Chorus Ltd v Commerce Commission
[2014] NZCA 440
•8 September 2014 at 8.30 am
| IN THE COURT OF APPEAL OF NEW ZEALAND |
| CA229/2014 [2014] NZCA 440 |
| BETWEEN | CHORUS LIMITED |
| COMMERCE COMMISSION First Respondent | |
| VODAFONE NEW ZEALAND LIMITED Second Respondent | |
| SPARK NEW ZEALAND LIMITED (FORMERLY TELECOM NEW ZEALAND LIMITED) Third Respondent | |
| ORCON LIMITED Fourth Respondent | |
| AND | OTHERS |
| Hearing: | 31 July and 1 August 2014 |
Court: | Ellen France, Stevens and White JJ |
Counsel: | D J Goddard QC, A M Peterson and T Smith for Appellant |
Judgment: | 8 September 2014 at 8.30 am |
JUDGMENT OF THE COURT
AThe questions raised in this appeal are answered as follows:
1. Did the Commerce Commission in fact fail to determine what inferences could reliably be drawn from the two benchmarks about the likely cost of providing the UBA service in New Zealand, and thereby err in law?
Answer: No.
2. Did the Commission in fact adopt the Swedish price as an upper band, and thereby err in law?
Answer: No.
3. Did the Commission in fact fail to consider s 18 other than in the context of price point selection, and thereby err in law?
Answer: No.
4. Did the Commission in fact fail to consider whether a price above $10.92 would be more consistent with s 18, and thereby err in law?
Answer: No.
5. Did the Commission fail to consider whether a price reduction of this magnitude was consistent with the s 18 purpose, having regard to the limited information available to it about the cost of providing the UBA service in New Zealand, and thereby err in law?
Answer: No.
6. If “yes” to any of the above, what remedy should be granted?
Answer: No response required.
BThe appeal is dismissed.
CThe appellant is to pay the costs of each of the first to fourth respondents for a complex appeal on a band A basis with usual disbursements to be fixed by the Registrar, including travel and accommodation expenses for out of town counsel. We certify for two counsel for the first respondent.
____________________________________________________________________
REASONS OF THE COURT
(Given by White J)
Table of Contents
Para No
Introduction [1]
Background [7]
Deregulation [8]
Regulation [11]
UBA prices [17]
The appeals [20]
Statutory framework [22]
The Commission [23]
Process [24]
Definitions [27]
Purpose [38]
Procedure [46]
Appeals [48]
Summary [49]
The Commission’s determination [51]
Discussion Paper [54]
Draft determination [56]
Chorus’s submissions on draft determination [64]
UBA conference [65]
Professor Vogelsang’s report [67]
Update Paper [72]
Chorus’s submissions on Update Paper [80]
The final determination [81]
Scope of appeal [107]
First question – Did the Commission in fact fail to determine
what inferences could reliably be drawn from the two benchmarks
about the likely cost of providing the UBA service in New Zealand,
and thereby err in law?
High Court judgment [115]
Submissions for Chorus [116]
Discussion [120]
Second question – Did the Commission in fact adopt the
Swedish price as an upper band, and thereby err in law?
High Court judgment [131]
Submissions for Chorus [133]
Discussion [135]
Third question – Did the Commission in fact fail to consider
s 18 other than in the context of price point selection, and thereby
err in law?
High Court judgment [147]
Submissions for Chorus [149]Discussion [151]
Fourth question – Did the Commission in fact fail to consider
whether a price above $10.92 would be more consistent with s 18,
and thereby err in law?
High Court judgment [160]
Submissions for Chorus [161]
Discussion [162]
Fifth question – Did the Commission fail to consider whether
a price reduction of this magnitude was consistent with the s 18
purpose, having regard to the limited information available to it
about the cost of providing the UBA service in New Zealand, and
thereby err in law?
High Court judgment [166]
Submissions for Chorus [167]
Discussion [168]
Sixth question – If “yes” to any of the above, what remedy should
be granted? [171]
Result [172]
Costs [174]
Introduction
When the third respondent, Spark New Zealand Ltd (formerly Telecom New Zealand Ltd) (Telecom), the provider of most of New Zealand’s wholesale and retail telecommunications infrastructure services, was split into two separate companies in 2011, the appellant, Chorus Ltd (Chorus), took over the wholesale services. Prices for some of Chorus’s core services are determined by the first respondent, the Commerce Commission (the Commission), under the provisions of the Telecommunication Act 2001 (the 2001 Act).
This case relates to the Commission’s determination of Chorus’s price for one component of the price that Chorus will be entitled to charge as from 1 December 2014 for its “unbundled bitstream access” (UBA) service, that is, its wholesale copper network service. Service providers, such as the other respondents, seek access to the UBA service in order to deliver retail broadband services to their customers.
Chorus exercised its right to appeal to the High Court against the Commission’s determination on questions of law. Six questions were identified.
Chorus was unsuccessful in the High Court.[1] It has obtained leave from that Court to appeal to this Court.[2]
[1]Chorus Ltd v Commerce Commission [2014] NZHC 690 [High Court judgment].
[2]Chorus Ltd v Commerce Commission HC Wellington CIV-2013-485-9923, 28 April 2014.
The appeal is opposed by the Commission and the other respondents who participated in the Commission’s determination process to set the disputed component of the price: the UBA additional component price (the UBA price). InternetNZ, ConsumerNZ and the Telecommunications Users Association of New Zealand (TUANZ) were also granted intervener status to oppose the appeal.[3]
[3]On the basis they are not entitled to costs.
Before addressing the six questions of law, we describe the background to the appeals, the relevant statutory framework, the Commission’s determination process and the scope of the appeal. We refer to the reasoning of the High Court Judge, Kós J, in the context of addressing the questions of law.
Background
The role of the Commission as the body responsible for the determination of prices for telecommunications services is a relatively new development in New Zealand.
Deregulation
Until the 1980s telecommunication services were provided by the state through the New Zealand Post Office which determined the prices for its services.[4] From 1987 the Government decided to restructure the Post Office into three autonomous state corporations.[5] The telecommunications section was the first to be corporatised (as Telecom Corporation of New Zealand Ltd) under the State-Owned Enterprises Act 1986.[6] Telecom then became a fully privatised company in 1989.
[4]Post Office Act 1959.
[5]Telecommunications Act 1987. A government-commissioned Post Office Review in 1986 suggested the inadequacies of this state-owned organisational entity and recommended that it should be re-organised into specific business units: RN Manson and MS Morrison “Post Office Review” (Government Printer, Wellington, 1986). Alan Bollard and David Mayes “Corporatisation and Privatisation in New Zealand” (paper prepared for conference on International Privatisation: Strategies and Practices, University of St Andrews, 1991) at Table 1.
[6]State-Owned Enterprises Act 1986, schs 1 and 2.
Telecom was initially a vertically integrated monopoly that owned the only universal telecommunications infrastructure in New Zealand and delivered the full range of retail services available.
As the New Zealand telecommunications market was opened up to competitive entry,[7] however, Telecom was recognised as having a dominant position in the market. This was recognised in Telecom’s “Kiwi Share” Obligations (KSOs), agreed between Telecom and the Government,[8] which obliged Telecom to offer free residential local calling and maintain a price cap on basic line rental service. The price and terms of access to Telecom’s networks were left to Telecom to determine subject to the constraints of the Commerce Act 1986 enacted to promote competition in markets for goods and services in New Zealand. Reliance on competition law rather than on industry-specific regulation as in other countries led to extensive litigation. The question whether Telecom alone should bear the cost of its unprofitable KSOs was contentious. Clear Communications, Telecom’s first competitor, sued Telecom in 1991 after failing to negotiate terms and conditions for access to Telecom’s Public Switched Telephone Network (PSTN) by means of an interconnection agreement.[9] The Commission was also involved in litigation against Telecom seeking to enforce the provisions of the Commerce Act.[10]
Regulation
[7]Telecommunications Amendment Act 1989.
[8]The Kiwi Share Obligations were recorded in cl 5 of the first schedule of Telecom’s Articles of Association.
[9]Clear Communications Ltd v Telecom Corporation of New Zealand Ltd (1992) 5 TCLR 166 (HC); rev’d (1993) 4 NZBLC 103,340 (CA); rev’d (in part): [1995] 1 NZLR 385 (PC).
[10]Telecom Corporation of New Zealand Ltd v Commerce Commission [1992] 3 NZLR 429 (CA).
Concern about the adequacy of reliance solely on competition law and Telecom’s ability to obstruct competition led to a Ministerial Inquiry into Telecommunications in 2000 (the Fletcher Inquiry) which recommended the introduction of a new regulatory pricing model for certain designated telecommunications services.[11] The Inquiry considered alternative means of establishing interconnection terms and conditions, and the pricing principles and processes applying to interconnection negotiations and KSOs. The Inquiry proposed a new suitably qualified Telecommunications Commissioner with responsibility for determining pricing for access by a provider to Telecom’s network on a two-stage basis: an initial determination in a short time frame followed, if necessary, by a final determination.[12] The Inquiry also recommended that international “benchmarking” based on OECD countries that regulate wholesale telecommunications services with a “TSLRIC type methodology”[13] and which did not have a state-owned incumbent be used for the initial determination.[14] Although new to New Zealand, the idea of using “benchmarking” for price determination purposes had already been adopted for other commodities in the United Kingdom and Australia.[15]
[11]Ministerial Inquiry into Telecommunications – Final Report [the Fletcher Inquiry] [4] and [6]–[8] and concluding at 104–108 and Ministerial Inquiry into Telecommunications – Draft Report [the Draft Report] at [4] and [6]–[7].
[12]At 71 referring to the Draft Report (see at [4]–[4.8]).
[13]Total Service Long Run Incremental Cost, referred to below at [29]–[30].
[14]At [7.2]–[7.3].
[15]Roger Carrington, Tim Coelli and Eric Groom “International Benchmarking for Monopoly Price Regulation: The Case of Australian Gas Distribution” (2002) 21 Journal of Regulatory Economics 191; Aoife Brophy Haney and Michael G Pollitt “International Benchmarking of Electricity Transmission by Regulators: Theory and Practice” (Electricity Policy Research Group Paper, Cambridge University, 2012); and Productivity Commission “Electricity Network Regulatory Frameworks” (Report 62, Canberra, 2013).
As Kós J pointed out and Mr Goddard QC for Chorus accepted, the following inferences may be drawn from the final report of the Fletcher Inquiry:[16]
(a)The initial determination might well be different from the pricing review determination if the latter was sought.
(b)Ideally, however, it would get close to the price expected on review, so that parties would not need to implement that step.
(c)Benchmarking (using OECD countries regulating the relevant designated service, using a TSLRIC-type methodology, and without a state-owned incumbent) would be used.
(d)The benchmarks would need adjustment for factors outside the incumbent’s control materially impacting cost of supply.
(e)Judgment would then be made by the regulator, on the basis of its best estimate of where New Zealand would fall if a full TSLRIC assessment were undertaken, evidently from within the benchmark range (as adjusted).
[16]High Court judgment, above n 1, at [27(a)–(e)].
Accepting most of the Fletcher Inquiry’s recommendations, the Government enacted the 2001 Act which:
(a)Created an industry-specific regulatory scheme, headed by a specialist Telecommunications Commissioner within the Commission.[17]
(b)Created an access regime under which the Commission could set terms, including the price, of access to Telecom’s wholesale services (“specified and designated services”) if commercial negotiations did not succeed. The terms were to be set for each access seeker separately, following a two-stage method.[18] The initial price of interconnection with Telecom’s fixed PSTN was presumptively to be determined by benchmarking against interconnection prices in comparable countries on a forward-looking cost-based pricing method.[19]
(c)Replaced the KSOs with Telecommunications Service Obligation (TSOs) by creating a statutory scheme for assessing the cost to Telecom for meeting its universal service obligations and how these costs were recovered across the industry.[20]
[17]Telecommunications Act 2001 [2001 Act], s 9 and sub-pt 2.
[18]Section 20 and sch 1, pt 2 (as enacted).
[19]Schedule 1, pt 2, sub-pt 1.
[20]Part 3.
