensure a level of fairness, balance and integrity, but that it could not influence editorial decision-making of the merging parties. The Commission referred to doubts expressed by the expert it had retained, Mr Foster, about the efficacy of internal plurality.
[335] In rejecting the appellants’ specific criticism, the Commission submitted that its view on the role of the Press Council had been signalled in the draft determination, which did not result in any further comment from the Press Council, although it did
submit a comment on a third party submission. The Commission also submitted that
141 CCFD, above n 1, at [653].
the appellants had not identified any material new information that might have been conveyed by the Press Council in a subsequent interview, had the Commission elected to do so.
[336] We consider those points are valid. We are not persuaded that the different views about the role of the Press Council received in post-conference interviews obliged the Commission to put such views to Press Council representatives before settling on a view as to the role it might play in maintenance of media plurality post- merger. Overall, we are not persuaded that the Commission’s post conference process was deficient.
Terms and timing of instructions to BDO
[337] The appellants advanced a number of criticisms of the process adopted by the
Commission in instructing BDO, and receiving a report from it.
[338] By mid-March 2017 the appellants had submitted three reports from PwC to the Commission in support of their applications. These comprised separate counterfactual forecasts for each of Fairfax and NZME provided on 25 November
2016, and a report on projected merger synergies provided on 17 March 2017. In addition to those reports, the appellants also provided, on 22 March 2017, an expert response from NERA to the Commission’s then thinking on the nature of the counterfactual that had been conveyed in a letter from the Commission to the appellants on 6 March 2017.
[339] At that point the Commission elected to engage BDO as independent accounting experts to comment on PwC’s merger synergies report and also on the earlier counterfactual reports. The Commission engaged BDO on 4 April 2017 on terms that the appellants complain were slanted in respects suggesting that the Commission was looking for views opposed to those of PwC.
[340] BDO provided its report to the Commission on 19 April 2017. That same day the Commission provided it to counsel for the appellants.
[341] After an inquiry from counsel for the appellants the Commission advised them on 21 April 2017 that it would consider any response to the BDO report provided it was received by 27 April 2017. The Commission also indicated that if that time frame was inadequate for a response to be prepared, the Commission would contemplate an extension of time. An extension for some few days was possible without disrupting the Commission’s timing for the issue of its determination, but the Commission indicated that if any extension for a longer period was required then that would involve deferral of the Commission’s decision until late June 2017. Counsel for the appellants confirmed that they did not want an extension.
[342] Some days before the 27 April 2017 deadline the appellants provided the Commission with critiques of the BDO report, one prepared by the appellants, and the other by PwC. The two complaints about the Commission’s process in relation to the BDO report that have subsequently been raised in the appeal were not raised at that time.
(i) Terms of instructions
[343] The terms of the Commission’s instructions to BDO were repeated in the firm’s
19 April 2017 report, and were accordingly conveyed to the appellants’ counsel when the report was copied to them. The specific questions posed by the Commission were in two parts, dealing first with the PwC counterfactual reports for NZME and Fairfax, and second with PwC’s Impact of Synergies letter. The instruction in relation to the first set of reports included the following:142
1. What are the shortcomings in the scenarios suggested in the PwC
counterfactual reports?
2. If the Commerce Commission can’t take PwC at face value, why not?
What are the gaps in the analysis? What else would the Commerce
Commission want to know?
…
4.Please highlight any other issues that you have identified with the counterfactual reports, e.g. potential inconsistencies with other documentation, etc.
142 BDO NZME/Fairfax Merger: Review of PwC Counterfactual and Synergies Submissions for
NZME Limited and Fairfax NZ Limited (19 April 2017, CC.FS.0930.0001) at 3–4.
[344] The instruction on PwC’s letter about the Impact of Synergies included the following:
1.The Commerce Commission are interested in whether the merger would necessarily prolong the lifespan of struggling … print publications … that are not profitable on a standalone basis. Specifically, can the Commerce Commission be satisfied that it is likely that they would use the cost savings from the merger to subsidise unprofitable parts of the business?
