Application by DBNGP (WA) Transmission Pty Ltd (No 3) [2012] ACompT 14

Decision under review 7 On 1 April 2010, DBP submitted to the ERA an access arrangement revision proposal for the DBNGP for approval by the ERA under the NGR. 8 On 14 March 2011, pursuant to rule 59 of the NGR, the ERA made a draft decision, the Draft Decision on Proposed Revisions to the Access Arrangement for the Dampier to Bunbury Natural Gas Pipeline - Submitted by DBNGP (WA) Transmission Pty Ltd (Draft Decision), not to approve the access arrangement revision proposal. The Draft Decision included a statement of the reasons for the decision and set out 109 amendments to the proposed revised access arrangement that would be required before the ERA would be prepared to approve the access arrangement revision proposal. 9 On 18 April 2011, pursuant to rule 60 of the NGR, DBP submitted revisions to the access arrangement revision proposal. On 20 May 2011, 11 August 2011 and again on 8 September 2011, DBP submitted further revised versions of the access arrangement proposal that incorporated corrections to several errors in the reference tariff calculation. This included corrected versions of the access arrangement, access arrangement information and tariff model. 10 On 31 October 2011, pursuant to rule 62 of the NGR, the ERA made a final decision, the Final Decision on Proposed Revisions to the Access Arrangement for the Dampier to Bunbury Natural Gas Pipeline - Submitted by DBNGP (WA) Transmission Pty Ltd (Final Decision of 31 October 2011) not to approve the revised access arrangement proposal as ultimately submitted by DBP on 8 September 2011. The Final Decision included a statement of the reasons for the decision and set out 73 amendments that the ERA required to be made to the revised access arrangement proposal. 11 By notice issued by the ERA on 1 December 2011, the ERA gave notice of its intention to amend the Final Decision of 31 October 2011 (exercising power under clause 20 of Schedule 2 to the NGL). The ERA invited submissions on its proposed amendments to the Final Decision of 31 October 2011, which specifically related to the forecast of operating expenditure, the extension and expansion requirements and correction of minor errors of fact in the Final Decision of 31 October 2011. 12 On 13 December 2011, DBP submitted to the ERA Submission 73 entitled "DBP Response to the ERA Final Decision on Proposed Revisions to the Access Arrangement for the DBNGP" (Submission 73) plus attachments and, on 16 December 2011, submitted to the ERA Submission 74 entitled "Further DBP Response to the ERA Final Decision on Proposed Revisions to the Access Arrangement for the DBNGP" (Submission 74) and attachments. Only sections 10 and 11 of Submission 73 were responsive to the proposed amendments to the Final Decision. 13 On 22 December 2011, the ERA published its Amended Final Decision (Final Decision) together with, pursuant to rule 64 of the NGR, its own access arrangement giving effect to its proposed access arrangement revisions, the Economic Regulation Authority's revised access arrangement determination for the Dampier to Bunbury Natural Gas Pipeline (Access Arrangement Decision).

Functions and powers of the ERA 14 The power to make the Access Arrangement Decision is governed by rule 64 of the NGR. 15 Relevantly, rule 64(2) provides that the ERA's proposal for an access arrangement was to be formulated with regard to: (1) the matters that the NGL requires an access arrangement to include; (2) the service provider's access arrangement proposal; and (3) the ERA's reasons for refusing to approve that proposal. 16 Rule 64(2) of the NGR provides that the ERA may, but is not obliged to, consult on its proposal. In this matter, the ERA did not consult on making the Access Arrangement Decision (having consulted on making the Draft Decision and the Amended Final Decision). 17 The making of the Access Arrangement Decision was an ERA "economic regulatory function or power". Accordingly, the ERA was required to make the Access Arrangement Decision in a manner that was likely to contribute to the achievement of the national gas objective: s 28(1) of the NGL. Rule 100 of the NGR requires that the provisions of an access arrangement be consistent with the national gas objective. 18 The national gas objective is defined in s 23 of the NGL as follows: 23 - National gas objective The objective of this Law is to promote efficient investment in, and efficient operation and use of, natural gas services for the long term interests of consumers of natural gas with respect to price, quality, safety, reliability and security of supply of natural gas. 19 The ERA was also required to take into account the revenue and pricing principles as defined in s 24 of the NGL when making those parts of the Access Arrangement Decision that relate to a reference tariff: s 28(2)(a) of the NGL. The ERA was permitted to take into account the revenue and pricing principles when exercising any other ERA "economic regulatory function or power" (for example, determining terms and conditions of the Access Arrangement Decision that did not relate to a reference tariff) if the ERA considered it appropriate to do so: s 28(2)(b) of the NGL. 20 The revenue and pricing principles are set out in s 24 of the NGL as follows: 24 - Revenue and pricing principles (1) The revenue and pricing principles are the principles set out in subsections (2) to (7). (2) A service provider should be provided with a reasonable opportunity to recover at least the efficient costs the service provider incurs in - (a) providing reference services; and (b) complying with a regulatory obligation or requirement or making a regulatory payment. (3) A service provider should be provided with effective incentives in order to promote economic efficiency with respect to reference services the service provider provides. The economic efficiency that should be promoted includes - (a) efficient investment in, or in connection with, a pipeline with which the service provider provides reference services; and (b) the efficient provision of pipeline services; and (c) the efficient use of the pipeline. (4) Regard should be had to the capital base with respect to a pipeline adopted - (a) in any previous - (i) full access arrangement decision; or (ii) decision of a relevant Regulator under section 2 of the Gas Code; (b) in the Rules. (5) A reference tariff should allow for a return commensurate with the regulatory and commercial risks involved in providing the reference service to which that tariff relates. (6) Regard should be had to the economic costs and risks of the potential for under and over investment by a service provider in a pipeline with which the service provider provides pipeline services. (7) Regard should be had to the economic costs and risks of the potential for under and over utilisation of a pipeline with which a service provider provides pipeline services. 21 In Application by Energy Australia [2009] ACompT 8 (Energy Australia), the Tribunal considered the revenue and pricing principles in s 7A of the National Electricity Law (NEL) which is in similar terms to s 24(2) of the NGL and the reference, in those principles, to the service provider being given an opportunity to recover "at least" its efficient costs incurred. It said at [81]: It might be asked why the NEL principles require that the regulated NSP be provided with the opportunity to recover at least its efficient costs. Why 'at least'? The issue of opportunity is critical to the answer. The regulatory framework does not guarantee recovery of costs, efficient or otherwise. Many events and circumstances, all characterised by various uncertainties, intervene between the ex ante regulatory setting of prices and the ex post assessment of whether costs were recovered. But if, as it were, the dice are loaded against the NSP at the outset by the regulator not providing the opportunity for it to recover its efficient costs (eg, by making insufficient provision for its operating costs or its cost of capital), then the NSP will not have the incentives to achieve the efficiency objectives, the achievement of which is the purpose of the regulatory regime. 22 An application for review by the Tribunal is limited in that: (1) the applicant must demonstrate that the ERA made an error of one of the four kinds described in section 246(1) of the NGL; (2) the applicant must not raise any "matter" (by way of evidence or submissions) that was not "raised in submissions in relation to the reviewable regulatory decision before that decision was made": s 258(2) of the NGL; and (3) the Tribunal must not consider any material other than that specified in section 261(1) and (2) of the NGL, which essentially comprises the material that was before the ERA, except for the purpose of determining the orders to be made by the Tribunal: s 258(3). 23 Each of those provisions attracted submissions in the course of the hearing of this application. It is convenient at this point to make some brief observations about them. 24 The grounds of review to which the applicants are limited by s 246(1) of the NGL are that: (1) the ERA made an error of fact in its findings of facts, and that error of fact was material to the making of the decision; (2) the ERA made more than one error of fact in its findings of facts, and those errors of fact, in combination, were material to the making of the decision; (3) the exercise of the ERA's discretion was incorrect, having regard to all the circumstances; (4) the ERA's decision was unreasonable, having regard to all the circumstances. 25 Consequently, the Tribunal's task is not to substitute a decision which the Tribunal may prefer to make on the material before the ERA: Application by ElectraNet Pty Limited (No 3) [2008] ACompT 3 (ElectraNet (No 3)) at [64] and [69]; Energy Australia at [70]. 26 Section 258(2) of the NGL prohibits any party other than the decision maker from raising "any matter that was not raised in submissions in relation to the reviewable regulatory decision". 27 In the context of s 258(2), the word "matter" is in a practical sense a synonym for "issue". The effect of section 258(2) is therefore to restrict the issues that an applicant can raise in a review: Application by Envestra (No 2) [2012] ACompT 4 (Envestra (No 2)) at [118]. If an issue cannot be identified as arising out of a matter that an applicant raised in a submission considered by the ERA, it is not permitted to be raised by the applicant in the review: Application by Ergon Energy Corporation Limited (Street Lighting Services) (No 6) [2010] ACompT 14 (Ergon (No 6)) at [36] and [37] regarding a similar provision of the Gas Pipelines Access (South Australia) Law (GPAL); Re Epic Energy South Australia Pty Ltd [2002] ACompT 4 (Epic Energy) at [24] and [25] regarding the GPAL; Re East Australian Pipeline Limited [2005] ACompT 1 (EAPL) at [2] to [4] and [9] and [10] regarding the GPAL. However, the identification or definition of a matter or issue in some circumstances may not be an easy one. That question is discussed later in these reasons. 28 Section 261 of the NGL regulates the matters that are required, and permitted, to be considered by the Tribunal in reviewing a reviewable regulatory decision. 29 The relevant provisions of section 261 are as follows: 261 - Matters to be considered by Tribunal in making determination (1) Subject to this section, the Tribunal, in reviewing a reviewable regulatory decision, must not consider any matter other than review related matter. (2) The Tribunal, in reviewing a reviewable regulatory decision, must (a) in all cases, have regard to any document - (i) prepared, and used, by the original decision maker for the purpose of making the reviewable regulatory decision; and (ii) that the decision maker has made publicly available; and (b) … (3) In addition, if in a review, the Tribunal is of the view that a ground of review has been established, the Tribunal may allow new information or material to be submitted if the new information or material - (a) would assist it on any aspect of the determination to be made; and (b) was not unreasonably withheld from - (i) in all cases, the original decision maker when the decision maker was making the reviewable regulatory decision; and … (6) In the case of a review of a reviewable regulatory decision of the AER that is a decision to make a full access arrangement decision in place of an access arrangement that the AER did not approve, the Tribunal may consider the reasons of the AER for its decision not to approve the access arrangement. (7) In this section review related matter means - (a) the application for review and submissions in support of the application; and (b) the reviewable regulatory decision and the written record of it and any written reasons for it; and (c) in the case of a reviewable regulatory decision that is an applicable access arrangement decision - any document, proposal or information required or allowed under the Rules to be submitted as part of the process for the making of the decision; and (d) any written submissions made to the original decision maker before the reviewable regulatory decision was made …; and (e) any reports and materials relied on by the original decision maker in making the reviewable regulatory decision …; and (f) any draft of the reviewable regulatory decision …; and (g) any submissions on - (i) the draft of the reviewable regulatory decision or the reviewable regulatory decision itself considered by the original decision maker; or (ii) …; and (h) the transcript (if any) of any hearing conducted by the original decision maker for the purpose of making the reviewable regulatory decision. 30 In short, the Tribunal's review is confined to the material that was before the ERA in making its decision. 31 As noted above, the ERA undertook consultations before making the Final Decision under rule 62, and undertook further consultations before amending the Final Decision. The ERA did not engage in consultation after the final publication of the Final Decision (amended ultimately on 22 December 2011) before making the Access Arrangement Decision under rule 64. Accordingly, the ERA did not consider Submission 74 and those parts of Submission 73 that did not respond to the proposed amendments to the Final Decision (being all parts other than sections 10 and 11). However, Submissions 73 and 74 may be "review related matter" as they come within the scope of section 261(7)(d) because they were received by the ERA before the Access Arrangement Decision, even though not considered by it. 32 If the Tribunal is satisfied that a ground of review is established, the Tribunal may consider whether to allow new information or material to be submitted for the purpose specified in s 261(3) if the new information would assist the Tribunal and was not unreasonably withheld from the ERA when it was making its decision. 33 Once the Tribunal has had the opportunity to consider any new material and the parties' submissions with respect to that material, the Tribunal must make a determination. 34 Section 259(2) of the NGL provides the Tribunal with four options; it may: (1) affirm the decision; (2) set aside the decision; (3) vary the decision; or (4) remit the matter to the ERA to reconsider the matter in accordance with any direction or recommendation the Tribunal considers appropriate. 35 There are two particular, but in a sense incidental, matters to note. 36 In participating in this proceeding, the ERA has pointed out that it is mindful of the principles stated in R v Australian Broadcasting Tribunal; Ex parte Hardiman: (1980) 144 CLR 13 (Hardiman) especially at [54] per Gibbs, Stephen, Mason, Aickin and Wilson JJ. The manner in which the Hardiman principles are to be applied in a given case depends upon a number of factors, including the presence of a contradictor, whether the matter is likely to be remitted to the initial decision maker, the nature of the review proceedings and the statutory scheme concerned. 37 Where there is no natural contradictor on an issue in the application, a decision maker may need to participate in the proceedings in order to assist the Tribunal. In addition, if the statutory regime is consistent with the decision maker taking an active role in the proceedings, it is then appropriate for the decision maker to respond substantively to the review application. The role that the decision maker takes in merits review proceedings is also different to the role taken in actions for judicial review. In the case of merits review, there is a more limited application of the Hardiman principles because the decision maker, as "administrator" of the particular statutory regime, is uniquely placed to assist in that it has an in-depth knowledge of the scheme: Macedon Ranges Shire Council v Romsey Hotel Pty Ltd and Anor (2008) 19 VR 422 at [30]; Bankstown City Radio Co-operative Ltd v Australian Communications and Media Authority [2007] FCA 2053 at [6]; Geographical Indications Committee v O'Connor (2000) 64 ALD 325 at [35]. 38 On this application, the ERA took a somewhat confined role, in particular where a particular intervener or interveners assumed the position of an active contradictor. The Tribunal nevertheless received extensive and helpful submissions from the ERA on a number of matters. 39 The second incidental matter concerns the background to existing access to the DBNGP. The present firm full haul capacity of the DBNGP is approximately 845 terajoules per day. The actual transmission capacity from time to time varies depending on ambient temperature, gas temperature and composition and other factors. However, all of that capacity is contracted with various users under access contracts, the durations of which extend beyond the period of the Access Arrangement Decision. These access contracts between the DBP and users of the DBNGP, namely the DBNGP shipper contracts, are currently substantially independent of the access terms and reference tariffs set under the prior access arrangement for the DBNGP (being the access arrangement for the period 2005 to 2010). With the exception of an access contract with one user (Alcoa of Australia Limited), the current shipper contracts with the major users of the DBNGP take the form of the standard shipper contract (SSC) that was negotiated between DBP and major users in 2004. Pursuant to clause 20.5 of the SSC, until 31 December 2015, the tariff under the SSC is set at a rate that is independent of the tariff in the access arrangement. However, from 1 January 2016, pursuant to clause 20.5 of the SSC, the tariff under the SSC transitions to the reference tariff for the reference service that is the most similar to the service provided under the SSC. Accordingly, the capacity of the pipeline which could otherwise be accessed by customers contracting for the firm full-haul reference service established by the Access Arrangement Decision is currently fully contracted by the shippers.