As a result of further concerns about a lack of competition, a significant gap in pricing performance compared with other OECD countries and slow broadband uptake, the Government made further major modifications to the regulatory framework by the Telecommunications Amendment Act (No 2) 2006, under which:
(a)Telecom was to undergo operational separation, which required it to set up a separate network, wholesale and retail business units and create rules concerning how they would deal with each other and with competitors. Telecom was obliged to supply regulated products on an “equivalence of inputs” basis. This meant it had to use the same systems to deliver services internally and externally.[21]
[21]Telecommunications Amendment Act (No 2) 2006 [2006 Act], s 32 inserting new pts 2A and 2B into the 2001 Act.
(b)The range of services subject to regulation was extended. In particular, two regulated services were introduced that comprised the broadband service (now Chorus’s UBA service):[22]
(i)the Unbundled Copper Local Loop (UCLL) service, which allowed access seekers (service providers) to install their own electronic equipment in Telecom’s local copper loop and deliver voice and broadband services to customers using the copper network; and[23]
(ii)the additional UBA service component, which allowed access seekers to supply broadband services over Telecom’s copper access lines without investing in their own equipment or software.[24]
(The UBA service carries broadband traffic back and forth between consumers of broadband and service providers across Chorus’s copper loop, through local exchanges (digital subscriber line access multiplexer network devices (DSLAMs)) and to the first (the local) data switch. This connects to a “hand-over point” to Chorus’s network).
(c)A standard terms determination regime was introduced so that the Commission could determine the terms on which a designated access service or specified service must be supplied in relation to all access seekers and providers of the service.[25]
(d)The price for the UCLL service per line per month was to be, like other existing fixed network prices, on a forward-looking, cost-based approach.[26]
(e)The price of the UBA service per line per month was to be “imputed” by the Commission on a “retail‑minus” basis (calculable by reference to the end-user services supplied by Telecom).[27]
(f)The functions of the Commission were clarified. It was given extensive monitoring powers, including powers to conduct studies, and in particular “international benchmarking” in relation to its inquiries, reviews and studies.[28] Benchmarking principles applied in the determination of a number of prices of regulated services.[29]
[22]Section 56, amending sch 1 of the 2001 Act. The relationship of the two services is demonstrated by the Commission’s diagram, set out below [31] of the High Court decision.
[23]Schedule 1, cl 3.
[24]Schedule 1, cl 2.
[25]Section 13 inserting a new s 30A.
[26]Schedule 1, cl 3.
[27]Schedule 1, cl 2.
[28]Section 7.
[29]Schedule 1, amending sch 1 of the 2001 Act (see also that schedule) (including fixed public telephone service, unbundled bitstream access backhaul, Chorus’s unbundled copper local loop network (UCLL) and mobile termination).
Telecom was also required to replace part of its copper access network with fibre running to a cabinet near end‑user premises.[30]
[30]Ministry of Business, Innovation and Employment Review of the Telecommunications Act 2001 (Discussion Document) (Wellington, 2013) at [67].
In 2011 further amendments were made to the Act following Telecom’s decision to participate in the Government’s Ultrafast Broadband (UFB) initiative.[31] Under the Telecommunications (TSO, Broadband, and Other Matters) Amendment Act 2011:
(a)Telecom underwent structural separation into two separate businesses. Chorus was separated from Telecom.[32]
(b)Chorus was prohibited from participation in providing retail services, and entered into undertakings to provide wholesale services on a non-discriminatory basis.[33]
(c)The structural separation meant a retail-minus approach could no longer be used to determine the price for the UBA service, as Chorus’s revenue would be determined by Telecom’s pricing strategy. The 2011 Act specified that Chorus’s UBA price set in Telecom’s standard terms determination of 12 December 2007 was to continue to apply to existing lines until three years from the 30 November 2011 separation of Chorus and Telecom (1 December 2014).[34]
(d)Section 18(2A) was inserted, in particular connection with the UFB rollout, providing that consideration must be given to the incentives to innovate for and the risks faced by investors in new telecommunications services.
UBA prices
[31]Telecommunications (TSO, Broadband, and Other Matters) Amendment Bill 2010 (250-2) (select committee) at 1–2.
[32]Telecommunications (TSO, Broadband, and Other Matters) Amendment Act 2011 [2011 Act], pt 2.
[33]Section 51, inserting new pt 2A into the 2001 Act, including new sub-pt 3 (line of business restrictions).
[34]Section 79(2).
The UCLL price was reset by the Commission on 3 December 2012 at $23.52 per line per month.[35] The UBA price is currently $21.46 per line per month and will apply in respect of existing users until 1 December 2014.[36] The Commission was required to review that price, to apply from 1 December 2014.[37]
[35]Commerce Commission Final determination on the benchmarking review for the unbundled copper local loop service [2012] NZCC 37.
[36]The current price was set to implement sch 1, cl 4 of the 2011 Act setting a geographically averaged UBA price on 24 November 2011 (this came into effect on 1 December 2011): 2011 Act, s 73 and sch 1.
[37]2011 Act, s 77(1).
The new UBA price is now to be determined initially on a forward‑looking, cost-based approach, in line with other price determinations under the Act.[38] As with services already priced on this approach, setting this initial price involves international benchmarking.[39]
[38]Section 79(3).
[39]Schedule 3, sub-pt 1 of pt 2, amending schedule 1 of the 2001 Act.
In determining the new UBA price, the Commission applied the new pricing principle and decided that the new initial UBA component price should be $10.92 per month.[40] This was a very substantial reduction from the previous price of $21.46 and led to a sharp fall in the price for Chorus’s shares and the present challenge to the Commission’s decision.
The appeals
[40]Unbundled Bitstream Access Service Price Review Commerce Commission Decision [2013] NZCC 20.
Chorus appealed to the High Court against the Commission’s decision on the basis of the six identified questions of law.[41] It was unsuccessful and obtained leave to appeal to this Court.
[41]High Court judgment, above n 1.
When granting leave the High Court did not stipulate the questions of law for this Court’s determination, but we are satisfied the appeal is appropriately determined on the basis of the six questions the High Court determined. That approach is consistent with s 60 of the 2001 Act. Moreover, while there was some discussion during the hearing before us about altering the wording of some questions, ultimately the parties were content to deal with the matter on the basis of the existing six questions.
Statutory framework
We attach as an appendix to this judgment the various relevant provisions of the 2006 and 2011 amendments to the 2001 Act.
The Commission
The starting point is that under the Act it is the Commission, as the regulator, that makes the pricing determination. The Commission is a statutory body established under the Commerce Act.[42] It is a specialist tribunal whose members must be qualified for appointment, having regard to the functions of the Commission, by virtue of their knowledge of or experience in industry, commerce, economics, law, accountancy, public administration, or consumer affairs.[43] The specialist nature of the Commission is well-recognised.[44]
Process
[42]Commerce Act 1986, pt 1.
[43]Section 9(4)(a).
[44]Unison Networks Ltd v Commerce Commission [2007] NZSC 74, [2008] 1 NZLR 42 at [55]; Commerce Commission v Woolworths Ltd [2008] NZCA 276, (2008) 8 NZBLC 102,336 at [49]; Matt Sumpter with Ben Hamlin and James Mellsop New Zealand Competition Law and Policy (CCH, Auckland, 2010) at ¶104.
The Commission’s responsibility for determining Chorus’s UBA price arises under s 77 of the 2011 Amendment Act, which provides for a review of the existing standard terms, including price, under s 30R of the principal Act for the purpose of making those changes necessary to implement the new forward-looking cost-based benchmarking approach to UBA pricing. The Commission was required to make reasonable efforts to review the UBA service standard terms, including component price, by 1 December 2012 (one year from the separation of Telecom and Chorus).[45]
[45]2011 Act, ss 72 and 77.
A two-stage process was envisaged in respect of the price: first, determination on the basis of what is described as the initial pricing principle (the IPP) and, second, in the event of an application under s 78 of the 2011 Act by a party to that determination for a review of the IPP under s 42 of the principal Act, determination on the basis of what is described as the final pricing principle (the FPP). In the event of a review of the IPP, as has occurred here, the Commission is required by s 78(3) of the 2011 Act to make reasonable efforts to complete that review by 1 December 2014.
We were told that it is anticipated that the review will not be completed until April 2015 and that questions of backdating the FPP may arise.[46] It is not necessary for us to say any more about that issue here.
Definitions
[46]This Court held in Telecom New Zealand Ltd v Commerce Commission CA75/05, 25 May 2006 at [44] that as a matter of statutory interpretation a price review determination relates back to the date of the initial determination. It was suggested before us that the 2001 Act may create a power to backdate, but that backdating may not be mandatory.
Clearly the definitions of the IPP and the FPP are of critical importance. The IPP is defined as:[47]
The price for the designated access service entitled Chorus’s unbundled copper local loop network plus benchmarking additional costs incurred in providing the unbundled bitstream access service against prices in comparable countries that use a forward-looking cost-based pricing method.
[47]2001 Act, Sch 1, pt 2, sub-pt 1.
The FPP is defined as:[48]
The price for Chorus’s unbundled copper local loop network plus TSLRIC of additional costs incurred in providing the unbundled bitstream service.
[48]Ibid.
“TSLRIC” (Total Service Long Run Incremental Cost) is also defined. It:[49]
(a)means the forward-looking costs over the long run of the total quantity of the facilities and functions that are directly attributable to, or reasonably identifiable as incremental to, the service, taking into account the service provider's provision of other telecommunications services; and
(b)includes a reasonable allocation of forward-looking common costs.
[49]Sch 1, pt 1, sub-pt 1, cl 1.
The TSLRIC model provides an estimate of the costs of an efficient access provider over a sufficient period of time (long run), on a “forward-looking” basis (reflecting the notional costs to an operator if it built a new network) rather than of Chorus’s actual costs.
A number of features of the two definitions should be noted. Both contain two components. The first component in each case is the same, that is the price for Chorus’s unbundled copper local loop network (the UCLL price). That price is not in issue in this case.
The second components have differences and similarities:
(a)For the IPP, the second component involves determination of “additional costs” incurred in providing the UBA service by “benchmarking” against “prices in comparable countries that use a forward-looking cost-based pricing method”.
(b)For the FPP, the second component is the TSLRIC of “additional costs” incurred in providing the unbundled bitstream service.
The differences between the second components are important. While the IPP is concerned with ascertaining the notional additional costs for the UBA service in New Zealand by an international benchmarking process, the FPP is concerned with ascertaining the additional costs by the TSLRIC method. Reflecting the distinction between “initial” and “final” pricing principles and the statutory time constraints, it is accepted that the former should be a relatively straightforward, modest, exercise compared with the latter.
The similarities between the second components are also important. The reference in the IPP to comparable countries that “use a forward-looking cost-based pricing methodology” shows that countries that use a pricing methodology similar to TSLRIC are to be chosen. It is accepted this means that the IPP is to be a preliminary estimate, based on benchmarking, of the price that may, if necessary, ultimately be determined for the FPP. In other words, the IPP is intended as “a proxy” for the FPP.
The reference to “additional costs” is to a notional figure. As Chorus accepts, the second component is not concerned with its actual costs.
The crucial expression in the IPP is “benchmarking” which is not defined in the Act. Its ordinary meaning – “evaluate or check (something) by comparison with a standard (we are benchmarking our performance against external criteria)” – is applicable in this context.[50] As Chorus accepts, the “benchmarking” exercise involves the selection of appropriate “comparable countries”, the appropriate adjustment of their “prices” to reflect New Zealand conditions in order to determine the notional forward-looking “additional costs” (the UBA price) and the appropriate UBA service price (a single figure).
[50]Patrick Hanks and Judy Pearsall (eds) The New Oxford Dictionary of English (Reissue, Oxford University Press, Oxford, 2001) at 161; and John Simpson and others (eds) Oxford English dictionary (online ed) >
The dispute in this case relates to the determination by the Commission of the UBA price. Critically it concerns how the Commission went about the benchmarking exercise.
Purpose
In determining the UBA price, the Commission is required by s 19(c) of the Act to make the determination it considers “best gives, or is likely to give, effect to the purpose set out in s 18”.