2. If not, why not?
3.[The] Commerce Commission are also interested in whether the merged entity would necessarily use merger-related cost savings to reinvest in the merged entity. Specifically, is there a real chance that such reinvestment would not occur, e.g. if merger savings were used to fund dividends to shareholders?
4.In light of these queries, please provide any other relevant comments you have regarding this letter, or specific paragraphs therein.
[345] The appellants criticised the closed form of the questions and terms of the instructions to BDO which they characterised as inviting criticisms of PwC and therefore not inviting a properly independent assessment. The deficiency in the terms of instruction was arguably compounded by the limited information provided by the Commission to BDO, and the relatively tight timeframe they were given to complete their report. Arguably the terms of the instructions suggested that the Commission wanted answers that found fault with PwC’s analyses which would pressure BDO to undertake its task in a way that produced the result the client wanted. Mr Butler submitted that the closed form of some of the questions risked BDO approaching their task with a less than fully open mind.
[346] In the first of two affidavits sworn in response to the process criticisms, Katie Rusbatch, the Competition Manager for the Commission, provided an explanation for the terms in which the instructions to BDO were cast. Those terms were settled by Commission staff and not the division members. The context as explained by
Ms Rusbatch was that the Commission had conducted a substantial analysis of submissions and expert reports. Preliminary views had been formed. The Commission did not necessarily disagree with the methodology that PwC had adopted in its counterfactual reports but was concerned about the validity of the underlying key assumptions behind PwC’s findings. The focus was accordingly on whether an
independent accountant confirmed the reasonableness of PwC’s underlying assumptions. Ms Rusbatch denied that the instructions were cast in terms seeking a report to support a particular view.
[347] If the Commission’s letter of instruction to BDO had been prepared by solicitors for a defendant in commercial litigation, inviting a critique of an expert report that was to be relied on by the plaintiff, then the lack of an open invitation to form views about the content of the work being critiqued would be relevant to challenging the objectivity and therefore the weight that might be given to the report provided in response. Instructions in those terms would certainly afford grounds for arguing that little or less weight should be given to a report produced in response to such instructions, compared with the weight that could be given to opinions that were sought in entirely open terms.
[348] However that was not the context in which the Commission sought an opinion from BDO in the latter stages of this inquiry. The Commission already had its own provisional views. It does not appear to have been intent on rejecting any particular component of the assumptions that PWC had worked on, where reliance on them was likely to influence the Commission’s ultimate analysis. In those circumstances there is not the same basis for criticising instructions in terms that focused BDO on the areas of concern to the Commission.
[349] It was reasonable for the Commission staff to adopt the view that if the instructions were cast in entirely open terms, it would have been more difficult for it to focus BDO on the matters on which it wanted their opinion. Time was relatively short. As independent accountants, if BDO agreed with the scenarios that PwC had suggested in their counterfactual reports, then it was open to them to respond that they did not identify material shortcomings. We are not satisfied that inadequacies in the terms of instructions provided to BDO ought to have required the Commission to reject BDO’s report provided in response to those instructions.
[350] Ultimately the determination cited the BDO report on only one point. This was to reject one aspect of PwC’s analysis as to [
].143 In adopting a likely outcome inconsistent with that in PwC’s November 2016 analysis, but which was supported by BDO, the determination also noted that PWC’s last contribution of
23 April 2017 acknowledged the prospect of that different scenario, consistently with that put to the Commission by BDO.144
[351] The Commission retained BDO because it wanted to test a number of the assumptions that PwC had relied on in that firm’s proposals as to the nature of the counterfactual. To the extent, if any, that the Commission adopted a different counterfactual from that PwC contended for (and without assessing what influence on that different definition of counterfactual that BDO’s report had) the Commission’s counterfactual has not been challenged on appeal.