GROUNDS OF REVIEW 40 Under s 245 of the NGL, DBP or any other affected or interested person or body may apply to the Tribunal for review of the Access Arrangement Decision, with the leave of the Tribunal. DBP made such an application, and received leave to apply from the Tribunal by decision of 15 March 2012: Application by DBNGP (WA) Transmission Pty Ltd [2012] ACompT 6. 41 The grounds of review are limited to those identified by DBP and in respect of which leave has been given. In this matter, they are described by DBP in the following way: (1) Rate of Return Issue - whether the appropriate rate of return on capital for the Revised Access Arrangement should be 5.74% (real, pre-tax), as determined by the ERA and as stated in the Revised Access Arrangement Information, or whether it should be some higher figure such as 10.03% (real, pre-tax), as proposed by DBP. This ground of review raised the following issues: (a) whether the ERA's construction of rule 87 of the NGR was correct; (b) whether the ERA's choices of the following inputs for the WACC modelling were in error: (i) Risk free rate of return; (ii) Market risk premium (MRP); (iii) Gamma; (iv) Inflation rate; (v) Debt risk premium (DRP); (vi) Debt raising costs; and (vii) Regulatory consistency. (2) Capital Base Issues - whether the ERA correctly adjusted the DBNGP capital base for the purposes of calculating a reference tariff in the following respects: (a) whether the opening capital base used for the Access Arrangement Decision should be escalated using a national or local rate of inflation (the selection of the appropriate inflation rate also affects certain other calculations relevant to the overall determination of the appropriate reference tariff); (b) whether a Project Management Retainer Fee paid to a related company in respect of certain pipeline expansion works should be counted as conforming capital expenditure; (c) as to the correct valuation of certain capital expenditure made by DBP in respect of the Burrup Extension Pipeline (BEP) for the purposes of the Revised Access Arrangement, having regard to: (i) the capital value of DBP's lease over the BEP (BEP Lease); and (ii) the temporal basis for depreciation of the asset. (3) Operating Expenditure Issue - whether the ERA should have taken into account the full amount of certain regulatory expenses which the ERA reduced in calculating the reference tariffs. (4) Reference Service Issues - concerning the specification of: (a) whether the ERA ought to have accepted the R1 Service which DBP proposed, as opposed to requiring DBP to offer T1, P1 and B1 Services (incorporating two grounds of review); and (b) if, contrary to DBP's contentions, the ERA was correct in rejecting the R1 Service, whether the definition of the P1 Service should be altered. (5) Terms and Conditions Issue (concerning Behavioural Limits in Standard Shipper Contracts) - whether the ERA ought to have adopted certain, more stringent, behavioural limits in the terms and conditions applicable under a reference service, which would mean that DBP did not have to set a lower amount as the available capacity of the DBNGP to provide pipeline services. (6) Coverage of Expansions Issue - whether any further expansions to the capacity of the DBNGP should automatically be covered by the Revised Access Arrangement, unless DBP demonstrates to the ERA's satisfaction that coverage is not consistent with the national gas objective contained in s 23 of the NGL. 42 The Tribunal will address the grounds of review under those headings and in that sequence. 43 On certain of those issues, entities which had made submissions to the ERA on the topic were given leave to intervene before the Tribunal. Where an intervener acted as a contradictor, the ERA took a confined role. Where there was no intervener acting as a contradictor, the ERA took a somewhat more expanded role to assist the Tribunal in its consideration. 44 BHP Billiton Nickel West Pty Ltd (BHP) intervened and made submissions in relation to the Rate of Return Issue and on the Reference Service Issues. 45 Electricity Generation Corporation trading as Verve Energy (Verve) also intervened in relation to the Rate of Return Issue and Verve and Alinta Sales Pty Ltd (Alinta) intervened in relation to the Reference Service Issues. 46 APT Parmelia Pty Ltd (a wholly owned subsidiary of Australian Pipeline Ltd) as trustee for the Parmelia Gas Trust (APA Group) also intervened in relation to the Reference Service Issues.

Construction of rule 87 of the National Gas Rules 47 The issue as to the proper construction of rule 87 of the NGR raised by DBP is, in essence, the same issue as that raised by WA Gas Networks Pty Ltd (now known as ATCO Gas Australia Pty Ltd) (ATCO) in Application by WA Gas Networks Pty Ltd (No 3) [2012] ACompT 12 (WAGN). 48 ATCO, in an application heard by the Tribunal at much the same time as the present application, sought review of a regulatory reviewable decision of the ERA made on 28 April 2011 entitled Economic Regulation Authority's revised access arrangement for the WA Gas Networks Pty Ltd Mid-West and South-West Gas Distribution Systems (the ERA ATCO decision). 49 Both DBP in this matter and ATCO contended that the ERA had failed properly to apply rule 87 of the NGR in determining the rate of return on capital. The issue arises in the following way: (1) The ERA made a calculation using a well accepted approach (a weighted average cost of capital (WACC)) incorporating the cost of equity and the cost of debt. For the purposes of estimating the cost of equity, the ERA used the Capital Asset Pricing Model (CAPM), and for the purposes of estimating the cost of debt the ERA summed the nominal risk free rate, the debt risk premium and debt raising costs. It then adopted the output of those processes for the purposes of its determination. DBP says that the models (and indeed modelling generally) are based on simplifying assumptions (as to risk and other parameters) and that there are a number of limitations as to the utility of their outcome. (2) DBP stresses that rule 87(1) requires that the determination which the ERA makes is one which is to be commensurate with prevailing conditions in the market for funds and the risks in providing reference services. (3) DBP says that the ERA did not consider and apply the requirement of rule 87(1) so as to determine a rate of return which is truly commensurate with prevailing market conditions and the actual risks of providing the reference services, but confined its approach to the calculation produced by the modelling it selected in accordance with rule 87(2), when that calculation should have been but a step in the process. (4) DBP then argues that the evidence available to the ERA established that a higher rate of return than that derived from the ERA's modelling was necessary in order to be commensurate with prevailing conditions in the market for funds and the risks involved in providing reference services; and consequently, it contends that the ERA ought to have made an adjustment (upwards) to the rate produced by the WACC modelling so as to satisfy the requirements of rule 87(1). This approach is also said to have been required by the revenue and pricing principles, in particular to afford DBP a reasonable opportunity to recover at least the efficient costs it incurs of providing the reference services and of complying with its regulatory obligations. 50 Because that issue was common to this review, and the review of the ERA ATCO decision, the commencement of the hearing of this review was brought forward so that the contentions of DBP on this issue on its review and those of ATCO on its review of the ERA ATCO decision were heard successively and before a decision on the common issue was given. 51 The Tribunal has made its determination on its review of the ERA ATCO decision (there were, broadly speaking, seven issues to address, including the issue as to the proper construction and application of rule 87 of the NGR): WAGN, delivered on 8 June 2012. It dealt with this issue at [39]-[71]. 52 In this matter, the Tribunal adheres to its conclusion in WAGN that the ERA did not misconstrue or misapply rule 87 of the NGR in determining the rate of return for the purposes of the Access Arrangement Decision. It also adheres to its reasons for that conclusion. As this matter is, of course, in a somewhat different factual context, and the contentions of counsel for DBP did not mirror those of counsel for ATCO (or, more accurately, the reverse is the case as the DBP submissions on the issue preceded those of counsel for ATCO) it is desirable to explain why the Tribunal reached that view. There will inevitably be substantial overlap and repetition in the two sets of reasons. The following also takes into account the submissions of BHP and Verve. 53 Under rule 76 of the NGR, total revenue is to be determined for each regulatory year of the access arrangement period using a "building block" approach in which the building blocks are: (1) a return on the projected capital base for the year; (2) depreciation on the projected capital base for the year; (3) if applicable, the estimated cost of corporate income tax for the year; (4) increments or decrements for the year resulting from the operation of an incentive mechanism to encourage gains in efficiency; and (5) a forecast of operating expenditure for the year. 54 The return on the projected capital base is calculated as the product of: (1) the rate of return; and (2) the projected capital base at the beginning of each year of the access arrangement period. 55 The rate of return to be applied in determining real pre-tax total revenue is a real pre-tax rate of return. A nominal pre-tax rate of return is calculated using a formula, to which it is not necessary to refer for present purposes. 56 Rule 87 of the NGR prescribes how the rate of return on capital is to be assessed by a regulator. It provides: (1) The rate of return on capital is to be commensurate with prevailing conditions in the market for funds and the risks involved in providing reference services. (2) In determining a rate of return on capital: (a) it will be assumed that the service provider: (i) meets benchmark levels of efficiency; and (ii) uses a financing structure that meets benchmark standards as to gearing and other financial parameters for a going concern and reflects in other respects best practice; and (b) a well accepted approach that incorporates the cost of equity and debt, such as the Weighted Average Cost of Capital, is to be used; and a well accepted financial model, such as the Capital Asset Pricing Model, is to be used. 57 The respective contentions of DBP on the one hand and the ERA on the other concerned the interaction of rule 87(1) and (2). Before turning to that construction issue, it is helpful to identify how DBP put its submissions to the ERA and how the ERA addressed them. 58 DBP's initial proposal of 1 April 2010 calculated revised reference tariffs using a real pre-tax rate of return of 10.76%, supported by a detailed submission of 14 April 2010. The method of determining the debt risk premium, which is one component of the rate of return on capital, is clearly a matter of general significance. On 1 December 2010, the ERA published for submissions a discussion paper "Measuring the Debt Risk Premium: A Bond-Yield Approach". DBP, and others, responded to that discussion paper. It will be necessary to discuss the consideration of the debt risk premium separately. 59 On 14 May 2011, the Draft Decision was published. It did not adopt DBP's proposed rate of return, but instead required the adoption of a real pre-tax rate of return on equity of 7.16%. 60 There followed further submissions, and the ERA's Final Decision of 22 December 2011 which required the adoption of a real pre-tax rate of return on equity of 5.74%. 61 Throughout its submissions DBP put its position in principle in the following way. 62 For the purpose of estimating the cost of equity, the derivation of the real pre-tax WACC under rule 87(2) then to be used to determine the rate of return, DBP used the Sharpe-Lintner CAPM. There is no dispute that the use of the Sharpe-Lintner CAPM conforms with the requirements of rule 87(2)(b). 63 The inputs to the WACC and the Sharpe-Lintner CAPM are separately addressed below. 64 As was the case with ATCO, the outcome of the modelling under s 87(2)(b) was not - in DBP's submission - the end of the process of selecting a rate of return. 65 It argued that after a WACC figure is arrived at, rule 87 requires consideration of whether that WACC figure is commensurate with prevailing conditions in the relevant market for funds and the risks involved in providing reference services. If it is, and it is thought to be, or to represent, the best forecast or estimate possible in the circumstances, then the WACC figure should be adopted as the rate of return determined for the purposes of rule 87. However, if the WACC figure is not commensurate with prevailing conditions in the relevant market for funds and the risks involved in providing reference services, then the WACC figure requires adjustment in order to ensure that the rate of return is so commensurate. That is required, DBP argues, to meet the objective provided by rule 87(1). 66 In addition, DBP says that the modelling is almost inherently likely to produce an unreliable outcome. That is because the use of a financial model such as the Sharpe-Lintner CAPM and, more generally, use of an approach involving the calculation of a WACC incorporating the cost of equity and the cost of debt, requires the making of estimates for the parameters to be used in that model, or for the parameters to be used with that approach. Use of any specific financial model, and specific approach, involves simplification and approximation. It also involves error in parameter estimation. Simplification, approximation and estimation error allow the possibility that the result obtained is not, of itself, commensurate with prevailing conditions in the market for funds and the risks involved in providing reference services. DBP pointed out in greater detail what it said were the weaknesses in such modelling. In the end, at [79] of its outline of submissions concerning rate of return, it firmly asserted that the result required by rule 87(1): cannot be achieved solely by reliance on the inputs and parameters of a well accepted approach and a well accepted financial model and at [82] it asserted that such modelling will not provide an adequate forecast of an appropriate WACC to satisfy rule 87(1). 67 DBP therefore assessed the estimate of cost of equity provided by the Sharpe-Lintner CAPM against: (1) estimates of the cost of equity obtained from using three other asset pricing models (Black's Capital Asset Pricing Model, the Fama-French three factor model and the zero beta version of the Fama-French three factor model); and (2) market based evidence on the cost of equity derived from advice provided by equity analysts to prospective investors in similar infrastructure businesses, supported by a report of NERA Economic Consulting (NERA) in relation to the other models, and a report of Strategic Finance Group Consulting (SFG) in relation to market based evidence. 68 It then adjusted the modelled estimate of the cost of equity so that the estimate was consistent with the estimates of the cost of equity obtained from the other asset pricing models and from the equity analysts' reports, thereby ensuring its commensurability with prevailing conditions in the market for funds and with the risks involved in providing reference services. 69 In reaching that result, it also adjusted certain of the input parameters into the WACC and the Sharpe-Lintner CAPM. As noted, whether the correct inputs were used is a separate issue. 70 In the Draft Decision, the ERA at [343] recognised the contention of DBP that the outcome of the modelling permitted by rule 87(2) does not dictate the outcome to the fixing of the rate of return on capital, but merely produces a guide or provisional result then to be tested and adjusted as required to give effect to the prescriptive standard of rule 87(1). It noted the contrast drawn between the more prescriptive approach of the National Electricity Rules (NER) and the NGR. It agreed at [352] that, in terms of rule 87(2)(b), there may be more than one well accepted CAPM; but it said that the Sharpe-Lintner CAPM is a well accepted financial model. It then discussed in detail at [362]-[395] the four different CAPM models proposed by DBP and explained why it chose to use the Sharpe-Lintner CAPM model. The ERA noted that it is the model used by other regulators in Australia. 71 As noted, DBP accepts that the ERA decision to use the Sharpe-Lintner CAPM model as a well accepted model for the purpose specified in rule 87(2)(b) is an available one. It is also the starting point for the DBP calculation. 72 The ERA then, in the Draft Decision, considered the other means of determining the cost of equity in the submissions of DBP. As ATCO had done, DBP had provided reports from SFG and NERA to suggest alternative methods of estimating the cost of equity, either as a check or as a substitute for the modelled outcome: the dividend yield technique, and the residual income model. The ERA explained why it did not accept those alternatives, even if their consideration were permitted under rule 87. 73 In the Final Decision at [601], [608] and [612], the ERA adhered to those views, notwithstanding the further submissions of DBP and the updated reports of SFG and NERA. It also adhered to its approach as discussed in the Draft Decision, to arrive at the real pre tax rate of return on capital of 5.74% at [615]-[620]. That conclusion was carried through to the Access Arrangement Decision. It is also reflected in the ERA's Access Arrangement Information for the Dampier to Bunbury Natural Gas Pipeline document of 22 December 2011. 74 DBP submitted, for much the same reasons as ATCO: see WAGN at [54]-[59], that the modelled outcome under rule 87(2)(b), assuming sound inputs, should be a starting point or a preliminary step in the determination of the rate of return on equity, so that the more generally expressed standard in rule 87(1) should then be applied to test and, if appropriate, to adjust the modelled outcome. 75 Rule 72(1)(g) requires the access arrangement information for a full access arrangement proposal (see rules 42, 43 and 46) to include: the proposed rate of return, the assumptions on which the rate of return is calculated and a demonstration of how it is calculated. It is clear that the fixing of the rate of return is a critical element in ensuring the objective of the NGL and in satisfying the revenue and pricing principles. The material provided by DBP includes a table showing the impact of the selection of different rates of return. The range between the figure proposed by DBP in its response to the Draft Decision and that adopted by the ERA in the Final Decision is very significant, both in terms of the value of the return on capital during the access arrangement period, and in terms of the reference tariffs which will apply at 31 December 2015 (after which the next regulatory period will come into operation). 76 There is no challenge to the validity of rule 87, but it is said by DBP it must reflect the national gas objective and the revenue and pricing principles contained in the NGL, ss 23 and 24 respectively. DBP notes that the rule making power in s 291 of the NGL empowers the rule maker Australian Energy Market Commission (AEMC) only to make a rule if it is satisfied that the rule "will or is likely to contribute to the achievement of the national gas objective", and that s 293 also requires the AEMC to take into account the revenue and pricing principles in making a rule which relates to topics specified in Schedule 1 (which include this topic). Indeed, rule 100 requires the provision of an access arrangement to be consistent with the national gas objective. The Tribunal does not consider that those considerations in any real way advance the question of the proper construction of rule 87. 77 Inevitably, a regulator such as the ERA has a public watchdog function. It is directed by the applicable rules such as ss 23 and 24 of the NGL and rule 87(1) of the NGR to attend to the proper interests of (in this case) the covered pipeline service provider on the one hand, but on the other hand it is required to be mindful of the long term interests of consumers of natural gas with respect to price, quality, safety, reliability and security of supply of natural gas and the concept of economic efficiency in setting tariffs under an access arrangement. Thus, for instance, rule 87(2)(a) requires that the service provider is assumed to be efficient and meets the benchmark standards and best practices of a going concern. In respect of capital expenditure, rule 79 imposes similar limiting criteria. So too does rule 91 in relation to operating expenditure. The national gas objective itself in s 23 of the NGL refers to promoting efficiency for the long term interests of consumers of natural gas (inter alia) with respect to price. That objective, focused on the interests of consumers, is balanced and informed also by the revenue and pricing principles to be applied by the regulator, but still in the context of ensuring the covered pipeline service provider acts efficiently: see also s 28 of the NGL. 78 In East Australian Pipeline Pty Ltd v Australian Competition and Consumer Commission (2007) 233 CLR 229 Gleeson CJ, Heydon and Crennan JJ, in the context of s 8.10 of the Gas Code said at [49]: The framework for third party access to natural gas pipelines set out above directs attention to the multiple objectives of an approved access regime. Stripped to essentials, such a regime is at least intended to allow efficient costs recovery to a service provider and at the same time ensure pricing arrangements for the consuming public which reflect the benefits of competition, despite the provision of such services by monopolies. The balancing of those objectives properly has a natural flow-on effect for future investment in infrastructure in Australia. 79 Their Honours at [50] also said that "the greater the degree of uncertainty and unpredictability in the regulatory process, the greater will be the perceived risk of investment". 80 Their Honours' observations would apply with equal force to the NGL and the NGR. 81 For the reasons given in WAGN, the Tribunal does not accept that, on the proper construction of rule 87, the two stage process suggested by DBP is the correct one. The following is, in essence, simply a repetition of the Tribunal's reasons in that decision at [60]-[71]. 82 In the Tribunal's view subrules 87(1) and (2) operate together in the following way. 83 Rule 87(1) describes the objective when the ERA is determining the rate of return on capital. It is an objective which is of course consistent with the national gas objective and with the revenue and pricing principles. It contains no guidance as to how the objective is to be achieved. In the interests of regulatory consistency, it is desirable that such guidance be provided. 84 Rule 87(2) provides that guidance. In particular, rule 87(2)(b) describes how the rate of return on capital is to be determined. It does so by prescribing the use of a "well accepted approach" and a "well accepted financial model". The explicit criticisms of modelling in the DBP submissions must be minimised, if not negated, by the requirement that the approach and the model used must be well accepted by those who undertake and use such approaches and models for that purpose. It is almost inherently contradictory then to say that the approach or the model is not likely to produce a reliable output - assuming the inputs are appropriate - if that approach and that model are well accepted. 85 In this matter, DBP accepts that the WACC approach is well accepted. It also accepts that the Sharpe-Lintner CAPM is a well accepted model. 86 Of course, the inputs into the model are critical. On that aspect, too, rule 87(1) informs the appropriateness of the inputs. That is one of the two explanations offered by DBP through SFG for an output which may be counter-intuitive. The inputs, too in this matter have been criticised by DBP. They form alternative grounds of review on this application and are addressed below. 