The purpose set out in s 18 reads:
18 Purpose
(1)The purpose of this Part and Schedules 1 to 3 is to promote competition in telecommunications markets for the long-term benefit of end-users of telecommunications services within New Zealand by regulating, and providing for the regulation of, the supply of certain telecommunications services between service providers.
(2)In determining whether or not, or the extent to which, any act or omission will result, or will be likely to result, in competition in telecommunications markets for the long-term benefit of end-users of telecommunications services within New Zealand, the efficiencies that will result, or will be likely to result, from that act or omission must be considered.
(2A)To avoid doubt, in determining whether or not, or the extent to which, competition in telecommunications markets for the long-term benefit of end-users of telecommunications services within New Zealand is promoted, consideration must be given to the incentives to innovate that exist for, and the risks faced by, investors in new telecommunications services that involve significant capital investment and that offer capabilities not available from established services.
(3)Except as otherwise expressly provided, nothing in this Act limits the application of this section.
(4)Subsection (3) is for the avoidance of doubt.
When considering the application of s 18 in the context of a UBA service price determination, the Commission is expressly required to consider:[51]
… relativity between this service and Chorus’ unbundled copper local loop network service (to the extent that terms and conditions have been determined for that service).
[51]2001 Act. sch 1.
In this case the parties accept that the reference in s 18(1) to “the long-term benefit of end-users of telecommunications services” relates to consumers.
The parties also accept that s 18(2A) is particularly significant. It requires the Commission to give “consideration” to the incentives to innovate that exist for, and the risks faced by, investors in new telecommunications services that involve significant capital investment and that offer capabilities not available from established services. This means that, in making an IPP determination, the Commission should “err on the high side” in circumstances where it has no evidential basis to find it was more likely a notional New Zealand cost of service on a TSLRIC formula would be below rather than above the highest benchmark used by the Commission. At the same time Chorus accepts that s 18(2A) is a subset of s 18(1).
Chorus does not accept, however, that the Commission complied with its obligation under s 19 to take s 18(2A) into account at all material times. We address this issue later in the context of considering the third question of law. We note here that, as a matter of statutory interpretation, the various provisions relating to the determination of the UBA price in the IPP and the requirements of ss 18 and 19
should be read together to ensure that the legislation works in a realistic and practical manner.[52]
[52]Interpretation Act 1999, s 5, Northland Milk Vendors Assoc Inc v Northern Milk Ltd [1988] 1 NZLR 530 (CA) at 538; Commerce Commission v Fonterra Co-operative Group Ltd [2007] NZSC 36, [2007] 3 NZLR 767 at [22] and compare Ben Nevis Forestry Ventures v Commissioner of Inland Revenue [2008] NZSC 115, [2009] 2 NZLR 289 at [103]. See also JF Burrows and RI Carter Statute Law in New Zealand (4th ed, LexisNexis, Wellington, 2009) at 205.
It is also reasonable to assume, on the basis of the principle of statutory interpretation that the provisions of a statute are likely to be internally consistent,[53] that the statutory definition of the UBA price reflects the requirements of s 18, including in particular subs (2A) which was enacted at the same time. In other words, the mandatory requirement for the Commission to carry out the “benchmarking” exercise for the IPP by reference to appropriate “comparable countries” is itself designed to implement the statutory purpose, not to contradict or undermine it.
[53]Burrows and Carter, above n 52, at 237–242.
For completeness we note that there is no issue in this case as far as the requirement for the Commission to consider relativity with the UCLL price component is concerned.
Procedure
In making a determination of this nature, the Commission is also required to comply with the procedural requirements of s 53 of the 2001 Act which provides:
53 Procedure for determinations
For a determination made under this Part, the Commission—
(a)is not bound by technicalities, legal forms, or rules of evidence:
(b)may inform itself of any matter relevant to the determination in any way it thinks appropriate:
(c)must consider all submissions made in relation to the determination and all information and opinions presented or expressed at any conference or public hearing in relation to the determination.
The determination process follows a dispute resolution model.[54] The Commission is a tribunal with both investigative and adjudicative powers.[55] As with other tribunals, the Commission properly incorporates inquisitorial elements into its processes, although as an adjudicative body its procedures must to some degree be adversarial.[56]
Appeals
[54]Telecommunications Bill 2001 (124-2) (select committee report) at 9.
[55]Law Commission Tribunals in New Zealand (NZLC IP6, 2008) at [1.56].
[56]At [2.51]–[2.53] and Fletcher Report, above n 11, at [4.4.1].
Appeals under pt 2 of the Act (dealing with determinations by the Commission) are limited to questions of law.[57] A party wishing to challenge the merits of a Commission determination of an IPP may also seek a review under s 42. We address the question of the scope of this appeal later.[58]
Summary
[57]Section 60.
[58]Below at [107]–[114].
The relevant statutory provisions demonstrate Parliament has decided that:
(a)The pricing of telecommunication services is to be regulated.
(b)The regulator with principal responsibility for determining the prices is the Commission, a specialist body with the necessary expertise in economics and finance to carry out the exercise.
(c)Using its expertise, the Commission is to make a series of value judgments in the “benchmarking” exercise, as part of making the IPP determination, including the selection of the “comparable countries”, determining the notional “additional costs” and selecting the UBA price.
(d)As far as consideration of s 18(2A) is concerned, the Commission is to make a further value judgment when considering, as required by s 19, what “best gives, or is likely to give, effect” to the purpose. This language reinforces the Commission’s role as the arbiter of the value judgments involved under the Act.
(e)The Commission must operate under significant time constraints, especially in determining the IPP UBA price.
(f)The IPP should be determined expeditiously and no FPP process may be required.
(g)Reflecting the primary role of the Commission, there should be no appeal on the merits from a Commission determination. A limited appeal on questions of law only is permitted.
(h)Moving from a price based on a combination of forward-looking, cost-based pricing and Telecom’s actual retail costs to price components based entirely on the costs of an efficient provider was a step which inevitably had to result in a significant fall in the UBA price.
Against this statutory framework we turn now to the Commission’s determination in this case.
The Commission’s determination
Notwithstanding the statutory time constraints and the preliminary estimate nature of the IPP, the Commission, in accordance with the procedural requirements of s 53 and its usual practice,[59] took a number of steps before issuing its IPP UBA price determination on 5 November 2013. In particular, it:
[59]Unison Networks Ltd v Commerce Commission, above n 44, at [15]–[24].
(a)Prepared a Discussion Paper which was published on 26 July 2012.
(b)Sought and received submissions and cross-submissions from interested parties on the Discussion Paper between August and September 2012.
(c)Sought and received responses to its Wholesale Bitstream Access Questionnaire from overseas national regulatory authorities as a starting point for its benchmarking exercise in 2012.
(d)Collected information by internal research, ongoing correspondence with the overseas regulators and consultant advisers.
(e)Prepared a draft determination in December 2012 which indicated a basic UBA price of $8.93 per month.
(f)Sought and received submissions from interested parties on the draft determination in February and March 2013.
(g)Held a two day conference with the interested parties in June 2013.
(h)Commissioned an independent expert report from Professor Ingo Vogelsang, an economics professor at Boston University specialising in the regulation of network industries, which was received in July 2013.
(i)Issued an “Update Paper” in light of the submissions received and the conference on 13 August 2013 with Professor Vogelsang’s report attached.
(j)Sought and received submissions from interested parties on its Update Paper in September 2013.
Chorus, as an interested party, participated in this process by making submissions to the Commission on the Discussion Paper, draft determination and Update Paper, and by attending the two day conference. There is no criticism of the comprehensive nature of the consultative process adopted by the Commission in reaching its determination of the UBA price under the IPP process.
The Commission’s Discussion Paper, draft determination, Update Paper and final determination, as well as Professor Vogelsang’s report, are well-summarised in the High Court judgment.[60] We refer to their essential features.
Discussion Paper
[60]High Court judgment, above n 1, at [46]–[99].
The Discussion Paper described the Commission’s proposed approach to the benchmarking exercise. It involved identifying countries which set forward-looking cost-based prices for services (or service components) similar to the UBA service, whether the countries were comparable and the relevance of UCLL and its costs.
Chorus’s response to the Discussion Paper did not reject the overall approach, but emphasised:
(a)that, as part of the “UFB regulatory reset”, the Commission’s s 18 analysis should pay attention to the impact of any decision on capital‑intensive and risky new investments that deliver innovative new services, as s 18(2A) explicitly provided, and that a low UCLL price could put that investment at risk and delay migration to fibre;
(b)the need as part of the benchmarking process to define a benchmark set composed of countries where the cost of providing bitstream services is not significantly different to the cost of providing the UBA service in New Zealand and, to do so, to establish comparability criteria which reflect the important cost drivers for the UBA service; and
(c)the need to make appropriate adjustments to the benchmark prices which would take into account and offset any remaining differences between New Zealand and the countries in the benchmark set in the cost of service provision.
Draft determination
In its draft determination the Commission used three criteria to identify the comparable countries to be included in the benchmark:
(a)Forward-looking cost-based pricing method: only countries regulating bitstream prices on a forward-looking cost-based pricing methodology would be included.
(b)Comparable countries: the Commission said that “the country characteristics that are relevant to the cost of providing a wholesale bitstream service must be similar to those of New Zealand”.
(c)Service similarity: whether the benchmark services were “sufficiently similar to the UBA service”.
Then, on the basis of information obtained from a number of sources, including the responses to the questionnaire sent to the overseas regulators, the Commission identified 31 countries or possible benchmark candidates.
In applying the first criterion – forward-looking cost-based pricing method – to the 31 countries, the Commission used a four-step filtering process conveniently described by Kós J:
[53] The first criterion – forward-looking cost-based pricing method – was broken down further. The Commission used a four-step filtering process. The first filter was whether cost-based regulated pricing was used. That eliminated 20 of the 31 countries. The second filter was whether countries used a forward-looking TSLRIC methodology (or equivalent) in calculating regulatory price. That eliminated a further five countries: France, Spain, Bahrain and the United Kingdom [sic] used a fully distributed cost (FDC) methodology. The third filter was whether current costs were used in regulated pricing. Poland used historic cost accounting in determining its TSLRIC based price. Out it went, then. The fourth filter was whether the cost model was designed, or expressly reviewed and approved, by the regulator. Three countries (Switzerland, Greece and Slovakia) did not have regulator-verified models.
[54] That left three countries that could pass through the four filters: Belgium, Denmark and Sweden.
In applying the second criterion – country comparability – the Commission checked whether the countries identified had broadband market characteristics broadly similar to New Zealand. Focussing on Belgium, Denmark and Sweden and considering both the number of broadband subscriptions and the degree of penetration in the three countries, the Commission was satisfied that all three were suitable as benchmarks for setting the New Zealand UBA service price.
In applying the third criterion – service similarity – the Commission found that only Denmark and Sweden had broadly similar UBA services to New Zealand. Belgium was different because the physical structure of its bitstream service, specifically the location of the handover point, was different. In New Zealand the DSLAM and local exchange and the handover point are physically separate in the case of approximately 84 per cent of exchanges. Belgium’s handover point is, as Kós J recorded, co-located with the DSLAM. This difference creates a difference in costs.[61] A bitstream service that has a handover point located closer to the DSLAM is likely to have higher active network costs and lower transport costs. The Commission considered whether it could undertake adjustments to allow for the difference in handover point, but felt there was “no intuitive and transparent way of determining such an adjustment”. Accordingly, any adjustment would introduce additional uncertainty to the comparability of the Belgian service.
[61]At [57].
The Commission’s preliminary determination of the UBA price on the basis of the two member benchmark set of Denmark and Sweden is described by Kós J:[62]
[59] The draft determination then considered how to determine UBA price on the basis of that two member benchmark set. The Commission first undertook a currency conversion exercise. That produced a Danish UBA price (excluding UCLL) of $10.51, and a Swedish price of $7.36. The median (which is the same as the mean in a two sample set) was $8.93. (Had Belgium been included, its bitstream price would have been $6.76, and the mean would have been lower, $8.21.) The Commission said its favoured approach was to start with that two sample set median of $8.93, and then consider whether there were grounds to deviate from that point because of expected differences in cost, or to address s 18 concerns such as relativity.