[352] The practical upshot is that there is no material prejudice to the quality of the consideration given to all the submissions presented by the appellants that can be attributed to any deficiency in the terms in which BDO was instructed.
[353] The appellants advanced a related criticism of the Commission’s instructions to BDO, in that the Commission was selective in the information provided, and did not provide BDO with sufficient background and information to enable it to competently carry out the instructions the Commission gave it. BDO’s report noted a limitation on its work by virtue of the limited extent of information and also acknowledged the absence of opportunity to talk with PwC or the appellants’ counsel.
[354] This was not advanced as a criticism of BDO. Rather, that the Commission erred in considering that it could obtain a report that was based on adequately informed consideration, when it was as selective as it was in provision of information to BDO. The appellants did not advance any respects in which they contended that the views expressed by BDO were wrong because BDO were unaware of other material
information.
143 [ ].
144 At [170].
[355] Again, the materiality of such criticism depends on the extent of reliance ultimately placed by the Commission on BDO’s report. We are not persuaded that that level of reliance was such as to treat any inadequacy in the information provided to BDO as constituting material error in the Commission’s process. Similar considerations to those noted in [349] above apply to the choices confronting Commission staff when they decided to instruct BDO at the stage of the inquisitorial process when that occurred.
(ii) Limited time to respond
[356] The second criticism of the Commission’s handling of the BDO report was that it was provided to the appellants only seven working days prior to the publication of the determination. Arguably that was insufficient time for the appellants to provide a thorough response, and also too near the Commission’s decision on its final determination (“in a real politik sense”) for any response the appellants provided to be adequately taken into account.
[357] The Commission’s response to these criticisms is that the sequence of events demonstrated that the appellants’ advisers were sufficiently resourced to provide a response some three days before the deadline, that no complaint was made at the time indicating the time allowed was too short and indeed that the appellants rejected an invitation to seek either a short or longer extension of time. Further the adequacy of the rejoinder on behalf of the appellants tends to be confirmed by the absence of any new grounds for criticising the BDO report, beyond those raised in April 2017, as being material on the appeal.
[358] We agree that the Commission’s points are a sufficient answer to the complaint of inadequate time for the appellants to provide a response.
[359] The appellants’ related concern was that the timing of receipt of their reply to the BDO report left too little time between receipt of that response and completing its determination for the Commission to give meaningful consideration to the appellants’ criticisms of the BDO report. The Commission’s first response to this was to dispute any obligation to provide the BDO report to the appellants at all. That is not
persuasive. We consider it was appropriate for the Commission to provide the BDO
report, and invite the appellants to comment on it, as it did.
[360] Having afforded that opportunity and received comments on the BDO report, the Commission was obliged to consider the appellants’ response. The Commission submitted that the BDO report was in the nature of clarification about a matter on which the appellants had made submissions. The Commission is right to characterise the BDO report as part of its deliberative process and there is no evidence to suggest that the Commission had closed its mind to any new or compelling contrary arguments raised in the appellants’ response to the BDO report.
[361] We accept the reasonableness of this response for the Commission. It is a question of fact as to how important the BDO report was in a relative sense to the Commission’s determination, and how different the outcome would have been if the Commission was then persuaded to reject any part of the BDO report it was otherwise intending to rely on, because of the final submissions on behalf of the appellants. They did not draw our attention to any content of that final response which was different from matters previously put to the Commission and which was not acknowledged by the Commission but arguably ought to have been. We accordingly reject this aspect of the appellants’ process criticism.
Nature of any relief
[362] We therefore do not find any material errors of process committed by the Commission on the criticisms advanced for the appellants. Had we done so, we would have real reservations about relying on such findings to grant any measure of the type of relief requested by Mr Butler in his oral submissions. The appellants’ proposition was that if any of their criticisms were upheld, then the appropriate relief was for the Court to disregard the content of the BDO report (or, depending on the criticisms upheld, the post-conference interviews) and ensure that no weight was placed on those matters. Mr Butler described this as a nuanced evaluation that could be left with the Court as an open invitation.