87 The selection of the appropriate input parameters is a critical step in ensuring that the well-accepted approach using a well accepted financial model produces an outcome which accords with the objective expressed in rule 87(1). For the purposes of the construction argument, it is valid to assume appropriate input parameters. As noted, some of those used by the ERA are in dispute. But there is no reason to conclude that it was contemplated by the drafters of rule 87(2)(b) that with appropriate input parameters it would not produce an outcome consistent with the objective. 88 That conclusion also reflects the mandatory expressions in rule 87(2)(b): "is to be used". The words of rule 87(2)(b) are clear. There is nothing which suggests that the outcome generated by the proper application of rule 87(2)(b) is but a starting point, or is somehow provisional. If that were intended, it could clearly have been expressed. The proper construction gives appropriate work to both rule 87(1) and (2). It avoids the suggestion that the rule 87(2)(b) process - as a preliminary or provisional step - has no real utility. 89 At a more general level, that construction avoids the suggestion that the selection of the rate of return on capital is to be done by the ERA under rule 87(1) without any real guidance. The measure of prevailing conditions in the market for funds, and of the risks involved in providing reference services - without prescribing finally how that is done - would be fraught and vulnerable to an evolutionary and possibly idiosyncratic series of regulatory decisions. It would provide less certainty. It would expose the process of selection of the rate of return on capital to the risk of prolonged debate about the relevant factors, their empirical measurement and their weighting. 90 Senior counsel for DBP placed considerable weight on the way in which rule 87 in the NGR came to be adopted, and its comparison with the more prescriptive provisions to fix a rate of return under the NER. It is argued that those two considerations support the two-stage process under rule 87 for which DBP contends. 91 It may be accepted, as DBP pointed out, that the historical derivation of the current NER is more clearly traced to the process of privatisation of state-owned electricity transmission and distribution lines, whereas the NGR and Gas Code emanated from long-standing private ownership of covered pipeline systems. 92 It is clear enough that the present form of rule 87 of the NGR appears to derive from ss 8.30 and 8.31 of the earlier Gas Code. 93 The "Review of the Gas Access Regime", Productivity Commission Report No.3, 2004, recognised that there is a range of plausible values which can be generated for the regulatory rate of return using the WACC CAPM approach. That Report also discussed the use of the CAPM (including the Sharpe-Lintner CAPM) in that context. The views of the Productivity Commission that its use was too formulaic, and based on debatable assumptions, led to the Expert Panel on Energy Access Pricing convened by the Ministerial Council on Energy expressing concern about those views. The 2006 Comprehensive Legislative Package: Overview and Response to Expert Panel on Energy Access Pricing, November 2006, of the Standing Committee of Officials of the Ministerial Council on Energy, in turn considered and responded to those views, including on the rate of return. 94 Rule 35 of the draft NGR released for consultation in November 2006 was simply expressed: proposed rule 35(1) was in terms not dissimilar from rule 87(1) and proposed rule 35(2) prescribed the use of a well accepted financial model, such as the CAPM, in determining a rate of return on capital. There was no provision like rule 87(2)(a). The draft rule was revised in July 2007 and again in March 2008 before its emergence as rule 87 in its present form. 95 That material, in the view of the Tribunal, does not assist in the proper construction of rule 87. 96 EAPL, in the view of the Tribunal, does not assist the construction of rule 87 either. It concerned the role of the Tribunal in reviewing a decision of the Australian Competition and Consumer Commission (ACCC) under the Code. The Tribunal set aside an ACCC prescribed access review because the ACCC had erred by substituting an access arrangement which included the calculation of the capital value of the pipeline assets at variance with the methodology prescribed by the Code. The ACCC had its decision restored on judicial review in the Federal Court because, it was found, there had been no reviewable error in the ACCC decision in terms of s 39(2) of Schedule 1 to the Code, the then equivalent of s 246(1) of the NGL. The High Court (Gleeson CJ, Heydon and Crennan JJ and in a separate concurring judgment Gummow and Hayne JJ) restored the decision of the Tribunal. The High Court concluded that the term "unreasonable having regard to all the circumstances" (see also s 246(1)(d) of the NGL) covered the case where the failure to exercise a discretion properly may be inferred from the character of the result in the sense that it is unreasonable or plainly unjust: see per Gummow and Hayne JJ at [80]; the plurality agreed at [13]. 97 Although the plurality reasons discuss at some length the role of the ACCC as the regulatory decision-maker having regard to the then (more or less equivalent) Reference Tariffs and Reference Tariff Policy, the Tribunal does not consider that there is anything in those reasons which is of particular assistance in addressing the construction of rule 87(1) and (2) of the NGR. Nor, in the Tribunal's view, do the reasons of Gummow and Hayne JJ about the meaning of s 8.10 of the Code (set out at [25] of their Honours' reasons) assist. The structure and terms of s 8.10 are quite different from rule 87 of the NGR. Indeed, their Honour's remarks at [50] and [59] suggest the desirability of a reasonably certain process to select a rate of return on investment for regulatory purposes. 98 As senior counsel for DBP argued, there are clear differences in the terms of the NER and the NGR relating to the determination of the rate of return on equity. Rule 6A.6.2 of the NER is considerably more detailed. However, as pointed out in WAGN at [70], rule 6A.6.2(j)(1) of the NER provides that, in undertaking a review concerning the return on capital, the regulator must have regard to: … the need for the rate of return calculated for the purposes of paragraph (b) to be a forward looking rate of return that is commensurate with prevailing conditions in the market for funds and the risk involved in providing prescribed transmission services; 99 In the Tribunal's view, that general direction - albeit in a different position than rule 87(1) of the NGR and in a more detailed set of rules - indicates that conceptually the aims of the NER and the NGR are the same or similar. 100 It is clearly inappropriate to follow slavishly the approach to the determination of rate of return under the NER. The differences in wording are enough to make good that proposition. The NER prescribe a different procedure to determine a rate of return, and substantively somewhat different steps. The differences are explained in detail in DBP's Submission 67 to the ERA of 13 September 2011. The fact that, over a long period of consultation, there has been no precise alignment of the NER and the NGR on this topic does not, on the other hand, point to a particular meaning for rule 87. It indicates simply that rule 87 must be construed and applied according to its own terms. The lack of alignment does not of itself point to a particular way in which rule 87 should be construed and applied. 101 There is no reason why the comparison of the NER and the NGR should lead to a strained construction of the NGR. Rule 87 of the NGR should be construed in its own context, and by consideration of the full terms of the NGR. Even if, as a principle of construction, it is legitimate to discern the meaning of rule 87 of the NGR by comparison with the quite differently expressed NEL, it must be borne in mind that each of those sets of rules has the common direction in rule 87(1) of the NGR and in rule 6A.6.2(j)(1) of the NER. 102 In view of the conclusion the Tribunal has reached, it is not necessary to comment on the alternative methods suggested in the SFG reports. It is, however, necessary to note that the selection of the brokers, the quality of their reports, the analyses of the so-called comparable infrastructure firms, the quality of their dividend yield forecasts and capital gain forecasts, and the compatibility of their recent capital raisings are all not fully argued or justified or, if those things were assessed by SFG, it is not transparent how that was done. Such matters would, or may, require very careful analysis on a case-by-case basis before a fair independent assessment acceptable to a regulator could be provided and such analysis would be necessary to satisfy rule 87(1). 103 For those reasons, the Tribunal rejects the contention of DBP about the proper construction of rule 87.

Risk free rate 104 In estimating the cost of equity to be used in the determination of the real pre-tax WACC under rule 87(2), both the ERA and DBP agreed that the Sharpe-Lintner CAPM was appropriate. A key underlying parameter in this model is the nominal risk free rate of return. 105 DBP estimated the risk free rate of return as the average of the daily yields on Commonwealth Government bonds with a ten-year term to maturity for the 20 trading days ending on 18 March 20l1, as reported by the Reserve Bank of Australia (RBA). This yield was 5.48%. 106 In contrast, the ERA in both its Draft Decision (at [687] to [715]) and Final Decision (at [467] to [475]) determined, in part of the basis of advice provided to the NSW Independent Pricing and Regulatory Tribunal (IPART) by Professor Kevin Davis, that it was preferable to base the estimate of the risk free rate of return on Commonwealth Government bonds with a five-year term to maturity. Professor Davis advanced the argument (summarised in the Draft Decision at [692] to [695]) that the maturity of debt relevant to the determination of the risk free rate of return should be the same as the length of the regulatory period. Support for this view was also found in the work of Associate Professor Martin Lally, in the AER's WACC Review of 2008/2009, and in a Draft Decision made by IPART dated February 2011. (The Tribunal notes that on 17 August 2011 Professor Davis was appointed as a part-time Member of the Tribunal; naturally the Tribunal as presently constituted has formed its independent views and has not discussed this application with him.) 107 In the Draft Decision, using the 20-day trading period ending on the same date as that chosen by DBP, the ERA's estimate of the risk free rate of return was 5.46%. That is, the difference between the five- and ten-year estimates was a mere 0.02 percentage points, i.e. two basis points. 108 In the Final Decision the ERA examined the debt issued by private- and government-owned energy network firms in Australia, and concluded that 52.5% of the total debt instruments of the former had an average term to maturity of under five years, and that for the latter some 44% of total debt instruments had an average term of less than five years. The ERA also considered trading in futures contracts for Commonwealth bonds for both a three-year and ten-year term to maturity, and found that both were actively traded, but that investors showed a preference for the shorter dated contracts. 109 Therefore, in the Final Decision, using a 20-day trading period ending 30 September 2011, a date that was as practical as could be expected to allow the timely issue of the Final Decision on 31 October, the ERA at [475] determined a risk free rate of return of 3.80%. This was based on bonds with a five-year term to maturity, a term which the ERA said was proper for "compelling reasons" at [462]. 110 DBP contended that the ERA was in error in its determination of the risk free rate of return for two reasons: (1) it wrongly used bonds with a five-year term to maturity; and (2) its method of estimating the risk free rate of return was not consistent with the basis on which it estimated the MRP, as this parameter was estimated by considering a long-term average of historical equity yields relative to the yields on Commonwealth Government bonds with a term to maturity of ten years. 111 DBP relied on research conducted for it by AMP Capital Investors (AMPCI), a firm which had been involved in procuring debt funding for it. DBP said that this research showed that when properly interpreted, the data on terms to maturity indicated that energy network firms issue debt with a term to maturity of longer than five years. However, it acknowledged that three-year bond futures were more heavily traded than ten-year futures, and that it could be inferred that the traded volumes of both three- and five-year futures contracts implied that an efficient market was in operation. 112 Given these inconclusive findings, DBP submitted that some other criteria needed to be found to help decide on whether a five-year or ten-year term to maturity should be used to establish the risk free rate of return. 113 In its April 2011 report at page 6, AMPCI recommended that a ten-year term to maturity should be adopted as this bond "is able to 'look through' temporary anomalies in business cycles, interest rate cycles and inflationary cycles and thus produce a more consistent measure of the true risk free rate over time". Consequently, in paragraph 6.3 of Attachment 1 to its Submission 55: Rate of Return to the ERA dated 20 May 2011 (Submission 55), DBP argued that the longer time frame was better as "recently reported yields incorporate the latest market information and expectations about future rates, but they also contain a random component ('noise')". 114 It was also submitted by DBP that use of bonds with a ten-year term to maturity, the longest issue of a Commonwealth Government bond, was appropriate for long-lived assets like those owned by DBP. 115 The ERA submitted that DBP's submissions in favour of the risk free rate of return being based on bonds with a ten-year term to maturity were recitations that did not demonstrate any error in its reasoning. 116 The Tribunal notes here that the risk free rate of return is a clearly defined, if abstract, concept. It measures the return on a bond that carries no risk for the investor. It is widely accepted that the closest approximation to such a bond will be government debt. The appropriate term to maturity of such debt, which will yield the best estimate of the risk free rate of return is not, however, written in stone. It is a calculation about which reasonable minds differ, and is one which may well differ depending on the date at which the rate must be identified. 117 DBP asks the Tribunal to find error in ERA's determination of the risk free rate of return, this error being "the unreasonable and incorrect use of yields on Commonwealth Government bonds with terms to maturity of 5 years". It does so on the basis of a different opinion on the appropriate term to maturity advanced by its adviser AMPCI. 118 It is not for the Tribunal to substitute another estimate for that chosen by the ERA, unless error or misuse of discretion or unreasonableness can be demonstrated. In both the Draft Decision and the Final Decision the ERA carefully explained its reasoning in coming to its chosen risk free rate of return based on bonds with a five-year term to maturity. It clearly did consider the report produced by AMPCI for DBP after the release of the Draft Decision (Final Decision at [463] to [465]). It chose not to accept what AMPCI recommended. 119 In its Draft Decision the ERA spent some time discussing the appropriate term to maturity. At [691] it discussed the reasons why, traditionally, regulatory practice had been to use a ten-year term to maturity when estimating the risk free rate of return. At [700] it referred to the AER's WACC Review in 2008/09 in which the AER expressed some concern that a risk free rate of return based on a ten-year term to maturity might not be correct and could lead to overcompensation, and noted in [703] that, its concerns notwithstanding, the AER had nevertheless retained a ten-year term to maturity in its ensuing determination of WACC parameters for electricity and distribution network service providers. 120 The main issue raised by DBP concerning the risk free rate of return was that the ERA was inconsistent, and therefore in error, in the way it estimated this parameter, compared with its determination of the MRP, where its estimate was based on the long-term average of excess returns relative to ten-year Commonwealth Government bonds. 121 This inconsistency arises, DBP submitted, because in the CAPM formula used to calculate the rate of return on equity, the risk free rate of return effectively appears twice - once in its own right and again as a component in the calculation of the premium for equity risk, calculated as the product of the MRP and the equity beta. Simply put, in its own right the risk free rate of return was based on five-year bonds, but when used as an input into another parameter it was based on ten-year bonds. 122 It is true that in a formula like the CAPM, where the same concept appears twice, it is desirable that it should be estimated consistently. The ERA did not dispute this principle. However, it submitted that its methodology did not involve any inconsistencies for two main reasons. 123 First, it estimated the MRP from several different sources that were clearly noted in its Draft Decision (at [730] to [744]), including the work of Associate Professor Handley, who based his estimates on ten-year Commonwealth Government bonds; other regulatory decisions, including those of the AER, IPART and the Queensland Competition Authority; surveys of market risk practice; and qualitative information on the state of the Australian financial market. The Tribunal notes here that the MRP estimates derived from the last three sources were clearly not expressly derived on the basis of a risk margin above ten-year Commonwealth Government bonds. 124 In reply, DBP submitted that ERA provided no indication how it used the non-quantitative information listed in [123] to modify the estimate provided by Associate Professor Handley, so the error relating to inconsistency that it had identified remained. 125 Second, the ERA argued that any estimate of the MRP will be subject to estimation error, and by implication any one specific point estimate will be subject to this error. 126 On this second point, the ERA submitted that the estimation error underlying the determination of the MRP value is substantially in excess of the difference between the nominal risk free rate of return derived from Commonwealth Government bonds with five-year and ten-year terms to maturity. The Draft Decision at [685] and [714] reveals that DBP's initial estimate of the risk free rate of return on ten-year bonds was 5.48% over the relevant trading period, compared with the ERA's estimate of 5.46% using five-year bonds. 127 Such a direct comparison is not possible from the Final Decision as DBP and the ERA chose different time periods as the basis for their estimates. Based on the 20 trading days prior to 28 February 2011 for ten-year bonds, DBP derived an estimate of 5.71%, while the ERA's estimate based on five-year bonds for the 20 days ending 30 September 2011, close to the date of its Final Decision and therefore more apposite as it better reflected the current market conditions, was 3.80%. 128 An indication of how sensitive the estimation of the risk free rate of return can be to both the period used to calculate it and the term to maturity of the bond chosen for the calculation is illustrated by the response to a question by the Tribunal. In response to that question the ERA calculated that the difference in the estimate based on bonds with a five-year term to maturity and that based on ten-year bonds was 45 basis points, that is, the ERA's estimate was 0.45 percentage points below the figure of 4.25% determined by DBP on the basis of ten-year bonds. This difference is considerably greater than the one estimated at the time of the Draft Decision. 129 In response to the Draft Decision, DBP argued that ERA's arguments relating to estimation error were not related to its inconsistency submission. Rather, it said, the point was that inconsistency between the different terms to maturity used as the basis for estimating the two components of the cost of equity under the CAPM meant that the ERA's estimated cost of equity could not be regarded as the best possible estimate in the circumstances, as required by rule 74(2)(b) of the NGR. 130 It goes without saying that the estimation of market financial parameters that are specified in theoretical finance models is contentious. Empirical methods, time periods and data can take on many dimensions and different degrees of relevance. No one empirical estimation method, period or data set can lay claim to absolute superiority. Counsel for the ERA accepted that "this is an area of debate". What is best in any one situation will depend on many conflicting and debatable assumptions and empirical factors. 131 The ERA clearly believed, as does the Tribunal, that the period over which the risk free rate of return should be estimated should be the same as that over which the DRP is estimated: Final Decision at [458]. 132 This consistency is important, as the DRP is in effect a risk premium for debt added to the risk free rate of return. A lack of consistency between the five-year term to maturity used to estimate the risk free rate of return and the term to maturity of the sample of bonds used to estimate the DRP would be of concern (the Tribunal will return to this issue later in this decision). The DRP is a much more granular parameter in the calculation of the WACC that is based on actual market data relating to issued bonds, and long-dated bonds are not especially common in Australia. In contrast, the MRP is a parameter relating to the equity market as a whole and is based much more on a very long-term average of excess market returns over a risk free rate. 133 DBP, on the other hand, argued that this consistency was "largely irrelevant in this context", but that consistency between the risk free rate of return and the MRP was relevant. 134 DBP went on to argue that the ERA, by not being consistent in its estimation of the risk free rate of return with the term used to estimate the MRP, had not satisfied its own standards for consistency in estimating the WACC parameters. 135 The Tribunal notes, however, that DBP in its submissions on the MRP appears to place more emphasis on the currency of the estimate (it argues that allowance should be made for the continuing impact of the recent GFC) rather than on taking a longer-term perspective, and argues that estimates should reflect prevailing market conditions. 136 The ERA had to use its discretion to determine an appropriate term to maturity for Commonwealth bonds over which to estimate the risk free rate of return. In the opinion of the Tribunal it carefully considered all the relevant material and arguments, including those of AMPCI. In its Draft Decision, it stated clearly its reasons for selecting the five-year term to maturity as the basis for its estimate of the risk free rate of return, to be consistent with the five-year bond term to maturity that it had chosen to use in order to estimate the DRP. DBP has not advanced any arguments which convince the Tribunal that the ERA's assessment involved reviewable error. 137 Accordingly, the Tribunal finds that the ERA committed no conceptual or empirical error in its choice of the length of the term to maturity. Nor did the ERA's chosen estimation method involve any capriciousness or lack of consistency between the term to maturity used in estimating the risk free rate of return and in estimating the MRP. It exercised its discretion under rule 40(3) to use a five-year term to maturity as the basis of its estimate of the risk free rate of return, and adequately explained its reasons for its selection of this five-year term to maturity, and this was a reasonable approach. 138 Accordingly, in terms of s 259(2)(a) of the NGL, the Tribunal affirms the ERA's decision in relation to the nominal risk free rate of return.