[62]Footnote omitted.
The Commission’s consideration of s 18 in its draft determination is conveniently summarised by Kós J:
[60] The Commission then discussed s 18. It noted that UBA pricing would affect UCLL unbundling in the near term, and might also affect transition to fibre in the medium term. Should a higher price point than the mean benchmark UBA price be chosen to encourage investment? Its preliminary view was not. The Commission felt that an adjustment to that price was not required to maintain appropriate incentives for access seekers to invest in copper loop-based services. The Commission noted the specific matters required by ss 18 and 19, and the complexity of setting such a price to start in two years time. It sought submissions on relativity between the pricing of UBA and UCLL, implications for investments in those services, whether there were asymmetric economic costs in setting the UBA price too high or too low, the likely impact on incentives to invest in broadband services, whether over copper or fibre, and effects on end users.
[61] The Commission specifically noted s 18(2A) in its draft determination. It considered, tentatively, that the issue was the same for both UCLL and ultra-fast broadband (UFB): whether the UBA price encouraged investment and a take-up of competing services that would benefit end users. The Commission was unclear whether a UBA price higher than benchmark sample mean was likely to lead to investment in new innovative services, over copper or fibre, given access seekers would have an incentive to upgrade to fibre in order to differentiate their services from copper-based ones. The draft determination concluded, “our preliminary view is that the mean price point best gives effect to the interests of end users”. So that would be a UBA price of $8.93 per month.
Pausing at this point in the process, it is significant to note that:
(a)the Commission was proposing from the outset to carry out its identification of the appropriate comparable countries for the benchmarking exercise required by the IPP before considering s 18; but
(b)the requirements of s 18 were expressly considered before the UBA price was selected.
Chorus’s submissions on draft determination
Chorus contended in its submissions on the draft determination that:
(a)The draft decision did not support New Zealand’s opportunity to get higher quality broadband via UFB. Rather, it promoted “legacy services” (that is, its proposed UBA price would slow UFB uptake by incentivising end-users to stay with copper longer and would discourage international investment in Chorus and the transition to UFB). Chorus had not anticipated that “s 18(2A) would potentially be interpreted as having no effect”.
(b)Although the current legislative scheme, correctly interpreted, was consistent with an approach to UBA pricing that supported migration to UFB, the Ministry of Business Innovation and Employment scheduled reviews of the Act in 2013 and 2016 ought to update the Act’s framework to move from a ladder of investment and TSLRIC regime only appropriate in a copper environment and not appropriate in an environment of an efficient transition to fibre.
(c)The Commission’s draft UBA benchmarking was problematic because reliance on two countries was not robust, the Commission had not adjusted for New Zealand’s lower population density (and higher unit costs) and high broadband speeds (two important costs drivers) and had ignored the costs of installing roadside cabinets (moving broadband equipment closer to the homes of end-users, to improve performance).
(d)If the Commission’s approach did not change, Chorus would have to apply for an FPP review to protect its interests, provide fairer investment signals to the industry and seek a more coherent pricing framework.
UBA conference
The UBA conference held in June 2013 lasted two days. As Kós J points out, considerations of particular importance at the conference included:[63]
(a)the robustness of setting the IPP using a two country benchmark set;
(b)adjustments and amendments to provide additional robustness; the potential inclusion of other countries in the benchmark set through the relaxation of the benchmarking criteria used by the Commission;
(c)comparable service speed;
(d)the impact of fibre on the migration from copper and the adjustments proposed by Chorus to reflect the potential impact;
(e)the application of s 18 of the Act; and
(f)selecting a price point and the possibility that the price determined could fall outside the benchmark range.
[63]High Court judgment, above n 1, at [62].
At the conference Chorus, Telecom and Vodafone made submissions on how to derive greater certainty from the benchmarking process to be considered by the Commission in its Update Paper.
Professor Vogelsang’s report
Professor Vogelsang was retained by the Commission as an expert to provide an independent analysis of the key s 18 economic considerations involved in undertaking a price review of the UBA. He proceeded on the basis that the Commission first had to determine the feasible benchmark set and the resulting range of outcomes before addressing the s 18 issues. As the IPP determination was unlikely to produce the “true UBA cost” (that is the TSLRIC, measured under the FPP), the key question for Professor Vogelsang was: what were the effects of an increase or decrease in the UBA price compared to the “true UBA cost” in terms of the s 18 objectives?
Professor Vogelsang noted at the outset that it was clear from the evidence in the Commission’s draft determination and his own knowledge of other areas of the world that the true UBA cost established in a TSLRIC model would be “substantially below” the current UBA retail-minus price of $21.46 per month. He said:
I would have expected a ballpark figure of about NZ$10 per month, based on international data, certainly nothing in the range of the retail-minus price of $21.46.
As a number of submissions on the draft determination (supported by evidence) favoured a price higher than the mean/median of $8.93 identified in the draft, Professor Vogelsang proceeded on the assumption that the expected true TSLRIC of the UBA price for New Zealand would lie above that median. After noting the highly uncertain nature of adjustments necessary to account for different line densities, with density in Sweden slightly lower than in New Zealand but substantially higher than in Denmark, the Professor said:[64]
A clean (econometric) analysis for adjusting the benchmark results to reflect the New Zealand circumstances would be quite complex (including taking care of the adjustments in equipment and network architecture for different densities) and would, in my view, totally defeat the purpose of the IPP. It would, in my view, be out of proportion and would catapult the IPP both in terms of resource use and time requirement to the same level as the FPP. We therefore do not want to go further into the nitty-gritty of explaining or adjusting the benchmark figures for Denmark and Sweden. Nevertheless, an adjustment by the weight given to each of the two countries may be warranted. For example, Sweden may actually be closer in geographic properties to New Zealand than Denmark is.
In conclusion, if the benchmarking shows Denmark with a lower UBA cost than Sweden, but Sweden has lower UBA density than Denmark, then the cost differences between the two countries may well be explained by the differences in density. Because New Zealand’s density is very close to that of Sweden, the Swedish observation is probably much closer to the true expected value for New Zealand than the Danish observation. Consequently, a value at the 75% or even 100% mark between the benchmark costs of Denmark and Sweden appears to be justified.
[64]At para 76–77 of his report (footnote omitted).
As Kós J notes, Professor Vogelsang then undertook “an indicative error analysis” that showed that the probability distribution of the true cost would not necessarily fit in the interval spanned by the corrected Danish costs at the lower bound and Swedish costs at the upper bound.[65] The reference to “corrected” costs reflected the fact that since the draft determination it had become apparent that the Danish cost should be $8.88 and the Swedish cost $10.92 (mean/median $9.90).
[65]High Court judgment, above n 1, at [68]–[69].
After explaining his indicative error analysis, the Professor concluded:
The objectives of s 18 and 19 are fulfilled by a price certainly not below but possibly above true costs. In addition, the error analysis indicates that the price should be above expected costs. Furthermore, the clear value of Chorus’ UBA costs appears to be above the median between the UBA cost of Denmark and Sweden. Combining these factors justifies a UBA price with the measured UBA cost of Sweden. Such a price at the 100% mark would still be compatible with the requirement to stay within the benchmarking range. In my view, it could even exceed this value. Since objective measures are not available, these statements are based on subjective probability assessments.
Update Paper
The Update Paper, which attached Professor Vogelsang’s report and corrected the Danish and Swedish figures, presented its views and sought further submissions on the Professor’s report and three particular matters:
(a)The option of weighting individual benchmarks most comparable to New Zealand to derive the “most likely forward-looking costs of the UBA service”.
(b)A proposed approach to deriving a plausible range of the forward‑looking costs of the UBA service.
(c)The application of s 18 to the choice of a price point for the UBA service.
The Update Paper set out the background to the UBA service, the price review, current pricing, the changes since the current prices were set and the statutory framework, including s 18. The need to consider relativity with the UCLL price was also mentioned. The Commission noted that as the IPP price was intended to be a proxy for the FPP price and, as the FPP was based on TSLRIC, it preferred to benchmark only against countries that set their prices using TSLRIC.
The Commission recognised the scope for error with a small benchmark set and recorded that it was still considering whether to include additional countries as proposed by parties in their submissions. It also recognised that if it faced a small number of benchmarks two practical issues arose:
(a)the increased uncertainty under the IPP in determining the price that is “an unbiased estimator” (previously the median) of the forward‑looking costs of the UBA service in New Zealand; and
(b)the increased uncertainty under the IPP as to “the plausible range” of forward-looking costs of the UBA service in New Zealand, the plausible range representing the benchmark range within which the actual forward-looking costs of the UBA service can sit.
The Commission then explained why a small benchmark set increased uncertainty in determining price and noted that various options for addressing uncertainty in determining an unbiased estimator had been referred to in submissions received, including submissions from Chorus. The Commission said it was considering those submissions, but also sought further submissions on Professor Vogelsang’s option of weighting benchmarks where one of the benchmarks is believed to be more comparable than others.
The Commission recorded its preliminary view that the weight of evidence suggested the Swedish benchmark was closely comparable to New Zealand. It noted that Chorus’s adviser agreed.
Turning to the issue of determining the plausible range, the Commission explained how that would affect its decision and described the options for estimating the plausible range, including Professor Vogelsang’s advice as to the use of “probability distribution”.
The Commission then dealt with the questions of selecting a price point within a distribution and determining a distribution of a plausible range in the face of adjustments to the unbiased estimator. The Commission observed that, assuming Sweden was the more comparable country and its price was adopted as the unbiased estimator, then one end of the plausible range had to be higher than the unbiased estimator (the Swedish price). It was accordingly “not plausible that these costs have no possibility of being higher than that benchmark”.[66]
[66]High Court judgment, above n 1, at [78] and see discussion at [76]–[78].
Assuming adoption of a distribution for which the median was an unbiased estimate of the UBA cost, the Commission examined the issue of price point selection and, in some detail, the application of s 18. The examination included consideration of unbundling and competition to date, competition on the copper network post 1 December 2014, competition between copper and other telecommunication networks and dynamic efficiency. The Commission concluded:
143.Our current view is that taking account of dynamic efficiencies, a UBA price above the median will best promote competition for the [long term benefit of end‑users].
Chorus’s submissions on Update Paper
Chorus, in its submissions on the Update Paper, suggested:
(a)The best approach to estimating the forward-looking cost of the UBA service in New Zealand was to make adjustments to the price points so that the adjusted benchmarks were comparable to New Zealand (rather than placing greater weight on particular benchmarks).
(b)The best approach to generating the plausible range was to add more benchmark countries to the data set.
(c)There was a need to apply s 18 to the choice of price point for the UBA service. The economic costs of error in estimating the UBA price were asymmetric: the costs of under-estimating the price gave rise to greater economic costs than over-estimating. These factors supported a price point that was higher in the range than the median.
(d)The Commission’s approach to s 18 and s 18(2A) that s 18(2A) correctly required the Commission to take UFB into account and that greater weighting should be placed on long-term benefits and dynamic efficiency/investment incentives over static short-term price drops (if not the weighting to be given to s 18(2A)).
The final determination
As Kós J points out, the Commission’s final determination draws heavily on the preceding analysis and consultation process.[67] It is structured in four parts: the statutory framework; a discussion about the determination of the benchmark set in applying the new IPP; an outline of the price point selection for the UBA service and a cross-check to determine whether the price point was robust; and finally an explanation of how the monthly prices per line were determined.
[67]High Court judgment, above n 1, at [34].
In summarising the Commission’s determination, we focus on those parts which are relevant to the questions of law raised on appeal.
In the first section on the statutory framework, the Commission refers to all the relevant provisions and how it intends to interpret them. No issue is taken with its interpretation of the definition of the IPP and FPP or s 18.
In describing how the relevant provisions require it to choose a price for the UBA service within the plausible range from benchmarking, the Commission said:[68]
49.The Act requires that we determine the cost of the UBA service by benchmarking against prices in comparable countries. In practice, the benchmarking process provides a plausible range for setting the cost of providing the regulated service in New Zealand.