[363] Our analysis of the extent of the appellants’ opportunity to respond to the post- conference interviews negates any prospect of excluding them from the evidence in
the appeal. The content of the BDO report does not figure materially in any aspect of our analysis of the merits of the appeal. Even if that had been the case, we would be disinclined to alter the substantive outcome by ignoring evidence that was before the Commission merely because of some inadequacy in the process by which that evidence was obtained by the Commission, or the opportunity the appellants were given to comment on it.
[364] We accordingly dismiss the process complaints, and reject the invitation that they should influence the evidence available on the appeal.
Costs
[365] The Commission is entitled to costs. Counsel for the Commission requested that, if the Court dismissed the process complaints, they be afforded an opportunity to make submissions in support of increased costs on this aspect of the appeal. Counsel for the appellants responded that, if their process complaints were entirely unsuccessful, the manner in which that part of the appeal had been pursued would not justify an increased award of costs in the Commission’s favour.
[366] My provisional view on costs for the whole appeal is that it warrants an award of costs on a 3C basis, with certification for second counsel throughout.
[367] I accept that the Commission was required to respond at short notice to a range of process criticisms that were then not pursued. This included preparation of affidavits, and some of the criticisms raised relatively serious matters. The responses would have taken significant resources of a time when their focus would otherwise have been on preparation of the Commission’s case on the substantive appeal.
[368] Had timing issues not been relevant, the appellants could well have been required to plead these process criticisms in a separate judicial review. The scale of work on allegations not proceeded with, and the discrete nature of the process criticisms that might equally have been advanced in a separate proceeding (and therefore result in a separate costs award on unsuccessful completion) do warrant an uplift. However, my provisional view is that there is no justification for its quantum to be increased to reflect criticism of the manner in which process criticisms were
advanced for the appellants. My provisional view is that a separate award for the
Commission’s successful opposition to the process complaints of approximately
$4,000 is likely to be warranted.
[369] I leave the parties to reflect, and hopefully confer, on an appropriate outcome on costs. If not agreed, the Commission may file a memorandum, not exceeding
10 pages, within 42 working days of delivery of this judgment. If such a memorandum is filed, the appellants are to respond, with the same limit as to length, within 10 working days of service of the Commission’s memorandum.
Summary
[370] We have dismissed the appeal against the Commission’s clearance decision. We have not found a likelihood of an SLC in the advertising market for Sunday newspapers, and dismiss the prospect of one of the appellants introducing a paywall for their online publication, post a merger. In other respects we come to the same conclusions as the Commission on the prospects of an SLC in the reader market for online national news, reader market for Sunday newspapers, and both advertising and reader markets for community newspapers in the 10 areas in the North Island where the appellants’ existing community newspapers compete.
[371] We have also dismissed the appeal against the Commission’s refusal to grant an authorisation for the proposed merger. We have upheld the jurisdiction of the Commission to consider detriments beyond economic or financial detriments applying in the market in which the Commission had found the likelihood of an SLC, and in particular for the Commission to take into account the material detriment arising from loss of media plurality. The Commission was also entitled to place significant weight on the prospect of reduced quality of the products produced by the merged entity.
[372] In the evaluation of public benefits and detriments, we have found that sufficient benefit to the public to warrant an authorisation cannot be made out.
[373] The appellants’ complaints of inadequate or improper process on the part of the Commission in the course of its work on its determination do not constitute criticisms that require us to assess the merits of the substantive appeal any differently.
It was legitimate for the Commission to select those with whom it conducted post- conference interviews, and whilst the terms of the Commission’s instructions to BDO as an external accounting expert might have been dealt with more felicitously, that aspect of the Commission’s process does not lead to a justiciable error.
[374] The Commission is entitled to costs.
Dobson J
Solicitors:
Russell McVeagh, Auckland for Appellants
Meredith Connell, Wellington for Respondent
4
2
0