Market risk premium 139 Rule 87(1) of the NGR requires that the rate of return on capital is to be commensurate with prevailing conditions in the market for funds and the risks involved in providing reference services. It is required by rule 74(2) that a forecast or estimate of this rate must be arrived at on a reasonable basis and must represent the best forecast or estimate possible in the circumstances. 140 DBP and the ERA have both accepted that the CAPM should be used to estimate the cost of equity for the purposes of determining the WACC referred to in rule 87(2). 141 In the context of the CAPM, a key parameter in estimating the cost of equity is the MRP, which is the expected return over the risk free rate of return that investors will require to invest in a well-diversified portfolio of risky assets. It is a forward-looking concept and thus its value has to be predicted. The Tribunal recently noted in Envestra (No 2) that, as with any variable whose values have to be forecast, there is unlikely to ever be a single "right" value of the MRP, and so considerable debate generally occurs as to how this parameter can best be calculated at any given point of time. This matter provides no exception to this state of affairs. 142 DBP submitted to the ERA that the MRP should be 6.5%, based on a series of reports produced for it by Dr Stephen Bishop of Value Advisor Associates (VAA). The ERA exercised its discretion and rejected it. Instead it adopted an MRP of 6%, arguing that DBP had provided no persuasive evidence to convince it to depart from either its previous estimation methods or the rate of 6% which it had used in its previous decisions. 143 In these proceedings, DBP submitted that the ERA was in error in the way it estimated the MRP. The alleged error appears to be that the ERA produced a value of the MRP that was not commensurate with the prevailing conditions in the market for equity funds. This, DBP submitted, was unreasonable and illogically relied on regulatory precedent. 144 There was no dispute between the ERA and DBP that an acceptable way to determine the MRP was to consider the level of excess market returns over a risk free rate. Rather, the dispute was over the relevant time period in which to make this observation. 145 In its many submissions made to the ERA, DBP stressed that attention should be paid to, and the MRP value should be determined from, measures of the implied volatility of option prices in the Australian market. 146 It also argued that the ERA misunderstood the opinion of Dr Bishop, who argued that current risk spreads above the historical average that were at the time being observed in the market for debt should be reflected in an elevated MRP. 147 The ERA submitted in reply that it had considered VAA's reports in its Final Decision (at [478ff]), and noted its view that the same conditions that had affected the DRP could be expected to affect the risk premium on equity. A key assumption in the VAA analysis was that equity and debt were priced in the same market. Therefore, the then higher currently observed risk spreads on debt should be reflected in higher spreads in the equity market, a position that DBP argued was supported by Professor Bruce Grundy's theoretical work that suggests the risk premium for equity should be 2.67 times the debt risk premium. Professor Grundy's paper "Calculation of the Cost of Capital: A Report for Envestra", September 2010 was referred to and relied on by VAA. Counsel for DBP conceded that Professor Grundy's work was untested in any formal academic sense. 148 The ERA discussed Professor Grundy's theoretical proposition at some length in [493] to [500] of the Final Decision but ultimately rejected it at [501] to [504]. Its rejection was based on the work of Professor Davis and Associate Professor Handley and their warning that the Modigliani-Miller theorem that provided the underpinning for Professor Grundy's proposition should not be used to imply any specific relationship between the cost of debt and the cost of equity. The ERA concluded that any theoretical relationship between the cost of equity and the cost of debt should not be used to form a conclusion about the relevant value to be assigned to the MRP, and so rejected VAA's argument as one that could not be relied upon. 149 DBP also submitted that the ERA's estimate was wrongly based on very long-term historical risk premiums, which do not provide a reasonable basis for concluding that a forecast MRP would be the same as the long-term average. The Tribunal notes, as mentioned in the discussion above on the calculation of the risk free rate of return, that with respect to the MRP, DBP takes a very short term view of the relevant estimation period compared to the ERA, whereas when it was estimating the risk free rate of return, DBP argued for a longer time horizon than that used by the ERA. Clearly, the 'right' period for the estimation of capital market parameters that are to be included in calculations of the WACC under the CAPM is one that is likely never to be agreed by parties in a rate of return calculation. 150 In reply, the ERA contended that it looked at a range of evidence (see the previous section on estimating the risk free rate of return) and not just at long-term historical risk premiums, and emphasised that it took special note of the impact of the GFC on market volatility in 2008 and 2009. It nevertheless concluded at [488] in the Final Decision that, at the time of making its decision, the current level of implied market volatility was "substantially below" its GFC level, and also argued that it was not aware of any reliable framework from which the MRP could be directly estimated from implied volatility over a long-term period. The Tribunal notes that the usefulness of the implied volatility framework is a matter of ongoing academic debate and dispute. 151 In addition, at [505] of the Final Decision, the ERA stated that a range of sources suggested that market conditions had stabilised. It also indicated at [489] to [492] in the Final Decision that recent evidence from several sources, including academics and market practitioners, supported an MRP figure of 6%. The ERA reported at [491] that an international survey that included 40 Australian respondents produced estimates of the MRP of 6.2% by academics, 5.4% by market practitioners and from managers it was 6.5%, giving an average of 5.8%, and a median of 5.2%. DBP submitted that in VAA's opinion such survey results are not a substitute for hard market data. However, the Tribunal notes, as will be seen later in these reasons, that DBP was content to rely almost solely on evidence from a single market player in its submissions regarding the DRP. 152 As well, the ERA cited at [506] to [507] of the Final Decision, an RBA August 2011 observation of a "new economic outlook", and the supporting views of the International Monetary Fund (IMF) published in October 2010 (although the ERA did observe that a very recent statement by the IMF had suggested that world growth was slowing and global activity had become more uneven). There is some substance in DBP's contention that these were but mere broad statements about the economic outlook which said little about investor expectations regarding the MRP which would be required to provide the incentive for them to invest in equities: see Envestra (No 2) at [161]. However, the Tribunal is satisfied that the references by the ERA to the RBA and IMF statements were used only as a confirmation of its findings, rather than as a substantive foundation for them. 153 DBP also submitted that the ERA's preferred figure of 6% for the MRP, being based on long-run historical risk premiums, does not provide "a safe or reasonable basis for implying that the expected risk premium for the immediate future . . . will be the same as the long term historical average". However, the Tribunal observes here that regulators in Australia have traditionally adopted the very long-term approach taken by the ERA. That approach provides some certainty based on past averages for a forward-looking projection. The approach advocated by DBP, however, entails a much shorter-term basis for such a projection which is subject to the short-run moods of equity markets and which could result in different estimates being produced over different short periods of time. 154 The danger of a short-term approach is well illustrated in the varying figures for the MRP produced by VAA over the course of the preparation of DBP's case before both the ERA and the Tribunal. Counsel for DBP referred both orally and through Submission 55 Attachment 1 at para 8.19 to a report prepared by VAA for WA Gas Networks in December 2009 that estimated an MRP of 12.2%. DBP then approached VAA for a report pertinent to its application to the ERA and, in a report dated April 2011, VAA proposed an MRP as of that date of 7%. This figure was based in part on Dr Bishop's contention that the MRP is affected by the same factors as had been observed to impact on the DRP (he said at page 2 that he was "of the view that the capital market for securities is integrated"), that the forward view of risk in the debt market was that it was high (page 5), and that there were no impediments to moving between debt and equity markets. DBP's claim at the hearing was for an MRP of 6.5%. 155 Counsel for the ERA submitted that the link between the MRP and the DRP was "asserted as a proposition, and asserted as a conclusion, and underpins" the argument put by VAA as to why the MRP should be set at 6.5%. The ERA was not convinced by these assertions. 156 Indeed, counsel for DBP in response to a question from the Tribunal agreed that "… we could argue long and hard about whether the effects of the GFC are still being felt and the answer to that might vary on a week to week basis". But DBP persisted in its claim that 6.5% was the best value for the MRP at the time of the Final Decision, in part because that was the value adopted by both the AER and the ERA during the GFC, and also because of VAA's opinion that movements in the DRP were linked to movements in the MRP due to the integration of the capital market. 157 In submissions in reply, counsel for DBP asserted that it had not been claimed by VAA that Professor Grundy's theoretical work could be used to establish the MRP, but rather that it was "a pointer" that the MRP should be "quite significantly greater" than the DRP. The Tribunal observes that this was not how Professor Grundy's proposition was advanced before the ERA, and finds that there was no error on the part of the ERA in rejecting the algorithm on the basis upon which this was advanced before it. 158 The fact that the AER in a May 2009 decision on WACC parameters for electricity determined that an MRP of 6.5% was commensurate with conditions in the market at that time, and that the ERA had determined the same figure in its October 2009 Draft Decision on the Goldfields Gas pipeline and in its December 2009 Final Decision on Western Power's south-west interconnected network, illustrates the great state of uncertainty at that time about the continuing impact of the GFC. That crisis was not a period of long-run or even short-run equilibrium, and a departure from the norm was thought by the two regulators to be necessary. That crisis scenario was no longer thought by the ERA to be valid at the time of the Final Decision. Instead, it adopted the reasonable assumption that market risk was in the process of resuming its long-run average value. 159 In the opinion of the Tribunal it is clear that the ERA did consider all of the material on the MRP (as to timing, level, averaging period, and Professor Grundy's proposition) that DBP had submitted to it. It did not agree with these submissions, for reasons which were carefully spelled out in the Final Decision. Counsel for the ERA submitted that it took the view that Professor Grundy's proposition provided the basis for the VAA analysis authored by Dr Bishop and so it was proper for it to consider that underpinning and that this was spelled out in the Final Decision at [501ff]. Indeed it referred to Associate Professor Handley's opinion that even if the Grundy proposition were correct, it does not necessarily follow that an accurate or even unique estimate for the MRP can be derived from the DRP. 160 Accordingly, in [504ff] the ERA decided that the derivation of an MRP estimate from a DRP value according to Professor Grundy's proposition was not a safe procedure to adopt. In the opinion of the Tribunal it did consider all the arguments put forward by DBP and showed that it understood them and indeed had tested them and carefully compared them with those of its own experts. In this context, it committed no reviewable error. While not explicitly identifying the alleged error in the terms of s 246 in its submissions, it appears that DBP contends that the ERA's decision to reject the proposed MRP of 6.5% and its replacement with an MRP of 6% was unreasonable in all the circumstances. 161 The ERA's estimate was one that could be viewed as being commensurate with prevailing conditions in the market for equity funds. That it rejected the recommendations of VAA, on both theoretical and empirical grounds, is not in itself reason for the Tribunal to find error or to conclude that the ERA misused its discretion or acted unreasonably. It did so only after carefully considering both the theory and the empirical evidence. It considered but did not accept Professor Grundy's proposition about the mathematical relativity between the MRP and the DRP. Nor did it accept, on the information before it, VAA's claim that market risk levels were so high looking forward that a 6.5% MRP figure was the current best forward-looking estimate. 162 For reasons stated in the preceding paragraphs, the Tribunal is of the opinion that DBP has failed to show reviewable error on the part of the ERA in deciding at [515] in the Final Decision that a reasonable point estimate for the MRP is 6.0%. 163 Accordingly, in terms of s 259(2)(a) of the NGL, the Tribunal affirms the ERA's decision that a reasonable point estimate for the MRP is 6.0%.