50.Once we have defined the range of plausible prices–which may extend outside the range of observed benchmarks–the IPP requires us to select a price point within that range. As set out in our update paper, our view, taking into account legal considerations, is that s 18 can affect our price point selection but cannot move us outside the process set by the IPP. In other words, we can only apply the s 18 purpose within the constraints of the benchmarking framework set out in the UBA IPP.
51.One party submitted that in selecting the price from the range defined by our benchmarking, we are limited to selecting the median. We do not agree that the statutory framework limits us in this way, and believe that our approach to price point selection (set out later in this decision) is consistent with the IPP and s 18.
52.We note that in our update paper we set out an approach that would potentially extend the plausible range beyond the observed benchmarks. A number of parties were critical of this approach.
53.The Commission remains of the view that inferring a larger plausible range is a conceptually valid exercise of the IPP, particularly where there are a small number of benchmarks. However, as we have not ultimately applied that approach in this determination, we have not responded to these aspects of submissions.
[68]Footnotes omitted.
Chorus argues that paragraphs 52 and 53 mean that the Commission decided not to consider a plausible range beyond the benchmarks and that as a consequence the Commission erroneously limited its approach. We address this argument when we consider the second question of law.
When referring to its interpretation and application of s 18(2A), the Commission expressly referred to aspects of the submissions for Chorus relating to implications for future network investment and the scope of the provision within the context of the benchmarking task required by the IPP which Chorus suggested required the Commission “to prioritise the successful migration to the UFB over short term price gains on the legacy copper network, where there is a conflict”. The Commission recorded its view that Chorus had overemphasised the scope of s 18(2A).
In the second section on the determination of the benchmark set within the IPP, the Commission referred to Chorus’s submissions relating to the addition of countries to the benchmark set and adjustments needed to ensure comparability. The Commission explained that:
85.Following our assessment, we have determined that our benchmark set will consist of the wholesale bitstream services from two countries–Denmark and Sweden. Our reasons for selecting these two countries are set out below.
86.In reaching our decision, we found that while Belgium, Greece and Switzerland met our criteria for forward-looking cost-based models, they did not meet our service characteristics criteria. Our decision is, therefore, to not include Belgium, Switzerland and Greece in the benchmark set. While we do not consider it appropriate to use these countries in making our final determination, we have used them as a cross-check to our decision.
The Commission’s detailed reasons for adopting Denmark and Sweden as the benchmark set are then given under the criteria which it had previously adopted in its draft determination: forward-looking, cost-based pricing methodology; service characteristics; and comparable countries.
The Commission reconsidered, in light of Chorus’s submissions, what countries should be included under the forward-looking cost-based criteria. It decided that in addition to Belgium, Denmark and Sweden, Greece and Switzerland should be added because it was satisfied that their regulators had sufficient involvement in review or validation of their prices.
The Commission reconsidered the service characteristics criterion, particularly the questions of the handover point and the weighted average speed price points. It decided that Belgium, Greece and Switzerland should not be included in the final benchmark, but at the same time observed they provided “considerable value as a cross-check to our two country core benchmark”.
The Commission reconsidered, in light of Chorus’s submissions, the comparable country criterion, particularly the questions of using line density as a cost driver and the impact of migration from copper. It decided not to adjust the benchmark countries for these factors because it considered they would increase the risk of error rather than improve the comparability of the benchmarks. The Commission noted that it would consider separately whether a price point adjustment for migration from the copper network was required.[69]
[69]See below at [103].
The Commission then set out its conclusions on the benchmark set:
195.Based on our analysis above, our view is that the most robust benchmark set is the two country benchmark set—Denmark and Sweden. Both countries meet our criteria for the IPP and our view is that they will provide the most robust estimate of the cost for the UBA service.
196.While a larger benchmark set is desirable, it potentially introduces error by including countries that are not directly comparable to New Zealand. In this case, we have decided to use an expanded benchmark set as a cross-check for the reasonableness of the core benchmark set of Denmark and Sweden.
197.The two country benchmark set is our preferred approach because there is no distortion from services that are not directly comparable to the UBA service. The approach is also based on countries most comparable to New Zealand, and our expectation is therefore that it more accurately reflects forward-looking costs for the UBA service.
198.We have rejected Chorus’ proposed adjustments for density and migration away from copper. Both adjustments require significant assumptions and manipulation of data. Our view is that given the uncertainty of the assumptions, we are uncertain that the adjustments provide more reliable benchmarks. In the next section we consider whether a price point adjustment is necessary for these factors.
199.The table below summarises the countries included in our final benchmark set and the countries used as a cross-check to determine the monthly rental for the UBA service in New Zealand.
Table 2: Countries included in final benchmark set and our cross-check
Country Comment
_____________________________________________________________
Denmark Included in final benchmark set
Sweden Included in final benchmark set
Belgium Used as a cross-check
Switzerland Used as a cross-check
Greece Used as a cross-check
_____________________________________________________________
The Commission introduced its third section on the price point selection for the basic UBA service as follows:
201.In the following sections we consider our price point selection within our two country benchmark set, and our cross-check using an expanded five country benchmark set. We then set out our decision on the price point for the Basic UBA service.
202.To determine our price point selection from within the two country benchmark set, we have started with the median of the benchmark set, and then considered whether we should move above or below the median due to any other additional factors. The factors that we consider relevant for this price review are:
202.1differences in comparability between the benchmark countries and New Zealand; and
202.2 section 18 considerations.
Chorus argues that the Commission’s statements that it was considering and determining its price point selection from “within our two country benchmark” meant that it had in fact adopted the Swedish price as an upper band and had thereby erred in law. We address this argument when we come to our consideration of the second question of law.
The Commission recorded that the monthly prices (in NZD) under the two country benchmark set were $8.88 (Denmark) and $10.92 (Sweden) with the median being $9.90.
The Commission then reconsidered whether a movement above the median price point was appropriate for differences in comparability. It decided that it was appropriate to adopt a weighting approach as a means to address the uncertainty of a small benchmark set.
After noting the view in the Update Paper that Sweden was the most comparable benchmark and the support from Professor Vogelsang’s report, the Commission referred to the submissions for the parties, including Chorus, and examined by way of a table measures of density and scale as comparability indicators for the UBA price.
The Commission described the table and its implications as follows:[70]
[70]Footnotes omitted.
212.Table 4 illustrates that urbanisation in New Zealand is comparable with the benchmark countries, while the difference in population density between New Zealand and Denmark appears to be substantial. However, DSL subscriptions and penetration in New Zealand is more comparable to Denmark. On balance, our view is that Sweden is more comparable to New Zealand than Denmark based on these indicators. However, we acknowledge that there are other cost drivers not observed in this table.
Table 4: Density measures for New Zealand, Denmark and Sweden
Characteristic New Zealand Denmark Sweden
____________________________________________________________
DSLAM sites 5,794 1,730 5,091
Population density 17 131 23
Urbanisation 86% 87% 85%
DSL penetration 27% 21% 15%
DSL subscriptions 1,169,014 1,190,954 1,470,000
____________________________________________________________
213.Given that New Zealand’s density factors appear more comparable to Sweden than Denmark, it is more likely that the forward-looking cost for the UBA service is closer to the Swedish bitstream service. This means we should select a price above the median, closer to Sweden.
214.However, while less comparable regarding population density, we do not consider that we should discount the Denmark price–we consider that price still relevant to our benchmarking purposes. Therefore, we do not believe that comparability factors alone would move us to the Swedish benchmark price. Doing so would give a 100% weighting to Sweden, and would discount the evidence provided by the Danish observation, effectively reducing the benchmark set to one.
215.We note that parties generally supported choosing a price closer to the Swedish price:
215.1Our expert advisor, Ingo Vogalsang [sic], indicated that a value at the 75% or even the 100% mark between the benchmark costs of Denmark and Sweden appears to be justified.
215.2NERA [Economic Consulting] [for Telecom, as it then was] indicated that a value at the 75% mark is reasonable.
215.3Covec [Ltd] [for Internet New Zealand, TUANZ and Consumer New Zealand] submitted that it is unlikely that Denmark provides no information about UBA costs in New Zealand.
215.4Network Strategies [for Vodafone] indicated that the price is likely to be between the median and the upper bound.
Chorus argues on appeal that these paragraphs support the error of law referred to in the first question of law. Mr Goddard submits that the second sentence in paragraph 213 is ambiguous, but whichever way it is read there is no evidential basis for saying that the density differences are likely to add less than $2.00 to the “Denmark price”. We address this submission when we come to our consideration of the first question of law.
As Kós J notes, the Commission then moved on to consider s 18 considerations.[71] In particular, it considered the asymmetric impact of error and the potential difference between IPP and FPP pricing. It re-endorsed the view expressed in the Update Paper and by Professor Vogelsang that a price above the median was “likely to ensure that s 18(2A) incentives are maintained”. The Commission put it this way:[72]
221.Our view remains that the negative impacts on competition of under-estimating the forward-looking costs are greater than over-estimating the forward-looking costs. This implies that we should err on the higher side to avoid the negative consequences of setting a price that is too low.
[71]High Court judgment, above n 1, at [94].
[72]Footnotes omitted.
After referring again to the submissions from the parties, including Chorus, the Commission repeated its view about “erring on the high side” when it said:
231.On balance, we accept in principle that the risk to dynamic efficiency of a low access price is asymmetric and that the balance of risk favours setting a price that errs on the high side. Consequently, we believe some adjustment is appropriate to take account of asymmetric risk.
232.We therefore consider that a price point above the median may be appropriate to minimise the risk to investment and to avoid dynamic efficiency losses that could arise from incorrectly setting a price below the forward-looking cost for the UBA service. This has been considered recognising that these benefits need to be balanced with the costs to end-users of raising the price above the median.
Chorus relies on these references to the need to “err on the high side” in support of its submissions on the fourth question of law.
As foreshadowed in its draft determination, the Commission again considered the possibility of adjusting the price point further to account for a potential drop in migration from copper to fibre.[73] It did so because Chorus and other parties had contended that migration to UFB should be a major consideration warranting selection of a higher price point. After examining the relevant considerations, the Commission concluded:
239.On balance, we do not consider a price point adjustment is appropriate to account for a potential drop in copper migration given that the effect of migration on the forward-looking costs of the UBA service is uncertain.
[73]Above at [62].
The Commission then concluded the third section of its determination as follows:
240.Having considered s 18, our view is that a price point at the top end of our two country benchmark set is appropriate to take into account the comparability of the benchmark countries and asymmetric cost.
241.This gives a monthly cost-based price for additional costs of providing the Basic UBA service of $10.92.
The Commission’s approach to the fourth section of its determination – the cross-check using an expanded benchmark set – is conveniently summarised by Kós J:[74]
[98] There was then a brief (five page) cross-check analysis using an expanded benchmark set. This brought back into account three further countries: Switzerland, Greece and Belgium. The pricing range for these countries (duly adjusted) ran from $6.56 per month (Belgium) to $11.45 per month (Switzerland). Taking this wider benchmark range into account, the Commission considered a 75th percentile price point appropriate (rather than the, in effect, 100th percentile chosen in the case of the two sample set). But the result was the same. The 75th percentile result was, again, the Swedish price (being one below the Swiss), at $10.92 a month.
[74]High Court judgment, above n 1, at [98].
The Commission then concluded:
268.Our conclusion is that the cost-based price for the additional cost of providing the Basic UBA service is $10.92 per month. We consider that that this price best gives, or is likely to best give, effect to the purpose of s 18 because:
268.1we systematically applied the IPP in order to reach our best estimate of the forward-looking cost-based UBA price for New Zealand;
268.2setting a forward-looking cost-based price promotes competition and promotes efficiencies in accordance with s 18; and
268.3we have taken into account s 18 including s 18(2A) in adjusting for the long term costs of potentially setting a price below the forward-looking costs of UBA.
Scope of appeal
This is an appeal under s 60(4) of the Act against the High Court judgment answering the six questions of law posed unfavourably to Chorus. It is not a general appeal against the merits of the Commission’s UBA price determination or an application for judicial review of the determination.
Moreover, Chorus does not challenge:
(a)Any aspect of the fairness of the consultative process the Commission followed in reaching its determination. Chorus accepts that it had ample opportunity to make submissions to the Commission during the process preceding the final determination and that in those submissions it raised the concerns which are now the subject of the questions of law.
(b)The Commission’s interpretation of any of the relevant statutory provisions. Chorus does not argue that in reaching its determination the Commission misinterpreted the definitions of IPP and FPP or the purpose provision (s 18). Chorus also accepts that it was open to the Commission, having regard to the purpose of the IPP as a proxy for the FPP, to decide to proceed on the basis that only countries that use a TSLRIC approach should be treated as relevant benchmarks.
Instead Chorus challenges the Commission’s determination on the basis that the proper application of the law required a different answer.[75] Chorus does this by alleging, in the first five questions of law, that the Commission made factual errors and thereby erred in law.
[75]Vodafone New Zealand Ltd v Telecom New Zealand Ltd [2011] NZSC 138, [2012] 3 NZLR 153 at [52], applying Bryson v Three Foot Six Ltd [2005] NZSC 34, [2005] 3 NZLR 721 at [24]–[27].
It is well-established, however, that to succeed on the basis of allegations of this nature Chorus must show that the Commission has exercised its judgment about the application of the IPP:[76]
… in a way that contradicts the true and only reasonable conclusion available on the facts and has thereby committed an error of law in terms of Edwards v Bairstow.
[76]Vodafone New Zealand Ltd v Telecom New Zealand Ltd, above n 75, at [58].
This is a high hurdle for Chorus to surmount. It is well-established that unless the Commission’s application of the statutory provisions is factually “unsupportable” it will not have erred in law.[77] It is for the Commission, as a specialist body, to exercise judgment in carrying out the requisite “benchmarking” exercise and in weighing up the relevant facts in that context.[78] It will therefore have erred only if there is no evidence to support the factual findings it made in reaching its determination.
[77]At [51].
[78]At [51] and [57].
In the absence of a right of general appeal, it is not the role of the Court in an appeal on a question of law to undertake a broad reappraisal of the Commission’s factual findings or the exercise of its evaluative judgments. Care should also be taken to avoid a technical and overly semantic analysis of the Commission’s determination in an endeavour to create a question of law. In making factual findings it is for the Commission, and not the Court, to decide what weight should be given to the relevant evidence and what inferences, if any, should be drawn from the evidence. An inference must be logically drawn from proven facts and not be mere speculation or guesswork.[79] At the same time, as counsel for the Commission acknowledged, if the Commission has made a factual error that makes its application of the statutory provisions “unsupportable” it will have erred in law.
[79]R v Puttick (1985) 1 CRNZ 644 (CA) at 647; Turners & Growers Ltd v Zespri Group Ltd (2011) 13 TCLR 286 (HC) at [97]; Caswell v Powell Duffryn Associated Collieries Ltd [1940] AC 152 (HL) at 169–170 per Lord Wright; and R v Kinghorn [2014] NZCA 168 at [20].
We now turn to address the six questions of law, but not in the same sequence as they were addressed in the High Court where question one was question three. In the High Court Kós J described the argument on what is now question 1 as “very much a tertiary argument to the two primary arguments”.[80] The two primary arguments related to what are now questions 2 and 3 and “dominated” the High Court appeal.[81]
[80]High Court judgment, above n 1, at [177].
[81]At [9].
The sequence of questions has been changed because at the hearing of the appeal in this Court Mr Goddard made it clear that what is now question one was to be treated as the primary submission for Chorus. We proceed on that basis.
First question – Did the Commission in fact fail to determine what inferences could reliably be drawn from the two benchmarks about the likely cost of providing the UBA service in New Zealand, and thereby err in law?
High Court judgment
Although this question assumed greater significance on appeal than it had in the High Court, it is still helpful to refer to the reasons of Kós J for answering this question “No”:
[153] The benchmark evidence was what it was. The benchmarks were based on overseas TSLRIC-derived prices only. They involved comparable country contexts. They involved comparable services. That is what the IPP required, both expressly and implicitly. They therefore provided reliable evidence, to the standard required by the Act. Section 18 was then applied as an overlay by the Commission. An approach I have found to be lawful. That added (as a result of Professor Vogelsang’s analysis) asymmetry to the analysis, and a recognition that the benchmark range and the plausible range were not necessarily (or in fact) co-extensive.
[154] In my view, if the Commission had reason to believe either:
(a)that the benchmark evidence was not reliable, probative evidence; or
(b)the IPP outcome (based on the benchmark evidence, and allowing for consideration of s 18) was nonetheless irrational or likely to produce an outcome at a substantial remove from the likely TSLRIC found under FPP:–
then the Commission would have had a duty to inquire further (and, possibly, re-consult).
[155] However, that was not the case here. First, the $10.92 figure adopted by the Commission is, on any view, within the plausible range for the IPP. Secondly, Professor Vogelsang’s independent advice to the Commission was, as I have set out at [66], that with all his relevant expertise he would have expected a “ballpark figure” of about $10 per month. Certainly nothing like the prior retail-minus price of $21.46. Thirdly, the cross-check analysis undertaken by the Commission with a wider range of countries, drawing back in Belgium, Greece and Switzerland which had been excluded because of their significantly different handover location points, produced a broader benchmark range, running from $6.56 to $11.45 per month. None of this information should have suggested to the Commission that the two sample set was unreliable, or that the $10.92 figure it adopted was irrational or likely to be at a substantial remove from the likely TSLRIC established under FPP.
[156] Beyond this, the mechanism to correct the IPP lay not in further protracted analysis to produce a more perfect IPP. It lay in the statutory mechanism, under s 42, to obtain a full pricing review using FPP.
Submissions for Chorus
Chorus contends that the Commission in fact failed to determine what inferences could reliably be drawn from the two benchmarks about the likely cost of providing the UBA service in New Zealand. In support of this contention, Mr Goddard submits that:
(a)The definition of the IPP required the Commission to determine the “additional costs” in New Zealand on the basis of benchmarked “prices” in comparable countries that use the requisite “forward-looking cost-based pricing method”.
(b)The Commission is not concerned with Chorus’s actual “costs”, but with the “costs” of a notionally efficient operation in New Zealand and, without more, the benchmarks are not evidence of costs in New Zealand.
(c)If the likely cost differences between New Zealand and the comparable country benchmarks cannot be reliably quantified on the basis of the available information, the raw benchmark figures will not be reliable guides and the target may lie outside the observed range. There is a “penumbra” of uncertainty.
(d)To determine “costs” in New Zealand will therefore require adjustments to be made to the benchmark costs, especially if there is only a single benchmark as in this case (Sweden).
(e)The absence of reliable evidence about the nature and scope of the differences in costs between New Zealand on the one hand, and Sweden on the other hand, does not imply that there are no such differences.
(f)The Commission could not, consistent with s 18 and the requirement that there be some reliable basis for its findings, proceed on the basis that the New Zealand costs were likely to be no greater than the Swedish benchmark price. Put simply, an assumption has been made that the cost in New Zealand is no greater than $10.92 without any supporting reasoning or evidence at all. With the “penumbra” of uncertainty, the Commission did not know enough about the cost drivers to say that it should not go above $10.92.
(g)The Commission’s reasoning was inherently contradictory because it identified reasons to think that the cost of providing the UBA service in New Zealand might well be in excess of the cost of providing the same service in Sweden, but failed to take that into account.
As already mentioned, Mr Goddard relies in particular on the information relating to density measures in paragraphs 212–215 and Table 4 of the Commission’s final determination and the statement in paragraph 213 that “we should select a price above the median, closer to Sweden”.[82] He submits that this statement could be read in two ways:
(a)a cost between the $9.90 median and the Swedish cost of $10.92 should be selected; or
(b)as Kós J held, as the plausible range extended beyond Sweden, the cost could be above Sweden, but the analysis supported the range between the median and the Swedish cost.[83]
[82]Above at [98]–[99].
[83]High Court judgment, above n 1, at [114].
Mr Goddard suggests that a price “above the median, closer to Sweden” could be above $10.92. On this basis he submits that there is no evidential basis for the Commission to limit its inquiry to a figure between $9.90 and $10.92. To do so involves an error of logic.
Mr Goddard also suggests that the Commission has identified a logical basis for thinking New Zealand costs will be above Denmark’s, but has no evidence for determining by how much. There is no evidential basis for saying the difference will be more or less than $2.00, being the approximate difference between Denmark and Sweden. If the Commission did not know what the difference was, it could not infer from the figure for Denmark that the best estimate lay between $9.90 and $10.92.
Discussion
The starting point is to recognise that the IPP definition requires the Commission to carry out the benchmarking exercise by identifying comparable countries that use a forward-looking cost-based pricing method. Parliament has decided that a benchmarking exercise must be carried out regardless of the number of comparable countries that meet the requirements of the definition. Once the Commission has identified the comparable countries, evidence of their prices will constitute the relevant evidence for the benchmarking exercise.
Here, the Commission went through an extensive exercise to make sure the benchmark set was truly comparable.[84] The exercise of winnowing out from the benchmark set those countries with differences that might affect the reliability of the comparison with New Zealand could give the Commission some comfort about the probative value of the benchmark set. As Kós J correctly found, there is no reason to believe that the benchmark evidence that the Commission obtained through its questionnaire was not reliable, probative evidence.[85] This evidence established not only that Denmark and Sweden with their TSLRIC pricing were the appropriate comparable countries but also that their benchmark costs were $8.88 and $10.92 respectively, with a mathematically calculated median of $9.90.
[84]See above at [56]–[60].
[85]At [154]–[155].
As Mr Farmer QC submits for the Commission, the benchmarking exercise is designed to set the price for the UBA service, not to ask whether New Zealand costs are likely to be above or below particular benchmarks or to adjust for all possible differences. An “evaluative exercise” of that nature would be problematic since the New Zealand TSLRIC costs are unknown and the purpose of the IPP is to avoid the modelling exercise if possible. At the same time, however, an evaluative exercise of a different nature is involved because, as Mr Farmer accepted, the Commission does have to form a “broad view” as to where New Zealand is placed vis-à-vis the two benchmarks “without being overly scientific” about it.
The next phase of the exercise, selection of the price point for the UBA service, involved an evaluative judgment on the part of the Commission. To that exercise, as Mr Wigley for Orcon and the interveners submits, the Commission brought to bear its expertise in areas such as statistics and pricing.
A “penumbra” of uncertainty was inevitable because the Commission was required to select a notional, rather an actual price, based on its benchmarking exercise. As Ms Fitzgerald for Spark submits, the existence of a level of uncertainty did not mean that the Commission was required to go beyond benchmarking on prices. Moreover, the Commission formed its own view on the reliability of the benchmark set having assessed the submissions from Chorus about the impact of differences in spatial density.
In our view the Commission proceeded correctly in making an evaluative judgment and selecting the price point for the UBA service by taking into account the two factors mentioned in paragraph 202 of its final determination:
(a)differences in comparability between the benchmark countries and New Zealand; and
(b)s 18 considerations.[86]
[86]The s 18 considerations are addressed below in the context of our responses to the third, fourth and fifth questions of law.
The Commission considered carefully the differences in comparability between the benchmark countries and New Zealand. The final determination is replete with this comparative analysis. In doing so, the Commission was entitled to take into account as relevant cost drivers the density measures set out in Table 4 of its final determination. This information constituted further relevant evidence for the Commission to evaluate and, in doing so, it did not proceed without evidence or in a contradictory or illogical manner.
The Commission did not find that costs in Denmark and Sweden were likely to be understated relative to New Zealand. Rather, it found that New Zealand was more comparable to Denmark in terms of digital subscriber line (DSL) (high-speed internet connection) subscriptions and penetration, while being more comparable to Sweden in terms of population density. It concluded that “on balance” New Zealand was more comparable to Sweden and accordingly gave it greater weight than Denmark. This did not mean that Denmark was to be ignored. On the contrary, it was to be taken into account in justifying the decision not to go above Sweden. This means that the Commission could be satisfied that Sweden’s price was the appropriate one for New Zealand.