Gamma 164 DBP contends that the ERA's decision on the appropriate value for gamma (0.25), involved error in terms of s 246(1) of the NGL. 165 The first contended error involved an attack on the ERA's adoption of different values of cash dividends, one (75 to 80 cents in the dollar) for estimating theta in the determination of gamma and another (100 cents in the dollar) for estimating the cost of equity using the CAPM. 166 The second contended error involved the ERA's interpretation of the practice of market practitioners not making any adjustment for franking credits in valuing a firm. 167 Gamma is a key component in estimating an Australian firm's cost of capital. In theory the cost of capital for an Australian firm should be lower if its domestic shareholders are able to use the franking credits it distributes to them to reduce their taxable incomes. Gamma measures the assumed utilisation of these imputation credits. 168 The cost of corporate income tax is one of the components of the annual revenue requirement that needs to be determined by the ERA. Gamma is an input into the calculation of this revenue requirement. 169 Other things being equal, a higher value of gamma decreases the allowance that must be made for the cost of corporate income tax in the annual revenue requirement, as a high value is assumed to mean that more of a company's tax liability will be recovered by investors through the imputation scheme. Thus, the higher the value of gamma, the smaller will be a regulated company's required revenue. Regulated entities typically argue for a gamma value that lies at the lower extremity of the range of possible values (zero to one) for this parameter. 170 In practice gamma is the product of the imputation credit payout ratio, or distribution rate (commonly referred to as F, and measured as the proportion of imputation credits paid out to investors as franking credits), and the assumed per dollar value to shareholders at the time of receipt of the distributed imputation credits (known as theta). Theta is commonly known as the utilisation rate. 171 Determining the appropriate values of F and theta has been a fiercely contested issue in Australia's regulatory history. There is no unique pair of values of F and theta that are regarded as universally correct. Therefore there is no value of gamma that is regarded as universally correct. The academic models, empirical research methods, data and relevant time periods all need to be carefully investigated. Debate is inevitable, and ultimately, which value is most relevant for the matter at hand must be decided on a case-by-case basis. 172 The Tribunal recently considered the ERA's process of determining the value of gamma in WAGN. Arguments were developed by the applicant in that case that are similar to those advanced by DBP in the matter now before the Tribunal. In WAGN, the ERA in effect acknowledged that, having regard to the Tribunal's decision in Application by Energex Limited (Gamma) (No 5) [2011] ACompT 9 (Energex (No 5)), its estimate of gamma under attack in WAGN was in error. In WAGN the Tribunal found, consistent with its decision in Energex (No 5) (a decision that was followed in Application by United Energy Distribution Pty Limited (No 2) [2012] ACompT 8), that the appropriate value of gamma was 0.25, being the rounded-up product of an F value of 0.7 and a theta value of 0.35. 173 In the matter involving DBP now before the Tribunal, the ERA determined that the best estimate of F, the payout ratio, at the time of its Final Decision was 0.7 (in its findings in the Draft Decision at [669] it had concluded that a range of 0.7 to 1.0 was supported by the existing evidence). 174 The Final Decision noted that in Energex (No 5) the Tribunal had: (a) determined that the best estimate of theta was likely to come from dividend drop-off studies; and (b) requested the AER to provide it with new estimates of theta from a specially commissioned dividend drop-off study. 175 The Tribunal's request resulted in a report by SFG entitled: Dividend drop-off estimate of theta, 21 March 2011 (SFG's March 2011 report). This report provided an estimated theta value of 0.35 which the Tribunal accepted as the best current estimate of this parameter. In the Final Decision the ERA recorded its concerns that estimates of theta derived from a dividend drop-off study were "inherently imprecise" and that, where available, a range of studies should be used to estimate this parameter. Nevertheless for the sake of "regulatory certainty" it adopted at [537] the Tribunal's determination of the value of 0.35 for theta. 176 Throughout the consultation process leading to the Final Decision, DBP advanced many arguments to support its submission that gamma should be valued at zero, but its case essentially centred on the two contentions summarised above. These contentions were based on the advice of its consultant Professor Stephen Gray of SFG, the author of the report requested by the Tribunal in Energex (No 5). The Tribunal notes that there is no debate here over the parameters in the model that need to be quantified. The argument is over the empirical techniques and assumptions used to estimate these key parameters. It should be recalled that rule 87(1) requires that the estimated rate of return, of which gamma is a component, must be commensurate with prevailing conditions in the market for funds and the risk involved in providing reference services. 177 DBP contends that the ERA did not consider the issue of consistency between the assumptions it used in estimating gamma and those it employed in the Sharpe-Lintner CAPM to estimate the return on equity. DBP submitted that the ERA was in error due to its inconsistent treatment of the value of cash dividends in both the Draft Decision at [655] to [657] and again in the Final Decision at [529]. In deriving the estimate of theta of 0.35 the ERA assumed that cash dividends were worth 75 to 80 cents per dollar, yet in its estimation of the cost of equity using the CAPM, it assumed the value of cash dividends to be 100 cents in the dollar. 178 This inconsistency was, DBP submitted, surprising given that the ERA elsewhere in its Final Decision had placed great emphasis on the need for consistency in regulatory decisions. 179 According to DBP, the ERA correctly assumed a value for F of 0.7, but was in error when it concluded that the value of theta was 0.35 instead of zero. Therefore, its determined gamma value of 0.25, being the rounded product of F and theta, should have been zero. DBP submitted that it had addressed all these issues in its Submission 55 in response to the Draft Decision, and that they had especially been considered in a report by Professor Gray (A regulatory estimate of gamma under the National Gas Rules: Response to Draft Decision, dated 13 May 2011) attached to the submission. DBP submitted that the evidence on which the Tribunal relied for its finding on gamma in Energex (No 5) should have been interpreted as implying that theta could take a value of zero and that a gamma value of zero was permissible (indeed was required) when cash dividends are valued at 100 cents in the dollar. 180 The Draft Decision rejected DBP's inconsistency arguments at [655] to [657] and the ERA maintained its position in the Final Decision at [529]. In addition, it argued that while dividend drop-off studies produced estimates of the value of dividends of less than 100 cents in the dollar, these drop-off study estimates were thought by the AER to be subject to "significant concerns" (Draft Decision at [680]) and provided no reason for valuing imputation credits (and thus gamma) at zero (Draft Decision at [646] to [647], Final Decision at [536]). 181 DBP relied heavily for its inconsistency contentions on the key points raised in [5.1] to [5.17] of Attachment 1: rate of return in the Draft Decision to Submission 55, namely, that when theta is estimated from a dividend drop-off study, this implies that cash dividends are valued at less than 100 cents in the dollar. In contrast, when pursuing dividend yield studies for the CAPM return on equity calculations, it is normally assumed the value is 100 cents in the dollar. 182 Counsel for DBP drew attention to the following paragraphs in Attachment 1 to its Submission 55: 5.15 In the SFG and other studies which estimate θ using a dividend drop-off method, θ is estimated jointly with a second parameter, the value of an additional dollar of cash dividend. Estimation produces an elliptical confidence region for pairs of values of θ and the value of an additional dollar of cash dividend. Parts of this confidence region can be rejected … SFG's estimate of 0.35 for θ is obtained jointly with an estimate of the value of an additional dollar of cash dividend of 0.85. That is, if an estimate of 0.35 is accepted for θ, then each additional dollar of cash dividend is valued at only 85 cents. 5.16 When the CAPM is applied to estimate the cost of equity, or that cost is estimated from the alternative asset pricing models which DBP has used, or is estimated from equity analysts' reports, the assumption is made that each additional dollar of cash dividends is valued at $1.00. The pair of estimates for the value of an additional dollar of cash dividend and θ, (1, 0), has the same statistical significance as a pair of estimates (0.85 and 0.35). If, in estimating the cost of equity, each additional dollar of cash dividends is valued at $1.00, then θ should be set at zero, and the appropriate estimate of gamma is also zero. (Footnotes omitted) 183 Those paragraphs were based on the conclusions of Professor Gray in his May 2011 report which states at [118]: e. Dividend drop-off analysis produces a pair of estimates. The theta estimate of 0.35 is conditional on an estimate of 0.85 for the value of cash dividends; f. If the dividend drop-off estimate of 0.35 is used as the basis for an estimate of theta (and consequently gamma), consistency requires that a value of cash dividends of 0.85 be used throughout the WACC estimation. It would be inconsistent to adopt an estimate of 0.85 when estimating gamma but an estimate of 1.0 - for the same parameter - when estimating the required return on equity. Put simply, Professor Gray's advice was that either pair of values (one and zero, or 0.85 and 0.35) could be used, but once the choice is made, the values for each of the two parameters must be used consistently throughout the process of estimating the WACC. Thus, when the value of cash dividends is set to one, as the ERA did when estimating the return on equity, the value of theta must logically be zero. 184 Professor Gray concluded at [119] that: ... an appropriate range for the point estimate of gamma is 0 to 0.25. We note that a value at the lower end of this range is consistent with market practice and with the ERA's approach in estimating the required return on equity. If a value toward the top end of the range is to be used, the corresponding 0.85 value of cash dividends must be used throughout the WACC estimation process. 185 The Tribunal is mindful that it was Professor Gray who did the work requested by the Tribunal that led to its determination of a gamma value of 0.25 in Energex (No. 5). Counsel for DBP submitted, however, that the issue being agitated in the current matter was not an issue that the Tribunal had been obliged to consider in Energex (No 5). In the matter now before the Tribunal, DBP was arguing for a zero value of gamma and in this context the opinion of Professor Gray was relevant. 186 In a later report (Regulatory estimates of gamma in light of recent decisions of the Australian Competition Tribunal, dated 20 July 2011), Professor Gray distinguished his advice to DBP from SFG's March 2011 report in Energex (No 5) by concluding at [6] that: … an appropriate range for the point estimate of gamma is zero to 0.25. We note that a value at the lower end is: a. consistent with market practice; b. consistent with the ERA's approach in estimating the required return on equity; and c. consistent with the estimate presented in Cannavan, Finn and Gray (2004), which is the only estimate published in a journal that is rated A* by the Australian Research Council. If a value toward the top end of the range is to be used, the corresponding 0.85 value of cash dividends must be used throughout the WACC estimation process. 187 Professor Gray also said in his 20 July 2011 report: 42. It is important to note that the Tribunal has ruled on only the specific matters before it. The matter in dispute in the Tribunal's final Reasons for Decision concerned the dividend drop-off estimate of theta … 43 Having determined that the appropriate distribution rate is 70% and that the best drop-off estimate of theta is 0.35, the Tribunal had no more work to do other than to multiply these two estimates together to obtain a gamma estimate of 0.25. 188 He went on to note in that an inconsistency would arise if a different estimate of the value of cash dividends were to be used in two different places in the same WACC estimation, and observed: 54. Clearly consistency is restored by using the same estimate in both steps. Logically, there are two possibilities: a. Use an estimate of the value of cash dividends of 100 cents in both steps of the WACC estimation; or b. Use an estimate of the value of cash dividends of 85-90 cents in both steps of the WACC estimation. 55. If consistency is to be restored, one of these courses of action must be taken. 189 Professor Gray concluded his 20 July 2011 report on the consistency issue as follows: 58. … the combined value of a one dollar cash dividend and the associated 43 cent franking credit is one dollar. If theta is to be estimated in a way that is consistent with cash dividends being fully valued, the estimate of theta is zero - the entire one dollar of combined value is attributed to the cash dividend in that case. 59. Consequently, if consistency is to be restored by using an estimate of the value of cash dividends of 100 cents in both steps of the WACC estimation, the appropriate (consistent) estimate of theta is zero. This, in turn, implies an estimate of zero for gamma. 60. Alternatively, if consistency is to be restored by using an estimate of the value of cash dividends of 85-90 cents in both steps of the WACC estimation, the allowed revenues must be increased to reflect the assumption that dividends are only valued by investors at 85-90 cents in the dollar. 190 Counsel for DBP submitted that the ERA failed to deal with the consistency issue in its Final Decision. Counsel contended that the ERA was in error in concluding that there was no inconsistency between an estimate of the value of cash dividends of 75 to 80 cents in the dollar with the estimate of theta of 0.35, and an estimate of the value of cash dividends of 100 cents in the dollar when the cost of equity was used to estimate the return on equity under the CAPM. 191 DBP's counsel argued that each set of two estimates of these two parameters had to be internally consistent. That is, the two elements within each pair ((100 and 0) and (75-80 and 0.35)) were strict complements, and that any one of the two values of theta could only be paired with its "partner" value of cash dividends figure. The ERA had, he submitted, opted for its two chosen values without addressing the logical basis for DBP's contention that gamma should be zero. 192 It was submitted by the ERA's counsel that estimates of the cost of equity using the CAPM approach do value dividends at 100 cents in the dollar. However, those measurements are affected by various factors, including differential taxes and other risk factors. It is because of this that estimates of dividends being worth less than 100 cents in the dollar are derived. 193 Counsel went on to submit that it is a particular feature of dividend drop-off studies and the way in which the relevant parameters are estimated that explains why dividends are valued at less than 100 cents in the dollar in these studies. This fact, counsel submitted, does not conflict, or indicate any inconsistency, with the valuation of dividends at 100 cents in the dollar in other contexts. 194 The Tribunal observes that dividend drop-off studies seek to measure the change in the price of a unit of stock when it goes ex-dividend. This measured change is assumed to represent the value to investors of two things - the dividend and the available franking credits. In perfect capital markets with no transactions costs, and with no differential taxation of dividends relative to capital gains, and no dividend imputation, the expected reduction in the price of a share on its ex-dividend day should equal the amount of the cash dividend. 195 With dividend imputation shareholders receive a gross dividend, comprised of a cash dividend plus a franking credit, where the franking credit takes the value of the tax already paid on that income at the firm level. Thus, in such a regime the expected ex-dividend share price drop-off should equal the amount of the gross dividend. However, in contrast to the theory, several studies have observed a price drop-off less than the size of the gross dividend. 196 The Tribunal considered dividend drop-off studies at length in Application by Energex Limited (No 2) [2010] ACompT 7 (Energex (No 2)) at [101] to [110], and in particular the three assumptions underlying simple dividend drop-off studies that were focused on by the parties in this matter. First, the studies assume that there are no transaction costs. Second, they take no account of the taxes paid by an investor, that is, no account is given to the fact that taxes on dividends might be different from those on capital gains - the differential taxes issue referred to above. Third, it is assumed that the share price is not subject to any other influence or risk over the immediate period of the payment of the divided. 197 As observed by the Tribunal in Energex (No 2) at [102]: The assumption that the share price is not subject to any other influence over the immediate period of the payment of the dividend would by no means be unreasonable if the share price could be measured from instant to instant before and after the dividend is paid, but in practice the share price change may be measured from the closing price on one day to the opening price on the next day. 198 The Tribunal also recognised at [110] to [111] a major estimation difficulty in these dividend drop-off regression studies - the fact that because most dividends are fully franked, cash dividends and franking credits are highly correlated (move together), leading to the econometric problem of multicollinearity (whereby it is difficult to disentangle the separate effects of each of these two factors on the share price drop-off). 199 Counsel for the ERA acknowledged that the inconsistency proposition as advanced by DBP was complex but suggested it could be summed up in simple terms to the effect that the value of cash dividends was inconsistently estimated in two places in the determination of the rate of return. He noted that while the inconsistency and market practice contentions were perhaps only canvassed briefly in the body of the Draft Decision, due to the length and complexity of the arguments put by DBP and its advisers, the ERA's discussion of the arguments was placed in Appendix 3 to the Draft Decision. 200 He also referred the Tribunal to an extensive 25 March 2010 report entitled Report to the AER, Evidence and submissions on gamma by McKenzie and Partington which was referred to by the ERA in its Draft Decision. He noted that these authors acknowledged that: (a) it is necessary to split the market value of the combined package of dividends and franking credits; and (b) the outcome of the regression analysis carried out in dividend drop-off studies is to produce a combined package of dividends and franking credits, a coefficient being estimated for each. 201 He also noted that there are, however, "substantial problems" in doing this, as McKenzie and Partington state at pp 12-13: … we cannot clearly observe the combined market value of dividends and franking credits as this measurement is confounded by other effects. We cannot, even in principle, observe the traded value of franking credits, because they are not separately traded. They come in a package with dividends. To split up the package we must make some assumptions. Depending on which assumptions we make, we will get different values for theta. The market value of a franked dividend on the ex-dividend date consists of a package that embeds the dividend, the franking credit, income taxes, capital gains taxes, discounting for the effect of time, and possibly some transactions costs . . . The problem is how to allocate the income taxes, capital gains taxes, and so on between the value of the cash dividend and the value of the credit. 202 In other words, dividend drop-off studies capture a range of factors. The problem is how to allocate these between the value of cash dividends and the value of the imputation credit. McKenzie and Partington refer to a paper The value of dividend imputation credits in Australia written by D. Cannavan, F. Finn and S. Gray (Cannavan et al.), that was published in 2004 in the Journal of Financial Economics, on which counsel for DBP and Professor Gray placed considerable emphasis, and note that the authors concluded at p. 175 that "… it is unlikely that the traditional ex-dividend day drop-off methodology will be able to separately identify the value of cash dividends and imputation credits". That study addresses the allocation problem, in what seems to the Tribunal to be a somewhat arbitrary procedure, by allocating a dollar to the value of the cash dividend and anything left over to the value of the franking credit. Cannavan et al. estimate that the value of franking credits is zero following the introduction of the 45-day rule that requires investors to hold the stock for 45 days to get the benefit of the franking credit for tax purposes. 203 Walker and Partington in a 1999 publication in the journal Accounting and Finance separated out the effects of taxes, transaction costs and discounting, so that they were removed from their estimate of the value of the franking credit, and estimated that the value of a franking credit was worth up to 96% of its face value. The difference between the Cannavan et al. estimate of zero versus this 96% estimate is clearly not due just to the different methods of allocation, but in the opinion of McKenzie and Partington allocation nevertheless appears to have driven a substantial part of this difference. 204 Section 5 of a 19 March 2010 report On the Estimation of Gamma produced for the AER by Associate Professor Handley, focuses on the suggested inconsistency in the AER's treatment of the value of cash dividends, and in particular on how the coefficient from regression-based dividend drop-off studies should be interpreted in order to estimate theta. Like McKenzie and Partington, he emphasises on page 24 that "the presence of differential taxes and risk complicates the interpretation of results from dividend drop-off studies". 205 Importantly, Associate Professor Handley then says that this estimation method "leads to what may at first appear to be an inconsistency regarding the AER's treatment of differential taxes", but this "apparent inconsistency is of no consequence". He rejected the suggestion of inconsistency (a point strongly made by Professor Gray in paragraph 66(b) of his 13 May 2011 report) as follows. What the AER had done was to rely on two classes of empirical evidence. That is, it relied on different studies and different techniques to derive different components of the CAPM - U.S. dividend yield studies were used in relation to the CAPM, and U.S. dividend drop-off studies were used in relation to gamma. 