When the Commission’s approach to the evaluative judgments it was required to make in selecting the price point for the UBA service is viewed in this way, it cannot be said that the Commission erred in the manner suggested by Chorus in the first question of law. In particular, we do not accept that the Commission’s determination was made without any proper evidential basis or on the basis of erroneous assumptions or inferences. Nor do we accept that the Commission exercised its judgment in a way that contradicts the true and only reasonable conclusion available on the facts or was factually unsupportable.
Chorus did not suggest any other information or cost drivers that ought to have been taken into account. Indeed Mr Goddard accepted that if the Commission had selected a figure a little above $10.92 it could have had no complaint. As Ms Fitzgerald submits, this confirms that $10.92 was within the plausible range.
Like Kós J, we therefore answer the first question of law – “No”.
Second question – Did the Commission in fact adopt the Swedish price as an upper band, and thereby err in law?
High Court judgment
Kós J found that the Commission did not in fact adopt the Swedish price as an upper bound because:
(a)It was improbable that the Commission, having correctly directed itself that a plausible range lay outside the observed benchmarks, then forgot those directions and in effect “bookended” itself.[87]
(b)The language in paragraphs 201 and 202 of the Commission’s final determination was unfortunate, but those passages had to be read in their whole context. Paragraphs 50 and 51 could hardly be clearer and it was more instructive to look at what the Commission did between paragraphs 201 and 241.[88]
(c)Having gone through the process in those paragraphs, the Commission finally lands on a number, the Swedish price of $10.92. The Commission’s judgment was that a price point at the top end of the benchmark range was appropriate to take into account both comparability and asymmetric costs (s 18 considerations). Nothing in the process suggested that the Commission reached that number because it was the highest number in the plausible range.[89]
(d)The fact that the Commission reached a number at the top end of the benchmark range strongly suggests that it must have regarded the plausible range as potentially broader. That was the very point it made in the Update Paper.[90]
(e)Price point selection at the top end of the plausible range would be unlikely to be statistically justifiable unless s 18 had profoundly skewed the benchmarking exercise. Yet here Chorus says s 18 did not skew it enough. Had the Commission considered that s 18 required it to select a price at the top of the plausible range, two things would have happened. First, the Commission would have defined the upper end of the plausible range. Secondly, it would have said that that is where s 18 took it. It did neither.[91]
[87]High Court judgment, above n 1, at [109]–[110].
[88]At [76]–[78] and [111]–[115].
[89]At [116].
[90]At [117].
[91]At [118].
Kós J concluded:
[119] The Commission has exercised a judgment, as it is required to. It has done so following a process probably more extensive than Parliament had in mind. It certainly took longer than Parliament had in mind. It is worth reminding ourselves that the Fletcher Inquiry, the progenitor of this IPP pricing process, had in mind that New Zealand could simply piggy-back international evidence, and that the result would be a “quick and cheap” process compared to the FPP cost-based modelling approach. In directing itself to that evidence, I do not find it to have limited the plausible range to the benchmark range. In other words, it did not “bookend” itself, confining the potential UBA price to the benchmark range.
Submissions for Chorus
Chorus contends that the Commission erred in law by adopting the Swedish price as an upper band. In support of this contention, Mr Goddard submits that the Commission had narrowed the range of the inquiry to the benchmark set of Denmark and Sweden. In other words, the Commission had erroneously concluded that the plausible range for the cost of the UBA service was capped or “bookended” by the Swedish price of $10.92. As already mentioned, Mr Goddard relies in particular on the Commission’s statements in paragraphs 52 and 53 and paragraphs 201 and 202 of the final determination.[92]
[92]Above at [85] and [94].
Mr Goddard also submits that Kós J erred in “charitably” not accepting that the Commission in fact adopted the Swedish price as the upper bound due to a failure to bear in mind its earlier directions. He submits that Chorus does not argue that the Commission forgot anything, but rather that the language of specific paragraphs in the Commission’s decision show it adopted a “weighting approach” to its analysis which could only ever result in a price in the range between the two benchmarks.
Discussion
We agree with Kós J that the Commission did not in fact constrain itself by the benchmark set. It is clear from the process it followed, including in particular Professor Vogelsang’s report and the Update Paper, and its final determination, read in light of that process and as a whole, that the Commission recognised that the plausible range extended beyond the benchmark range. It is no doubt true, as Mr Goddard submits, that when the Commission came to its final determination its workings proceeded on the basis of the outcome it knew it had reached, but this does not mean that the Commission had fettered its decision-making. On the contrary, the Commission recognised that, conceptually, it could go outside the range of the benchmark set, but on the basis of the material it had there was no reason to do so.
It is important to read the specific paragraphs in the final determination, relied on by Mr Goddard, carefully and in the context of the process followed as well as the final determination itself.
Adopting this approach in the case of paragraphs 52 and 53, we are satisfied for the following reasons that the Commission did not indicate that the cost of the UBA service was capped by the Swedish price of $10.92:
(a)It is clear from paragraphs 49 and 50 that the Commission recognised not only that in practice the “benchmarking process” provided a plausible range but also that the range of plausible prices might extend “outside the range” of observed benchmarks.
(b)It is equally clear from paragraphs 52 and 53 that all the Commission was saying was that it had not responded to submissions critical of “an approach” that would potentially extend the plausible range beyond the observed benchmarks because it had not ultimately applied “that approach” to its determination. This did not mean that it had rejected the approach as erroneous.
(c)We agree with Kós J that when these paragraphs are read together the Commission was making it clear that it did not regard itself as limited to choosing the median of the benchmark range.[93] It remained of the view that inferring a larger plausible range was conceptually valid under the IPP, particularly where there was a small number of benchmarks.
[93]High Court judgment, above n 1, at [86].
Similarly, adopting this approach in the case of paragraphs 201 and 202, we are satisfied for the following reasons that the Commission did not decide that its determination of the UBA price was “bookended” by the Swedish price.
First, while the Commission has stated in these paragraphs, which appear in the third section of its determination, that it is considering and determining its price point selection from “within” the two country benchmark set, it has not suggested that by doing so it has disregarded the views previously expressed in the second section of its determination about the plausible range extending beyond the benchmark range. Indeed in paragraph 201 the Commission also refers expressly to its “cross-check using an expanded five country benchmark set” which recognises the relevance of the wider plausible range.
Second, read in the context of the whole determination, the statements in these paragraphs should not be interpreted as an indication that the Commission intended to depart from the approach that it adopted in the second section of its determination and then proceeded to apply in the fourth section when it carried out its cross‑check using the expanded benchmark set.
Third, in section three of its determination the Commission considered the possibility of adjusting the price point further to account for a potential drop in migration from copper to fibre.[94] It did so because Chorus and others had claimed that migration to fibre warranted selection of a price figure higher than $10.92. An increase of some $2.75 was mentioned before us. The Commission declined to make the adjustment because of uncertainty about the effect of migration, not because the increase would have taken the price point beyond the Swedish price.
[94]Above at [103].
Fourth, as Mr Farmer submits, if the evidence or application of s 18 had led the Commission to consider that a price above $10.92 was the appropriate price point, it is clear from the determination that the Commission had not constrained itself from considering such an outcome. On the contrary, if the cross-check in the further section of the determination had come out at a materially higher (or lower) number, the Commission would have given further consideration to the reliability of the $10.92 price point and might have determined a price outside the two country benchmark range.
Fifth, again as Mr Farmer submits, Chorus did not point to any evidence or arguments that establish that a higher price than the Swedish price was appropriate. The issues it raises on appeal relating to the density measures, the existence of uncertainty and s 18 were all fully considered and addressed by the Commission following extensive input from the parties at various stages of the determination process.
Finally, the Commission had put forward the idea in its Update paper that the plausible range of comparable countries might be expanded, but this was rejected by submitters, including Chorus. Instead Chorus suggested adding various countries and making various data adjustments. In its final determination the Commission considered these suggestions and then did its five-country cross-check.
Accordingly, we do not accept that the Commission exercised its judgment on this issue in a way that contradicts the true and only reasonable conclusion available on the facts or was factually unsupportable.
Like Kós J, we therefore answer the second question of law – “No”.
Third question – Did the Commission in fact fail to consider s 18 other than in the context of price point selection, and thereby err in law?
High Court judgment
Kós J found that the Commission did not in fact fail to consider s 18 appropriately because:
(a)The Commission was not required to take s 18 into account when identifying as the first step in its analysis the comparable countries that use a forward-looking, cost-based pricing methodology as that required an essentially evidence-based approach.[95]
(b)The Commission correctly took s 18 into account when choosing a price for the UBA service “within the plausible range derived from benchmarking”.[96]
[95]High Court judgment, above n 1, at [129]–[137].
[96]At [136]–[139].
Kós J said:
[135] In my view the statutory language is not prescriptive that s 18 is a consideration borne into (and interwoven through) the benchmarking analysis to be undertaken by the Commission, at the evidential stage. Indeed, I can see every reason why it would not be. That sort of external consideration does not seem to me to form a relevant part of the evidence-gathering exercise. Rather, it is relevant to what to do with the evidence once it is obtained.
[136] In this respect the Commission was entitled, in my view, to take the approach it did, and apply s 18 after the point at which the potential benchmarks had been identified. In directing itself that it must choose a price for the UBA service “within the plausible range derived from benchmarking” the Commission was readying itself for just that inquiry. The concept of a plausible range, in context (particularly given Professor Vogelsang’s work), anticipated a weighted outcome either within or beyond the immediate benchmark range.
Submissions for Chorus
Chorus contends that the Commission erred in law by failing to consider s 18 other than in the context of price point selection. In support of this contention, Mr Goddard submits that s 18 is a mandatory relevant consideration that the Commission was required to consider at each stage of its analysis when making judgments that had a material effect on the ultimate result. That included process decisions, decisions to exclude eligible benchmarks, decisions on whether to carry out further information gathering on the range within which a price point would be selected, as well as selection of the fixed price point.
Mr Goddard accepts that the Commission took s 18 into account in the process of selecting the final price point, but erroneously failed to do so when making its decisions during the first stages of the benchmarking exercise which also involved value judgments. If, as required, the Commission had complied with the need to “err on the high side”, it would have selected a price point above $10.92.
Discussion
There is no dispute that s 18, the purpose provision, is a mandatory requirement for the Commission to apply when making a determination under the Act. The mandatory nature of the requirement is clear from s 19. At the same time, however, the mandatory nature of the requirement is qualified in two significant statutory respects.
First, in terms of s 19(c) the Commission is required to make the determination it considers “best gives, or is likely to give, effect to the purpose set out in s 18”. As we have already noted, this means that Parliament has left it to the Commission to make a further value judgment when considering and applying the s 18 purpose provision.[97] The language of “best” giving, or “likely” to give, “effect” to the provision recognises the essentially evaluative nature of the requisite consideration. It also recognises that the Commission has a choice between a current (“best gives”) and future (“is likely to give”) assessment of the effect of the determination. This emphasises the Commission’s role as the expert arbiter under the Act.
[97]Above at [49](d)].
Second, determination of the IPP in accordance with the requirements of the statutory definition of the IPP will of itself involve implementation of the s 18 purpose provision. As we have already noted, it is reasonable to assume that Parliament will have settled on that particular definition because it is consistent with and implements the requirements of the statutory purpose.[98]
[98]Above at [44].
This means that when, for the purpose of the benchmarking exercise required by the IPP definition, the Commission obtained and analysed all the information about prices in potentially “comparable countries that use a forward-looking cost-based pricing method” it was carrying out the first stage of a process which would ultimately lead to a determination implementing the s 18 purpose. In carrying out that stage of the process which involved the identification of benchmarks that complied with the statutory definition and a plausible range of prices derived from those benchmarks, the Commission was not required to consider separately the application of s 18. To the extent that the s 18 purpose was relevant, it was already reflected in the statutory IPP definition.
As Mr Farmer submits, in assessing whether candidate countries had regulated forward-looking cost-based prices for a similar service and were otherwise comparable, the Commission was engaged in an evidence-based task where it was concerned to ensure that the benchmarks were as reliable as possible. The Commission considered all proposed adjustments on the merits and rejected adjustments where they would have made the estimate less reliable. In this context there was no requirement or need for the Commission to consider s 18 separately.
When it came to the next stage of the benchmarking exercise, the selection of the appropriate price point, there is no doubt that the Commission expressly considered the requirements of s 18 and subs (2A) in particular. It is clear that at all stages of the process followed by the Commission and in its final determination it correctly considered the purpose provision. In particular, it did so when deciding to select a price point closer to Sweden than the median.
It is clear from the Commission’s final determination it accepted, in accordance with s 18(2A), that it should “err on the high side” because of the need to consider the incentives to innovate that exist for, and the risks faced by, investors in new telecommunications services that involve significant capital investment and that offer capabilities not available from established services.[99] The Commission then in the exercise of its judgment “erred on the high side” by selecting the price for Sweden and not a price below Sweden.
[99]See above at [100]–[101].
Once again we do not accept that the Commission exercised its judgment on this issue in a way that contradicts the true and only reasonable conclusion available on the facts or was factually unsupportable.
Like Kós J, we therefore answer the third question of law – “No”.
Fourth question – Did the Commission in fact fail to consider whether a price above $10.92 would be more consistent with s 18, and thereby err in law?
High Court judgment
Kós J found that the Commission did not in fact fail to consider whether a price above $10.92 would be more consistent with s 18 because:
(a)As already decided, the Commission had not “bookended” itself with the Swedish price.[100]
(b)In considering the implications of s 18, the Commission was well aware of Chorus’s submission that the IPP outcome should be above $10.92.[101]
(c)The simple fact was that the Commission did not accept Chorus’s submissions.[102]
Submissions for Chorus
[100]High Court judgment, above n 1, at [165].
[101]At [166].
[102]At [167].
Chorus contends that the Commission erred in failing to consider whether a price above $10.92 would be more consistent with s 18. In support of this contention, Mr Goddard submits that:
(a)The Commission failed to consider whether a price in excess of the Swedish price would give better effect to the s 18 purpose because it incorrectly proceeded on the basis that s 18 was relevant only to selection of a price within the benchmark range.
(b)The Commission failed to take the s 18 purpose, and in particular s 18(2A), into account when determining the range within which the New Zealand costs might plausibly fall because it excluded from its analysis prices in excess of the Swedish price without asking whether this risked underestimating the New Zealand costs in a manner inconsistent with s 18.
Discussion
As is apparent from the reasons of Kós J and the submissions for Chorus, this question overlaps to a significant extent with the previous questions and has therefore already essentially been answered.
The short point is that, as an examination of the Commission’s process and final determination shows, it did in fact give careful consideration to the question whether a price above $10.92 would be appropriate and, equally, when carrying out that consideration in the context of selecting the appropriate price point, it took into account the s 18 purpose, including the requirement of subs (2) relating to the effect of efficiencies that will result or be likely to result in the long-term benefit of end‑users of telecommunications services in New Zealand.
Consequently we do not accept that the Commission exercised its judgment on this issue in a way that contradicts the true and only reasonable conclusion available on the facts or was factually unsupportable.
Like Kós J, we therefore answer the fourth question of law – “No”.
Fifth question – Did the Commission fail to consider whether a price reduction of this magnitude was consistent with the s 18 purpose, having regard to the limited information available to it about the cost of providing the UBA service in New Zealand, and thereby err in law?
High Court judgment
Kós J found that the Commission did not err in law in respect of this “proportionality” argument because:
(a)It was more a complaint to be addressed to Parliament than to the Commission or the Court as Parliament had defined the IPP.[103]
(b)A broad “proportionality” consideration did not add anything useful to s 18 which provided the only relevant guidance necessary and, as the Commission correctly recognised, required the negative impacts on competition of underestimating forward-looking costs to be treated as greater than those of overestimating.[104]
Submissions for Chorus
[103]At [169].
[104]At [170].
Chorus contends that the Commission erred in law by failing to consider whether a price reduction of this magnitude was consistent with the s 18 purpose, having regard to the limited information available to it about the cost of providing the UBA service in New Zealand. In support of this contention, Mr Goddard submits that, applying the proportionality test inherent in s 18, the Commission’s decision to reduce the UBA price by almost 50 per cent required especially cogent justification in terms of both probative material and reasoning in order to provide a high level of confidence that setting a price at that level would not impair incentives to innovate and invest in new telecommunications services. The material before the Commission and its analysis of that material could not, and did not, provide the necessary level of confidence that the s 18 purpose was not compromised by the Commission’s decision.
Discussion
There are serious difficulties with the submissions for Chorus on this issue in addition to those identified by Kós J:
(a)As Mr Farmer submits, the Commission’s decision-making was governed by the new IPP and s 18. There is no basis for suggesting that s 18 imports proportionality as an independent consideration.
(b)As Mr Radich QC for Vodafone submits, the evidence before the Commission and its analysis of that evidence did provide the Commission with the necessary level of confidence that the s 18 purpose was met as required by s 19. It is not for the Court in an appeal of this nature to second‑guess the Commission’s evaluative judgment in that respect.
Consequently we do not accept that the Commission exercised its judgment on this issue in a way that contradicts the true and only reasonable conclusion available on the facts.
Like Kós J, we therefore answer the fifth question of law – “No”.
Sixth question – If “yes” to any of the above, what remedy should be granted?
In view of the negative answers to the first five questions, question six does not require a response.
Result
For the reasons given, we have answered the first five questions in the negative and therefore need not answer the sixth question.
The appeal is dismissed.
Costs
Costs should follow the event. Chorus is to pay the costs of each of the first to fourth respondents for a complex appeal on a band A basis with usual disbursements to be fixed by the Registrar, including travel and accommodation expenses for out of town counsel. We certify for two counsel for the first respondent.
Solicitors:
Chapman Tripp, Wellington for Appellant
R J Bernau, Wellington for Commerce Commission
T Thursby, Wellington for Vodafone New Zealand Ltd
Russell McVeagh, Auckland for Spark New Zealand Ltd
Wigley & Co, Wellington for Orcon LtdSCHEDULE OF 5th TO 32nd RESPONDENTS
5th respondent 2Talk Limited
6th respondent Atrix Networks Limited
7th respondent Airnet NZ Limited
8th respondent Baycity Communications Limited
9th respondent Brennan Voice & Data Pty Limited
10th respondent CallPlus Limited
11th respondent CityLink Limited
12th respondent Compass Communications Limited
13th respondent Fastcom Limited
14th respondent Fusion Networks Limited
15th respondent Gisborne.net.NZ Limited
16th respondent ICONZ – Webvisions Limited
17th respondent Inspire Net Limited
18th respondent Korida Limited
19th respondent Link Telecom (NZ) Limited
20th respondent Network Edge NZ Limited
21st respondent New Zealand Technology Group Limited
22nd respondent NZ Wireless Limited
23rd respondent Planet Communications Limited
24th respondent Snap Limited
25th respondent Solarix Networks Limited
26th respondent Telnet Telecommunications Limited
27th respondent Total Consumer Services Limited
28th respondent Vanco Australasia Pty Limited
29th respondent Vector Communications Limited
30th respondent Vibe Communications Limited
31st respondent Voyager Internet Limited
32nd respondent WorldxChange Communications LimitedAPPENDIX
Relevant provisions of the Telecommunications Amendment Act (No 2) 2006 and Telecommunications (TSO, Broadband, and Other Matters) Amendment Act 2011
Amendment Act:
New section 9A inserted
The following section is inserted after section 9:
9AFunctions of Commission in relation to sector monitoring and information Dissemination
(1)In addition to the other functions conferred on the Commission by this Act, the Commission—
(a) must monitor competition in telecommunications markets and the performance and development of telecommunications markets; and
(b)may conduct inquiries, reviews, and studies (including international
benchmarking) into any matter relating to the telecommunications industry or
the long-term benefit of end-users of telecommunications services within New Zealand; and(c)must make available reports, summaries, and information about the things referred to in paragraphs (a) and (b).
…
New subpart 2A inserted
The following subpart is inserted after section 30 [of the principal Act]:
…
30A Overview of subpart
(1)This subpart enables the Commission to make, as an alternative to a determination made under section 27, a determination of the terms on which a designated access service or specified service must be supplied with reference to all access seekers and all access providers of the service.
(2) Accordingly, this subpart—
(a) provides a process for the development of standard terms for the supply of the service (sections 30C to 30J):
(b)provides for the Commission to make, and review, a standard terms determination (sections 30K to 30R):
(c)specifies how a standard terms determination is to apply (section 30S):
(d)clarifies the interface between a determination made under section 27 and a standard terms determination (section 30T):
(e)provides a mechanism that allows parties to a standard terms determination to apply for a residual terms determination if they wish to adjust, as between themselves, the application of terms specified in the standard terms determination (sections 30U to 30ZD).
(3)This section is intended only as a guide to the general scheme and effect of this subpart.
…
Schedule 1
…
2
New bitstream and bitstream backhaul designated access servicesSubpart 1 of Part 2
Omit the items relating to “Access to, and interconnection with, Telecom’s fixed PDN” and “Telecom’s fixed PDN backhaul” and substitute the following items:
Telecom's unbundled bitstream access:Description of service: A digital subscriber line enabled service (and its associated functions, including the associated functions of Telecom’s operational support systems) that enables access to, and interconnection with, that part of Telecom’s fixed PDN that connects the end-user's building (or, where relevant, the building distribution frames) to Telecom's first data switch (or equivalent facility), other than a digital subscriber line access multiplexer (DSLAM)
To avoid doubt, unless requested by the access seeker, the supply of this service must not be conditional on a requirement that the access seeker, the end-user, or any other person must purchase any other service from the access provider
…
Initial pricing principle: Retail price (as imputed by the Commission having regard to the price of any other digital subscriber line enabled service, including the imputed price of any such service offered as part of a bundle of retail services) minus a discount benchmarked against discounts in comparable countries that apply retail price minus avoided costs saved pricing in respect of the service
Plus, if no person is also purchasing a local access and calling service from the access provider in relation to the relevant subscriber line, all or any of the costs of Telecom’s local loop network that would usually be recovered by the access provider from an end-user of its local access and calling service, as determined by benchmarking against comparable countries (unless the Commission considers that the price already takes into account all of the relevant costs)
Final pricing principle:
Either—
(a)retail price (as imputed by the Commission having regard to the price of any other digital subscriber line enabled service including the imputed price of any such service offered as part of a bundle of retail services) minus a discount comprising avoided costs saved, in a case where Telecom faces limited, or is likely to face lessened, competition in a relevant market; or
(b)retail price (as imputed by the Commission having regard to the price of any other digital subscriber line enabled service including the imputed price of any such service offered as part of a bundle of retail services) minus a discount comprising actual costs saved, in a case where Telecom does not face limited or lessened competition in a relevant market
Plus, in either case, if no person is also purchasing a local access and calling service from the access provider in relation to the relevant subscriber line, all or any of the costs of Telecom’s local loop network that would usually be recovered by the access provider from an end-user of its local access and calling service, as determined by identifying the relevant costs (unless the account all of the relevant costs)
Amendment Act:
New Part 2A substituted
Part 2A is repealed and the following Part substituted:
…
69O No participation in supply of retail services
(1)Chorus, or any related party of Chorus, must not participate in the supply of a telecommunications service to a person (A) if 25% or more of the services supplied, or to be supplied, by Chorus to A in any year are or will be supplied—
(a)for A’s own use or consumption; or
(b)for the use or consumption of persons who are related parties of A.
…
Schedule 3
Amendments to Schedule 1 of principal Act
Subpart 1 of Part 2
Items headed Telecom’s unbundled bitstream access, Telecom’s unbundled bitstream access backhaul, Telecom’s unbundled copper local loop network, Telecom’s unbundled copper local loop network co-location, Telecom’s unbundled copper local loop network backhaul (distribution cabinet to telephone exchange), and Telecom’s unbundled copper local loop network backhaul (telephone exchange to interconnect point): omit and substitute:Chorus’s unbundled bitstream access
…
Initial pricing principle applicable after the expiry of 3 years from separation day:
The price for the designated access service entitled Chorus’s unbundled copper local loop network plus benchmarking additional costs incurred in providing the unbundled bitstream access service against prices in comparable countries that use a forward-looking cost-based pricing method.
5
1