206 As observed by the ERA at [656] in the Final Decision, Associate Professor Handley was of the view that the AER, in its WACC review in 2009, was relying on the appropriate evidence in the appropriate context when it relied on these studies. 207 To summarise, as counsel for the ERA did, dividend drop-off studies and dividend yield studies do different work and measure different things. Estimation using dividend yields and theta estimation using dividend drop-off studies are separate steps in the determination of the CAPM-based rate of return. They are both necessary and appropriate steps but they are not linked steps. They seek independently of each other to estimate two different parameters. 208 In the opinion of the Tribunal, the use by the ERA of these two different classes of studies does not indicate that it was lacking in reason or was in error, nor does such estimation involve inconsistency. The ERA's embrace of the two different methods simply reflects the realities of the estimation process, a usage that is by no means novel. 209 It is important to understand that the errors argued for by DBP are the consequences of different studies doing different things. It is for this reason that there is no inconsistency between them because they are seeking to estimate two different parameters, using different empirical methodologies. There needs to be a distinction drawn between arguments for consistency based on theoretical logic, and the empirical realities of estimation that require separate and independent estimation of two parameters. 210 The material (and conclusions drawn from it) that were relied on by the ERA were objectively produced by respected academic researchers. Professor Gray advanced alternative interpretations. This whole area of discourse about inputs into the CAPM and the correct approach to estimating the relevant parameters of the CAPM, including gamma, is a continuing area of sophisticated debate involving competing opinions. An agreed position appears to be a distant outcome. In reaching its decision the ERA relied on expert opinions that were contrary to those of Professor Gray, who had been engaged by DBP. Such a difference of opinion is common amongst academics in this as in other similar areas. Thus, so long as the ERA acted reasonably in preferring one expert to another, it will not have committed error. 211 The Tribunal reiterates that there is no single agreed-upon correct value of gamma. While the value of F is relatively settled in Australia, great controversy has surrounded the relevant value of theta. Many papers on the measurement of theta, and thus gamma, were produced for the ERA's consideration, from its own and from DBP's experts. The gap between their estimating models, and their ensuing calculations, was wide. 212 At [538] of its Final Decision the ERA concluded after a detailed consideration of the F and theta values estimated by various experts: Based on an estimate of the payout ratio of imputation credits of 70 per cent, together with an estimate of theta of 0.35, the Authority concludes that a reasonable value of gamma, for the purpose of the Authority's final decision on the DBNGP Access Arrangement, is 0.25 (or 25 per cent). The estimate of gamma of 0.25 is consistent with the Tribunal's decision on gamma. 213 In Energex (No 5) at [45], the Tribunal noted that "… the estimation of a parameter such as gamma is necessarily, and desirably, an ongoing intellectual and empirical endeavour", and that its decision was based on the material presently before it. The ERA has recognised the Tribunal's decision in Energex (No 5) and has acknowledged the desirability of (inter-temporal and inter-sector) regulatory consistency and that regulated entities should be able to be confident of regulatory certainty in such matters. 214 The Tribunal observes that this is not to say, however, that a gamma value of 0.25 is the only possible value for this parameter. It is simply the best estimate currently available for use in this matter now before it. As with the estimation of many economic and financial parameters, finding the "right" value is a process of continual refinement as new models and paradigms emerge and as better data and estimating techniques become available. 215 It also was submitted by DBP that while the ERA may have made its decision on the basis of maintaining regulatory certainty with respect to gamma, DBP had in this matter raised a new point not previously argued in previous regulatory decisions and Tribunal determinations, and so consistency in the form of regulatory certainty should not have been the main driving factor. DBP said that the ERA had failed to address this. 216 In the opinion of the Tribunal a gamma value of zero, while possible, would be a very unusual practical or commercial outcome, one not seen before in Australia. ERA appeared to have been concerned about the presence of Australian investors in DBP and the need to make allowances for the value of imputation credits. This heavily influenced its conclusion that zero was not an appropriate value for gamma. The detailed arguments submitted by DBP were clearly considered by the ERA and were rejected. 217 In conclusion, the Tribunal finds that the ERA's method for determining gamma lacked neither in reason nor in evidence, nor did it lack strong support from respected academic researchers. Accordingly, the Tribunal finds that DBP has not demonstrated that the ERA's approach to determine gamma involved error. 218 The Tribunal turns now to the "market practice" ground submitted by DBP as another area in which the ERA was said to be in error. 219 The ERA noted in both its Draft Decision at [637] to [640] and in the Final Decision at [527] to [529] that advice provided to the AER in March 2010 by McKenzie and Partington suggested that it was not the case that market professionals made an explicit adjustment for imputation credits when estimating the value of a firm. This was not because these credits had no value, but because the adjustment was not easy to make. 220 Such a market practice based on the realities of empirical measurement of a firm's market value should not, the ERA submitted, lead to a conclusion that franking credits have no value, and this does not create any inconsistencies within the WACC formula. The work of McKenzie and Partington, and of Associate Professor Handley also provided strong support for this view, and indicate that it is not necessary to value franking credits in order to value a firm. Associate Professor Handley's 2010 report on gamma to the AER shows that the cost of equity can be derived without making any assumptions about the value of gamma and thus is independent of the value of gamma, and therefore it is not necessary to assume that imputation credits have zero value. 221 Accordingly, counsel for the ERA submitted that the fact that market practitioners do not expressly value franking credits when they value a company does not mean that it should be concluded that franking credits have no value. Nor, he submitted, can it be reasoned from that fact that gamma should be set to zero in the calculation of the WACC. 222 In reply, DBP acknowledged the above-mentioned position of McKenzie and Partington, but pointed out that this advice does not deny the econometric evidence that a reasonable estimate for theta, when jointly estimated with the value of cash dividends (the Tribunal's emphasis), is zero. It also took issue with the ERA's contention that estimates of the value of dividends were imprecise, arguing that this was contrary to the advice provided by SFG in a report submitted as Attachment 1 to DBP's "Submission 65: Regulatory Estimate of gamma in light of recent decisions of the Australian Competition Tribunal" (Submission 65). 223 DBP also submitted that a value for gamma higher than zero was not reasonable because, even though imputation credits must have some value to Australian investors, gamma is a measure of the effect of franking credits on the market equilibrium rate of return on equity, and a non-zero value of gamma means that share prices are not affected by franking credits, even though Australian investors gain benefits from these credits. 224 What the ERA says about market practice is effectively that it is too difficult to value cash dividends, but the market believes that gamma is not zero. The ERA also says that the market does not use gamma to derive a WACC. However, counsel for DBP submitted that neither of those propositions means that there is a basis for saying that gamma should be set above zero. Therefore, it is necessary to consider the consistency question. 225 The Tribunal finds no error in the ERA's acceptance that the market practitioners do not use a measure of gamma to derive a firm's value does not mean that gamma should be assigned a value of zero. This practice is simply a necessary response to the realities of estimation. It does not imply that imputation credits have no value to investors. The ERA considered the material put before it and weighed up the differing arguments. The choice of a non-zero value for gamma was open to it and it did not misuse or unreasonably use its discretion in determining that gamma should be 0.25, consistent with recent regulatory and Tribunal decisions. 226 For reasons stated in the preceding paragraphs, the Tribunal is of the opinion that DBP has failed to show error on the part of the ERA in deciding at [538] in the Final Decision that gamma should be 0.25, consistent with recent decisions of the Tribunal. 227 Accordingly, in terms of s 259(2)(a) of the NGL, the Tribunal affirms the ERA's decision that gamma is to be measured at 0.25 in order to determine DBP's WACC value.

Inflation Rate 228 DBP contends that the ERA incorrectly exercised its discretion or unreasonably decided to adopt an expected rate of inflation of 2.75% to arrive at the real pre-tax WACC. 229 The following paragraphs summarise the process leading to the ERA adopting the 2.75% inflation rate. 230 The Draft Decision noted that: (1) DBP used a widely accepted method to estimate the inflation rate, namely, the geometric mean of the RBA' inflation forecasts for the next ten years; (2) the ERA's own approach has been to use that method; and (3) based on that method, DBP proposed an expected inflation rate of 2.52%. 231 The Draft Decision went on to note, however, that while the ERA proposed to adopt the same general approach, for consistency with the estimates of the DRP and the calculations of the nominal risk free rate, it had decided to depart from its previous position of using an assumed ten-year term to maturity. 232 Instead, it adopted an assumed five-year term to maturity and calculated a forecast inflation rate of 2.65%. 233 The ERA attributed the difference between its 2.65% and DBP's 2.52% to: (1) DBP's use of the RBA's forecasts from its February 2010 Statement on Monetary Policy and the ERA's use of the RBA's forecasts from its February 2011 Statement on Monetary Policy, which was closer to the date of DBP's revised access arrangement proposal that was lodged with the ERA; and (2) the ERA adopting a five year term to maturity and DBP adopting a ten year term. 234 The Draft Decision's consideration of the inflation rate concluded that DBP's proposed revisions to its access arrangement should be amended to allow for a forecast inflation rate to be calculated using a five year term. This, it said, may result in a changed rate at the time of the Final Decision. 235 DBP's Submission 55 in response to the Draft Decision calculated a real pre-tax WACC in two steps. In the first step a nominal pre-tax WACC of 12.20% was calculated. In the second step, the real pre-tax WACC was calculated by removing expected inflation, estimated by DBP to be 2.57% in Submission 55, from the nominal pre-tax WACC. 236 Section 7 of Submission 55 explained DBP's estimate of 2.57% as follows: 7.13. The estimate of expected inflation of 2.57% used in calculating the real pre-tax WACC of 9.39% is an estimate made using forecasts published by the Reserve Bank of Australia in February 2011, and which were based on economic conditions early in 2011. It is, at the date of this response, an estimate commensurate with prevailing conditions in the market for funds. 7.14. The estimate of expected inflation used in calculating the WACC is an estimate of the expected change in the level of prices in Western Australia during the next access arrangement period. Change in those prices may influence the risks involved in providing reference services, but the estimate of expected inflation is not meaningfully assessed for commensurability with those risks. 237 Section 7 of Attachment 1 to Submission 55, quoted below, responded to the ERA's method of estimating the expected rate of inflation as explained in the Draft Decision. Expected inflation 7.1. Paragraph 716 of the Draft Decision noted that DBP had proposed a widely accepted method (calculation of a geometric mean of Reserve Bank of Australia inflation forecasts for a period of 10 years) to estimate expected inflation, and in paragraph 72l the ERA indicated that it had adopted the same general approach for the Draft Decision. However, for reasons of consistency with the estimates of the debt risk premium and the calculations of the nominal risk free rate, the ERA decided not to calculate the geometric mean for a period of 10 years, and had made the calculation using Reserve Bank inflation forecasts for a period of five years. 7.2 Beyond perceived requirements for consistency with the estimation of the debt risk premium, and the calculation of the nominal risk free rate, the Draft Decision provided no reason for the ERA's not accepting DBP's proposal. 7.3 This consistency which the ERA seems to require between the method of estimating expected inflation and the methods of estimating the debt risk premium and calculating the nominal risk free rate is, in the absence of more substantial argument, consistency for reasons of appearance only. 7.4. If, as noted in paragraph 6.9 above, the ERA's method of estimating the debt risk premium is accepted, the premium above the nominal risk free rate required by investors purchasing the relevant sample of bonds is a premium on bonds with terms to maturity of around five years. In these circumstances, to ensure the correct measure of risk in the aggregate, the debt risk premium should be added to a nominal risk free rate estimated from the yields on government bonds (the assumed risk free security) with terms to maturity of five years. 7.5 However, unlike the debt risk premium and the nominal risk free rate of return, expected inflation is not estimated from the return on (or other attributes of) specific financial instruments. There is, then, no reason, other than for appearance, to require the use of a geometric mean of inflation forecasts calculated for a period of five years in place of a geometric mean of inflation forecasts calculated for a period of 10 years. 7.6. In the absence of any substantial reason for change, DBP has made the estimate expected inflation used in Submission 55 by calculating a geometric mean of Reserve Bank of Australia inflation forecasts for a period of 10 years. 238 Notwithstanding DBP's response in Section 7 of Attachment 1 to Submission 55, paragraph [582] of the Final Decision noted that: "DBP has not made any response in relation to the method of estimating the expected rate of inflation." 239 The Final Decision also noted that no third party submissions were received in relation to the ERA's method. 240 The Final Decision went on to say: 583. In its amended access arrangement, DBP adopted an expected rate of inflation of 2.57 per cent, which is a geometric mean of the RBA's inflation forecasts for the next two years and the mid-point estimate of the RBA's long-term inflation forecasts of 2.5 per cent for the remaining eight years. … 585. The Authority has adopted the same approach for this final decision as was used in the draft decision. However, the expected rate of inflation has been calculated as a geometric mean of inflation forecasts by the RBA for the next two years and the mid-point estimate of the RBA's long-term inflation forecasts of 2.5 per cent for the remaining three years … 586. Using the above forecasts, the Authority has calculated the forecast inflation rate for this final decision of 2.75 per cent. 241 DBP particularised its contention of the ERA's incorrect exercise of its discretion or unreasonable decision as follows: (1) the ERA's calculation of the expected inflation rate by averaging RBA inflation forecasts for a five year period, rather than for a ten year period as the DBP proposed, solely on the basis that there needed to be a consistency between the maturity period for bonds used for calculation of the nominal risk free rate; and (2) there is no logical reason for requiring consistency between the maturity period used to estimate the nominal risk free rate for bonds and the period over which the RBA inflation forecasts are averaged. 242 It submitted that: (1) unlike the DRP and the nominal risk free rate of return, expected inflation is not estimated from the attributes of specific financial instruments; (2) there is no reason, other than for appearance, to require the use of a geometric mean of inflation forecasts calculated for a period of five years in place of a geometric mean of inflation forecasts calculated for a period of ten years; and (3) expected inflation should be forecast as the geometric mean of the inflation forecasts made by the RBA and published periodically in its Statement on Monetary Policy, the mean being calculated for a period of ten years as it had proposed. 243 The ERA responded by submitting that: (1) DBP did not make any submissions in response to the ERA's method of estimating the expected rate of inflation in the Draft Decision; and (2) there was no error as its decision was a permissible exercise of its discretion under rule 40(3) of the NGR. 244 DBP responded by submitting that the ERA was incorrect to allege that DBP made no submission in response to the Draft Decision outlining any error. In support of its submission DBP referred to Section 7 of Attachment 1 to Submission 55. The Tribunal notes that paragraph [582] of the Final Decision might have been better expressed along the lines: "DBP has not made any new or substantive response in relation to the method of estimating the expected rate of inflation" (Tribunal's emphasis). 245 Counsel for the ERA did not reply to DBP's submission outlined in the previous paragraph. Rather, he emphasised what was only implicit in the above mentioned passages from its Draft and Final Decisions but more apparent on a reading of those decisions as a whole, namely, that: The authority took the view that the inflation measure that was relevant to the WACC calculation was the rate of inflation that is expected during the regulatory period for the simple reason that that is the period to which the WACC [sic] pre-tax WACC rate of return applies. So the rate of return through the WACC formula is being estimated for a five year period. One starts with a nominal rate. One converts it to a real rate to deduct inflation. And the simple proposition is the appropriate measure in inflation is a measure that is relevant to the period during which the rate of return will apply. 246 In support of his submission counsel referred the Tribunal to paragraphs [718] and [721] of the Draft Decision and paragraphs [582] to [587] of the Final Decision. While it is not immediately apparent from a reading of those paragraphs alone that the ERA adopted a five year term because it is the period to which pre-tax WACC rate of return applies, counsel for DBP agreed justification for it was stated in paragraph [721] of the Draft Decision. And the Tribunal is of the opinion that justification for it is to be found in reading the Draft and Final Decisions as a whole. 247 Simply stated, the inflation rate input issue is whether the ERA should have adopted a five year period or DBP's ten year period for calculating the inflation rate. 248 There is no question that the ERA had a discretion to choose one or the other. The exercise of that discretion is unfettered: NGR rule 40(3). A decision by the ERA based upon the exercise of its discretion may, however, be reviewed and set aside on the ground that its exercise was incorrect having regard to all the circumstances: NGL s 246(1)(c). Likewise, it may be set aside if the decision were unreasonable having regard to all the circumstances: s 2416(1)(d). It is for the applicant to establish a ground of review: s 246(2). 249 While the ERA is mistaken in its statement that DBP did not respond to the Draft Decision in relation to the method of estimating the expected rate of inflation, a reading of the Draft and Final Decisions as a whole does indicate that the ERA was cognisant of the thrust of Section 7 of Attachment 1 to Submission 55. Also, when read as a whole, the Draft and Final Decisions do not evince mere consistency for reasons of appearance only as suggested at paragraph [7.3] of Section 7 of Attachment 1 to Submission 55 and DBP's submissions to the Tribunal. 250 There may be some apparent disconformity between the passages in the Draft and Final Decisions addressing the DRP, on the one hand, and those addressing the inflation rate, on the other hand. Read as a whole, however, the passages make it clear that the ERA was aligning the WACC inputs with the five year regulatory period and why it was doing so. 251 As observed, it is for DBP to establish a ground of review. Counsel for DBP all but conceded the s 246(1)(c) ground. Counsel then characterised the ERA's decision as "… not one that withstands logical scrutiny"; that is, it is a decision which is unreasonable in all the circumstances: s 246(1)(d). 252 That is not, however, the Tribunal's view. A reading of the Draft and Final Decisions as a whole evinces a logical and reasonable basis for the ERA's decision to align the WACC period with the regulatory period and to adopt a five year term to arrive at a forecast inflation rate of 2.75%. Nor does a reading of those decisions suggest that the ERA was incorrect in its decision in that regard. 253 The Tribunal finds no error in this process undertaken by the ERA. In so doing it notes that the RBA's inflation figures are but estimates after year two. That is, the five-year rate comprises actual estimated rates for each of the next two years, plus three common values for the next three years that are perhaps best regarded as averages themselves based on past experience. Similarly, the ten-year rate comprises different rates for each of the first two years plus a common number for each of the other eight years. This means that any such average rate based on data for more than two years is adding layers of uncertainty into the final number determined, and the larger the number of years, the less accurate the final number might be. 254 It is a simple mathematical observation that the geometric mean for a ten-year average will be much more heavily influenced by the common number than will be the five-year geometric mean. It will be included eight times, rather than three. Therefore if this common value is greater than the first two values in the calculation, the ten-year geometric mean will be higher than the five-year geometric mean. Conversely, if the common value is lower, then the ten-yeast average will lie below the five-year average. 255 For reasons stated in the preceding paragraphs, the Tribunal is of the opinion that: (1) the ERA's decision to adopt an expected rate of inflation of 2.75% to arrive at the real pre-tax WACC was correct and reasonable having regard to all the circumstances; and (2) DBP has failed to establish a ground of review in relation to the ERA's decision. 256 Accordingly, the Tribunal affirms the ERA's decision to adopt an expected rate of inflation of 2.75% to arrive at the real pre-tax WACC.

Debt Risk Premium 257 DBP contends that: (1) the ERA's estimate of the DRP by means of its bond-yield approach involved error in terms of s 246(1) of the NGL; and (2) the ERA should have adopted the estimate of the DRP provided by DBP's adviser, AMPCI. 258 The cost of debt for a company is typically estimated as the sum of the nominal risk free rate of return on debt and the DRP (plus estimated debt raising costs, which are discussed in the following section of these reasons). The DRP is the margin over the risk free rate that a debt investor in a benchmark efficient service provider would be expected to seek in order to provide debt finance to the business. 259 Neither the NGL nor the NGR prescribe a particular method that should be used to determine the DRP. And, as is the case with the MRP, there is no one single empirical method that is universally accepted as providing the "right" value for the DRP at any point of time or for any specific regulated firm. Consistent with the use of an expected, not historical, cost of equity, the DRP is an estimate. It is not determined by a regulated entity's actual cost of debt, as might be inferred from the arguments of some proponents of the use of market-based evidence to determine the DRP. 260 DBP asked AMPCI (which had previously raised funds for DBP and which had a comprehensive knowledge of its business and funding requirements) to estimate the cost of debt, assuming a representative large Australian regulated utility with benchmark gearing, as required by rule 87(2)(a), and to be guided by AMPCI's understanding of DBP's financial and business risks. AMPCI advised that the DRP may be estimated, largely in a hypothetical form, from the current costs of raising debt for a large regulated utility like DBP. Relying on this advice, DBP's May 2011 Submission 55 estimated the nominal pre-tax cost of debt at 9.95%. Attachment 1 to Submission 55 explained DBP's reasons for not adopting the Draft Decision's estimate of the DRP (3.124%) and the ERA's bond-yield approach to arriving at that estimate. Previously in the Draft Decision the ERA had noted at [491] that DBP had proposed a total cost of debt of 9.73%. DBP advocated to the Tribunal a total cost of debt of 9.52%, leading to a DRP including an allowance for debt raising costs of 5.71%. 261 Given the intensity of argument in regulatory appeals about the appropriate level of the DRP that should be included in the calculation of the WACC, the Tribunal has in two recent decisions (Envestra (No 2) in January 2012 and WAGN in June 2012) urged regulators and regulated firms to derive an agreed methodology for determining the DRP. These two decisions were not, of course, available to the ERA when it issued its Final Decision in this matter, but in its Final Decision the subject of WAGN it adopted what it called a bond-yield approach arrived at following its consideration of submissions from interested parties in response to its 1 December 2010 discussion paper Measuring the Debt Risk premium: A Bond-Yield Approach. 262 The ERA also used its bond-yield approach in the present matter. Briefly, that approach relies on the ERA observing bond yields of companies in the Australian financial market. More particularly, believing that it would yield a more reliable estimate of the DRP than the firm-specific process advanced by AMPCI, the ERA selected a sample of 15 companies which had been assigned credit ratings by Standard and Poor's in the range BBB-/BBB/BBB+. The specific sample selection criteria used to estimate DBP's DRP were developed by the ERA in its Final Decision the subject of the Tribunal ATCO decision and accepted by the Tribunal in that matter. 263 In determining the DRP in the matter now before the Tribunal, the ERA's Draft Decision (at [517] to [540]) and Final Decision (at [551] to [560]) considered several different estimating methods (including the one advocated by AMPCI) to determine the DRP. It concluded that the preferable method was the bond-yield approach, incorporating bond yields observed directly from the Australian bond market. 264 Noting AMPCI's view that the ERA's use of the Standard and Poor's credit rating as the key determinant of the peer reference sample group it developed was "an overly simplistic assessment of credit risk", the Final Decision stated, at [551], that in the ERA's opinion there "… is no better alternative approach, which is as simple, independent and transparent as the Standard and Poor's method, in assessing credit risk". As presently informed, the Tribunal agrees with the ERA. 265 DBP, however, submitted that such credit ratings were not complete or "precise indicators of the default risk for which lenders are to be compensated through the debt risk premium". Again, as presently informed, for want of better, the Tribunal is prepared to accept such credit ratings. 266 In its Draft Decision, at [550], the ERA listed three criteria that it considered should, ideally, be satisfied when determining which companies' bonds should be included in its sample. In the Final Decision, however, these were replaced by a set of less restrictive criteria that had been developed by the ERA in its Final Decision the subject of WAGN. Relevantly, the ERA decided that in order to obtain a statistically viable sample it should include companies that were not in the same industry as DBP. This was consistent with the sample selection criteria endorsed in several recent decisions by the Tribunal: see Application by Jemena Networks (No 5) [2011] ACompT 10 at [74]-[75]; Application by United Energy Distribution Pty Ltd [2012] ACompT 1 at [438]; Envestra (No 2) at [98]. 267 The ERA submitted that its methodology was consistent with the views previously expressed by the Tribunal (for example in Envestra (No 2) at [92] and [98]) that a wide coverage of bonds was needed to estimate the DRP and that a bond should be excluded only if there were sound reasons for its exclusion. In particular, the Tribunal stressed that a bond should not be excluded only because its issuing firm was in an industry different from that of the regulated entity for which the DRP was being estimated. 268 In the Final Decision, the ERA determined, at [569], that a DRP of 3.082% was "reasonable" (and added to this number an allowance for debt raising costs of 0.125%). This, it said at [570], reflected a conservative position on several grounds, including its belief that regulated firms had access to bank finance which at the time was likely to be cheaper than debt finance, and because its sample of 15 firms included BBB- and BBB bonds which in principle could be expected to have higher yields than BBB+ bonds for regulated firms. 269 AMPCI posited in its April 2010 Cost of Debt Summary Paper that, due to DBP's size, it would need to raise debt from the Australian bank and bond markets and from the U.S. public and private placement markets. In assessing what it described as the appropriate mix of portfolio funding, AMPCI focused on those Australian and US markets. It made no mention of any other international capital market, nor did it attempt to produce any assessment of the likely optimal allocation of borrowings between the Australian and US markets, an omission that led the ERA to note, at [510] in the Draft Decision, that neither DBP nor AMPCI provided convincing evidence for a different allocation of debt into different markets and that AMPCI's "allocation is very arbitrary and the cost of debt will be significantly different when the allocation changes". The Draft Decision also noted, at [512], that: (1) were the DRP to be estimated from U.S. data, then for consistency reasons all of the other WACC parameters such as the nominal risk free rate of return and the MRP should also be estimated with U.S. data, which would be "contrary to current practices applied by Australian regulators"; and (2) under the NGL and the NGR, the market for funds "is meant as the Australian financial market" and concluded that borrowing "must be treated as sourced from the Australian capital market, using Australian data". The Tribunal notes that senior counsel for the ERA said that there is nothing in the NGL or the NGR that expressly states that requirement and that it appears that it has been the AER's practice to look only at Australian markets for debt funds; his comment was not controverted by counsel for DBP or the interveners. 270 DBP submitted that this constituted an error in the ERA's construction of rule 87 of the NGR, arguing that the NGL and NGR do not limit the market for funds to financial markets within Australia. Such a limitation, it submitted, "would have imposed an artificial and unnecessary restriction on the assessment of whether a particular rate was the rate of return required by Rule 87". However, the Tribunal observes that if a regulator like the ERA had to consider a swathe of Australian and overseas markets in order to estimate the cost of debt and the DRP, the regulator's task would be of considerably greater dimensions and the scope for disagreement over allocations would likewise be considerably greater. 271 The ERA also submitted that DBP's current refinancing needs were not representative of market conditions and that to adopt AMPCI's approach would be inconsistent with rule 87 that requires the ERA to assume a benchmark efficient service provider in order to estimate the regulated rate of return. 272 In its Final Decision, the ERA considered an April 2011 update of AMPCI's Cost of Debt Summary Paper. In the updated paper AMPCI made significant changes in the level of debt allocated to different markets and in terms to maturity. The Final Decision concluded at [557] that it was "of the view that the different allocations are not transparent and ad hoc". It maintained its position that its transparent bond-yield approach was preferable to the method proposed by DBP. The Tribunal observes here that the ERA's assessment in this statement could have been better expressed, and it is the Tribunal's understanding, in the context of the ERA's overall argument, that it meant this statement to read "the different allocations are not transparent and are ad hoc" (emphasis added). 273 DBP appeared to read what the ERA said here in the same way as it has been read by the Tribunal, that is, its method was ad hoc, as it submitted that AMPCI's allocation was certainly not ad hoc and, like all such allocations, was dependent on market conditions that were known to AMPCI as an active participant in domestic and international capital markets. It argued that the ERA could have tested this allocation by seeking the view of its own capital market adviser, but did not do so. 274 The Tribunal is for the following three reasons of the opinion that the ERA was not obliged to seek the view of its capital market adviser. First, the Tribunal notes that much can change in international capital markets in a short time and that different capital market advisors often have quite different views of current finance market conditions and trends. Even if the ERA had felt obliged to act in the way suggested by DBP it is quite likely that the advice it received would have been at odds with that provided by AMPCI. 275 Second, the April 2011 AMPCI update of its Cost of Debt Summary Paper provided no substantive details on how the final estimate of the cost of debt was determined. Five broad categories of information were listed as the source of the pricing data, but no discussion of the data was provided (two of the sources were listed as "indicative"). 276 Third, AMPCI provided no detail or justification of how the proposed hypothetical debt was allocated across the different markets. Several criteria were listed as determinants of the appropriate mix of portfolio debt funding, but no discussion of the weights assigned to the various sources or types of funding was provided, except to say they would vary depending on the circumstances of the large regulated utility in question. It was submitted by counsel for the ERA that the process was "entirely opaque and furthermore it is entirely subjective". The Tribunal agrees with this assessment. 277 DBP also submitted that it had not discarded the cost of debt model (which, in any case, was not required by rule 87(1), as had been submitted by BHP), but rather had applied that model and sought to test whether its application produced a rate of return that would meet rule 87(1), using the expertise of an experienced capital markets adviser in establishing compliance with the requirements of rule 87(1). It also submitted that the ERA was in error as it had not provided an estimate of the DRP which reflected the specific risks faced by DBP. 278 The ERA rejected this approach and opted to use the bond-yield approach, directly observing bond-yields from the Australian market. It chose not to include bond market data from other countries to inform its analysis (see [465]-[466] of the ERA ATCO Decision), one of the main reasons being that solely Australian financial data have in the past been used consistently by Australian regulators to estimate the DRP. 279 Rule 87 places no limitation on the ERA's exercise of its discretion in determining DBP's rate of return. Thus, rule 40(3) of the NGR provides, in effect, that the ERA has full discretion to withhold its approval to the DRP element of DBP's access arrangement proposal if in the ERA's opinion there exists a preferable alternative that, inter alia, complies with the applicable requirements of the NGL. 280 It falls to the Tribunal to determine whether: (1) it was open to the ERA, acting reasonably, to prefer its bond-yield approach to the firm-specific, market-based approach advanced by DBP; and (2) the ERA's ensuing estimate was arrived at on a reasonable basis and was the best estimate possible in the circumstances, as required by rule 74(2)(a) and (b) of the NGR. 281 As in WAGN, in this matter the Tribunal must evaluate competing estimating methodologies. For the Tribunal to reject the bond-yield approach of the ERA, which it has already endorsed in principle in WAGN, DBP must demonstrate error in the ERA's bond-yield approach. DBP has not done so. Its submissions are but assertions that its own method for estimating its DRP is better than the ERA's approach. 282 Nothing that DBP submitted has persuaded the Tribunal to change from its in-principle acceptance in WAGN of the ERA's bond-yield approach. To succeed in its submissions that the DBP-specific estimates prepared by AMPCI were to be preferred, DBP had to show that the ERA was in error in its choice of the bond-yield approach. It failed to do so. It is not for the Tribunal to choose between competing methods, unless the choice of one over another is shown to be in error or the one chosen is misapplied. 283 The applicant in WAGN pressed the Tribunal to accept a method (not unlike that proposed by AMPCI for DBP) that adduced evidence from the market of what in fact it would be charged by a major financier for debt finance. In other words, ATCO proposed estimating the cost of debt based on market practice and on the likely pricing of debt by a lender. 284 ATCO also expressed the view that the ERA's bond-yield approach did not satisfy rule 87 of the NGR and submitted that its suggested market-based approach would result in a "more considered and pragmatic approach, which recognises current conditions in financial markets". 285 The Tribunal rejects what may be called the "direct market evidence" approach of DBP that was based on its enquiries of "an experienced capital markets adviser" as to what would be charged for debt finance to a provider of the relevant reference services. As perceived by the Tribunal, the direct market evidence approach as proposed by DBP in this matter is not transparent and is highly firm-specific. Even if all the relevant assumptions on which the estimate was based were known, it would not be possible to easily replicate such an estimate, as different advisers take different views of market conditions and these conditions and opinions can change over quite short periods. A regulated entity seeking to advance a direct market evidence approach must be prepared to address what the Tribunal perceives as the deficiencies in DBP's approach in this matter. That is, an entity must be prepared to provide a reliable and objective estimate of the cost of debt which is forward-looking and which is commensurate with prevailing conditions in the market for funds and the risks involved in providing reference services and which, most importantly, is verifiable by the regulator. 286 Rule 74(2) of the NGR requires the estimation of a cost of debt which has been arrived at on a reasonable basis, and which represents the best estimate possible in the circumstances. The method advocated by AMPCI does not meet this requirement. 287 Accordingly, the Tribunal discerns no incorrect exercise of discretion or unreasonableness in all the circumstances of the ERA's decision making in relation to its development of the bond-yield approach to estimate the DRP. 288 Having determined its method for selecting a relevant sample of firms from which to estimate the DRP, the ERA then proceeded (Final Decision [563ff]) to determine a value for the DRP by means of the statistical approach it used in the ERA ATCO Decision which the Tribunal found to be in error. 289 It is not necessary for the Tribunal to re-state in detail its criticisms in WAGN of how the ERA quantified its estimates of the DRP. The ERA in the present matter followed the same statistical procedures assessed in WAGN. The Tribunal's assessment of these procedures in WAGN at [168] to [187] apply equally to the ERA's calculation of DBP's DRP in the Final Decision at [561] to [569]. 290 The Tribunal accepts that the ERA's bond-yield approach is a valid one. However, the ERA chose a simple averaging approach when considering the various categories of estimates for the DRP which the bond-yield approach produced, in order to establish a single value for the DRP. It is in that averaging process where the ERA fell into error, not in its development of the bond-yield approach. The Tribunal has previously stressed the importance of choosing the most appropriate method of averaging. The averaging method adopted by the ERA is flawed given the unexplained and clearly subjective implicit differential weights assigned to the bonds of some companies in its sample. The ERA's estimate of a value of the DRP for DBP based on such a methodology in the circumstances constitutes an incorrect exercise of its discretion, or alternatively was unreasonable. 291 DBP focussed its DRP submissions on the ERA's bond-yield approach to estimating this parameter rather than its flawed averaging. Senior counsel for DBP accepted that the flaw in the averaging was not advanced before the ERA or by DBP in its written submissions to the Tribunal. This acknowledgment gave rise to an issue whether, having regard to s 258(2) of the NGL, the Tribunal is at liberty to find error in the ERA's averaging. 292 Section 258 of the NGL provides: (1) An original decision maker whose decision is the reviewable regulatory decision being reviewed under this Division may, in the review, raise- (a) a matter not raised by the applicant or an intervener that relates to a ground for review, or a matter raised in support of a ground for review, raised by the applicant or an intervener; (b) a possible outcome or effect on the reviewable regulatory decision being reviewed that the original decision maker considers may occur as a consequence of the Tribunal making a determination setting aside or varying the reviewable regulatory decision. (2) A party (other than the original decision maker) to a review under this Division may not raise any matter that was not raised in submissions in relation to the reviewable regulatory decision before that decision was made. Counsel for the ERA submitted that section 258(2) confines the ground of review to issues that were raised before the ERA. Asked whether the issue of the ERA's averaging was something that cannot be looked at by the Tribunal, he replied: "Yes. If it wasn't agitated, that is one example … that … properly applying 258(2) it would not be a matter that would be the subject of review as an error unless it was raised as an error previously." 293 Addressing the issue of whether DBP's focus on the process of the ERA's bond-yield approach method of estimating the DRP would be a sufficient foundation for a finding by the Tribunal that the outcome of the process was flawed because of the averaging, the ERA's counsel submitted it would take the word "matter" in s 258(2) to a level of abstraction that goes too far. In support of his submission he suggested that the history of regulatory access regimes illustrated a limiting and narrowing of the Tribunal's jurisdiction over time, the purpose of which was akin to the purpose underpinning s 258(2), namely, to make decision making more efficient and improve its quality at the first instance by bringing forth the debate at that level. Within that context, he submitted, s 258(2) is performing a very important task and ought to be given work to do, admittedly at a practical level, concluding that what is a practical level in any given instance may, of course, be the subject of debate but the section should not be overridden by an excessive abstraction of "matter". 294 The Tribunal at [27] above briefly adverted to the sort of problem which is now raised. At one extreme, one might take the view that s 258(2) is intended to confine the review to the particular precise and confined argument put by an applicant to the regulator on a particular topic. At the opposite extreme, one might take the view that s 258 enables an applicant to raise entirely fresh and unconsidered grounds to review a decision on a topic provided the applicant put any argument on the topic, even if it did not pursue that argument. It becomes a question of a sensible constructional choice: neither of those extremes is likely to have been intended by the relevant legislature. 295 The proper scope and effect of s 258 is likely to be determined on the basis of a series of decisions by the Tribunal over time. However, some things are clear. 296 Section 258 must be approached in its context in the NGL. It operates only where leave to apply for review has been given. Leave to apply for review is subject to constraints: the grounds for review are confined by s 246, and must give rise to a serious issue or issues: s 248, and must overcome the revenue threshold: s 249. It is not relevant for present purposes to refer to the procedural constraints in ss 247, 250 and 251. 297 Subject to an intervener raising a new ground of review under s 256, the regulator is then confined by s 258 to a "matter" relating to a ground of review or raised by the applicant in support of a ground of review, and the consequences or possible consequences of a particular decision of the Tribunal. The applicant is not entitled to raise any "matter" that was not raised in submissions in relation to the reviewable regulatory decision before that decision was raised. The word "matter" is used in both s 258(1)(a) and in s 258(2). It should be given the same meaning in each subsection: Registrar of Titles (WA) v Franzon (1975) 132 CLR 611 at 618 per Mason J. 298 Under ss 75-77 of the Constitution, the jurisdiction of the High Court and other Courts exercising federal jurisdiction is expressed by reference to the word "matter". It has been explained to mean the subject matter for determination in a proceeding, where there is an immediate right, duty or liability to be established or determined: see eg Re Judiciary and Navigation Acts (1921) 29 CLR 257 at 265; Philip Morris Inc v Adam P Brown Male Fashions Pty Ltd (1981) 148 CLR 457 at 507-509 and Fencott v Muller (1983) 152 CLR 570. 299 Of course, it would not be appropriate simply to adopt that expansive meaning. But it may be taken to indicate that "matter" in s 258(1) and (2) extends to the controversy or thing in dispute, so long as it is further confined (in the case of the regulator) to a controversy or thing in dispute that relates to a ground of review or relates to a matter raised in support of a ground of review, and (in the case of an applicant) to a controversy or thing in dispute that was raised in submissions by the applicant. 300 In EAPL, the issue arose as to whether the regulator (in that case the ACCC) was prohibited by s 39(5) of the Gas Pipelines Access (South Australia) Act 1997 (SA) (GPA Act) from taking a position on the method of depreciating optimised replacement cost which was a departure from its previous position. Section 39(5) of that Act is in much the same terms as s 261 of the NGL, so the issue ventilated was in a somewhat different context. The Tribunal (Gyles J, Deputy President and Members Mr R Davey and Ms M Starrs) said at [9]: The issue has arisen in the course of the review and arises directly from the decision which is being reviewed. On the other hand, there is much to be said for the submission that the sub-section as a whole is aimed at restricting the substance of the limited review on the part of the Tribunal to "matter" (meaning subject matter) raised before the decision under review was made. That view would prohibit consideration of any substantive issue arising thereafter even though it may be relevant to, or arise out of, the decision being reviewed. 301 However, that question was not finally resolved as the Tribunal at [10] declined to consider a substantive change in the regulator's position, as being contrary to the purpose of a review, having regard to the range of interests at stake in the initial regulatory decision: see at [11]. 302 Epic Energy also concerned the GPA Act. Section 39(2)(b) is in terms relevantly the same as s 258(2) of the NGL, so it is more directly on point. The Tribunal (Cooper J, Deputy President, and Members Professor D Round and Ms M Starrs) said at [24]: Section 39(2)(b) limits the matters to which recourse may be had to those that may be identified in the submissions which, in fact, were made prior to decision. The matters include the subject matters raised, the issues raised and the materials relied upon in support of the position or proposal put forward in the submission as being relevant to the decision being made. Thus, if any matter, whether by way of argument or evidentiary material, cannot be identified as broadly arising out of a matter fairly raised in the submissions to the relevant Regulator before the decision under review was made, it will not be permitted to be raised in the review. This is not to say that a reformulation of an argument or contention previously put to the relevant Regulator on material which was before it before the decision was made would be excluded. A similar comment is made by the Tribunal in Energy Australia at [316f] 303 As appears in [26] of Epic Energy, the issue under consideration was whether the applicant should be allowed to rely on "new matters including expert opinion evidence or material of a rebuttal nature" which were not before the regulator. Thus, the useful observations of the Tribunal in context addressed a somewhat different set of circumstances to the present issue. Order 2 of the Ruling in that decision confirms that. Here, the issue is not whether fresh evidentiary material could be adduced by DBP to support its review. 304 Ergon (No 6) relevantly concerned s 71O(2) of the NEL, which is also in terms alike to s 258(2) of the NGL. It focused upon the word "submissions". The Tribunal (Middleton J, Deputy President and Members Mr R Davey and Mr R Shogren) said at [43] that "submissions" means submissions made during the process leading to the reviewable decision. That is, the Tribunal adopted a purposive approach to the interpretation of that provision. 305 The Tribunal adopts the purposive construction of s 258(2) referred to in the decisions discussed. The process leading to an access arrangement decision by a regulator involves a series of steps designed to secure the identification of contentious issues and then resolution by the regulator in the light of the material and submissions of the covered pipeline provider and those entities whose interests may be affected by the decision. In practical terms, it is a matter for assessment by the Tribunal in the particular circumstances whether the issue which is in contention is one where, as a result of that process, an informed decision has been made by the regulator and which is the subject of the review. 306 In this matter, the issue which is presently the subject of contention is whether the ERA selected a correct input for the DRP for the purposes of the modelling under rule 87(2). There were quite different proposals nominated by DBP on the one hand, and utilised by the ERA on the other, about which was the correct or preferable approach. The Tribunal has found that the approach of the ERA was a proper one. Among the concerns of DBP for urging its proposition to the ERA was the desirability of satisfying rule 87(1) and the national gas objective. It submitted that the ERA's approach would not achieve that outcome, both generally and because some of the inputs (including the DRP) were not satisfactory. The Tribunal has found that the ERA's approach is capable of achieving that outcome, but that in a limited respect it did not do so. The limited respect is the value of the DRP used by the ERA. 307 Viewed in that way, the Tribunal is satisfied that the submissions of DBP raised the matter about the quality of the steps taken by the ERA to ensure its approach led to a reliable input for the DRP in the modelling prescribed by rule 87(2). That matter is one which other entities whose interests may be affected also had the opportunity to make submissions about to the ERA. 308 Accordingly, the Tribunal takes the view that the issue under consideration is properly capable of consideration under s 258(2). It does not therefore need to consider the alternative route by which the issue may have been eligible for consideration. That route involves identifying the ERA as contending under s 258(1)(a) that its approach to ascertaining the correct DRP, including the quality of the data it used and the method of assessing that data to identify the correct input value of the DRP, is correct. That would then support the proposition that the modelled outcome under s 87(2) is itself a proper and reliable one. So far as the Tribunal is aware, apart from the decision in EAPL, s 258(1)(a) and its scope is yet to be carefully considered. 309 For the reasons stated in the preceding paragraphs, the Tribunal is of the opinion that DBP failed to show error on the part of the ERA in deciding to base its determination of the DRP on its bond-yield approach, an approach recently endorsed by the Tribunal in the WAGN decision. 310 Accordingly, in terms of s 259(2)(a) of the NGL, the Tribunal affirms the ERA's decision to use its bond-yield approach to determine DBP's DRP. 311 Also, as indicated above, the Tribunal finds that the ERA's choice of a value for the DRP based on its averaging procedure constitutes an incorrect exercise of its discretion, or alternatively was unreasonable. 312 The Tribunal is not in a position to determine the best possible single estimate for the DRP using the bond-yield approach. The ERA is the appropriate body to undertake the necessary revisions, after receiving fresh submissions from DBP confined to how best to derive a value for the DRP using the ERA's bond-yield approach. 313 Accordingly, in terms of s 259(2)(b) of the NGL, the Tribunal remits the matter back to the ERA to determine a value for the DRP using its bond-yield approach having regard to submissions from DBP confined to the averaging issue, each of the ERA and DBP having liberty to apply for further directions on the issue.

Debt Raising Costs 314 DBP's contention on this matter, as senior counsel explained, is pursued only if the Tribunal were to reject (as it has done) DBP's contention on the appropriate DRP. That is because DBP's contention on the appropriate DRP incorporated the current costs of issuing debt. 315 This alternative contention depends also on the Tribunal accepting as correct the contention of DBP about the proper construction and application of rule 87(1) and (2). 316 The starting point for the claim for debt raising costs is the assertion that they are additional to the DRP as part of the cost of debt. They include underwriter fees, upfront fees, other transaction costs, any cross currency swaps, credit margin on AUD swaps, and borrower's advisor fees. AMPCI provided the expert advice on these costs to DBP. AMPCI initially assessed them as ranging between 44 and 65 basis points per year, and after the Draft Decision as 31 to 45 basis points per year. 317 In the Draft Decision, the ERA allowed 12.5 basis points as part of the cost of debt, based upon the 2004 ACG study and on Professor Handley's report to the AER: A Note on the Completion Method, April 2010. 318 In the Final Decision, the ERA adhered to that view. It said at [576] that no persuasive evidence in support of the claim of 31 to 45 basis points had been given, nor (after the Draft Decision) had the revised claim - reduced by some 30% - been substantiated. Hence at [578] it maintained its reliance on the two sources referred to in the Draft Decision, which it noted reflected an amount widely used in the past by other regulators. 319 The ground of review is based on s 246(1)(a) and (b), namely the assertion of material errors of fact in the Final Decision at [576] and [578], because: (a) the two AMPCI reports "April 2011 (Update) - Cost of Debt Summary Paper" and "April 2011 Draft Determination - Issues Paper" do provide persuasive evidence in support of debt raising costs in the range of 31 to 45 basis points, and showed what would be experienced in prevailing conditions in the market for funds for a large regulated entity operating in Australian markets with a capital base in excess of $2 billion; and (b) the ERA's reliance on the ACG report of December 2004 was misplaced because it was obsolete and its reliance on the Handley report was misplaced because it was irrelevant. 320 DBP's submissions point out that AMPCI used its own knowledge and experience to build the costs of borrowing as a series of explicitly identified costs to be added to the lender's base rate. It says also that the decline in its estimate between April 2010 and April 2011 reflected the change in current market conditions over that period. It points out that that explanation was proffered to the ERA in its Submission 55, Attachment 8. 321 Apart from emphasis on the relative contemporaneity of the AMPCI reports, DBP has not explained why - apart from being a 2004 report - the ACG report is "obsolete", or why the Handley report is not relevant. It also says that, if the ERA had doubts about the "veracity" of the AMPCI reports, it should have sought further expert advice. 322 The ERA's short response to these contentions is that the 12.5 basis points it provided for in the DRP rate used in the Sharpe-Lintner CAPM modelling for debt-raising costs involved no error of fact. It was an available factual finding as a step in the selection of that input. It had evidentiary material to which it referred, and on which it could reasonably have made that finding. 323 Australian Competition and Consumer Commission v Australian Competition Tribunal (2006) 152 FCR 33 (ACCC v ACT) addressed the meaning and operation of s 246(1)(a) and (b) of the NGL. The Full Court (French, Goldberg and Finkelstein JJ), when discussing the review function of the Tribunal under s 39 of Sch 1 to the Gas Pipelines Access (South Australia) Act 1997 (SA) said at [171]-[172] of the ground of review of error in the findings of fact: Findings of facts may include findings of the following kind: 1. The existence of an historical fact being an event or circumstance. 2. The existence of a present fact being an event or circumstance. 3. An opinion about the existence of a future fact or circumstance. The ACCC's function under the Code involves assessment not only of historical and present facts but also of expert opinion on various matters relevant to the fixing of a Reference Tariff. The term "findings of fact" should be interpreted broadly enough to be meaningful in relation to the function of the ACCC under review. It should encompass opinions formed by the ACCC based upon approaches to the assessment of facts or methodologies which it has chosen to apply. The question of what constitutes a finding of "fact" varies according to the statutory context in which that word or like words are used. … In reaching findings of fact in this broad sense the ACCC will necessarily make choices of a discretionary character as was pointed out in Re Michael. An example is the choice between permitted methodologies for the calculation of Total Revenue mentioned in s 8.4 of the Code. Such a choice is not a finding of fact. Nor is a finding of fact in error because it is based upon the use of one methodology rather than another. The relative weight to be given to the factors set out in s 8.10 is also a matter of discretion rather than a finding of fact which can be impugned as such. That decision was reversed on appeal in the East Australian Pipeline v Australian Competition and Consumer Commission case, but not on that issue. 324 In Application by ActewAGL Distribution [2010] ACompT 4 (ActewAGL), the Tribunal said at [31] that a "fact" is something that exists independently of its acknowledgment in the mind of the perceiver. It based that comment on the dictionary definition as something capable of being experienced or perceived and known to be true. It may not be the case that a fact includes the statement of opinion by an expert (unless the issue is whether or not that opinion was stated). It is not clear to the Tribunal as presently constituted that the Full Court in ACCC v ACT said that was the case. 325 However, an expert opinion may be material relevant to a fact - whether it be the speed of a motor car or (as here) the costs likely to be incurred by a substantial regulated entity in securing funding for its operations during a regulatory period. The Tribunal considers that the kinds of fees which might be incurred, or would be incurred, and the range of fees which might be incurred, or would be incurred, by DBP as debt raising costs are matters of fact. The expert evidence is material which informs the answers to those issues and so informs the decision as to the fact or facts. Almost inevitably, the more refined and detailed that factual analysis is, the greater the difficulty in showing that any error in the factual findings is material. No doubt that is why, both when seeking leave to apply for the review and on this application, DBP focused on the more general and complex conclusionary fact about the allowance to be made for debt raising costs as part of the DRP to be input into the modelling. 326 It is not necessary in this matter to explore any nuances in the meaning of "error of fact" in s 246(1) of the NGL: see also Application by ElectraNet (No 3) at [67]. It is clear enough that a finding of fact by the ERA is not shown to be erroneous simply because there is material which could support a different finding of fact. Nor is a finding of fact by a regulator shown to be erroneous simply because the Tribunal might, if it considered the material afresh, prefer to make a different finding of fact. The reference to "error" clearly requires more. In some instances, the error will be apparent: there will be no material to support the finding, or the only material will support a different finding. However, where the finding is a complex one, that is one which involves the assessment of expert opinion material or conflicting material or (as here) conflicting expert opinion material, an error of fact is not shown simply because one expert opinion or set of opinions has been preferred over another expert opinion or set of opinions. Something more will be required, before the Tribunal is satisfied that there is an error of fact (and, necessarily, a material one) on the part of the regulator. 327 The Tribunal has considered the two reports of AMPCI, as well as the ACG report and the Handley report. They do not lend themselves to ready comparative analysis. The ERA was aware that the ACG report was some years old, but there is nothing in it which specifically suggests that it is "obsolete" or not relevant to the issue of debt raising costs. The Handley report is contemporaneous, as are the two AMPCI reports. They reach different conclusions. Their underlying factual premises appear to be somewhat different. 328 Those matters do not, however, go far enough in this matter to enable the Tribunal to conclude that the preference of the ERA for the view expressed by ACG and by Professor Handley involves an error of fact about the correct amount to be allowed for debt raising costs. Put shortly, whilst the differences are clear, the material does not persuade the Tribunal that the ERA's decision on this topic was an error of fact or facts. 329 An alternative submission, noted above, is that the ERA should have sought further comment from AMPCI, or from another expert, on the differences between the experts' views and presumably sought to better clarify the correct factual premises for the expressions of opinion on debt raising costs. The Tribunal is not persuaded that the fact that the ERA made its decision on the material it referred to itself involved any reviewable error in terms of s 246(1)(c) or (d) of the NGL. 330 For these reasons, the ground of review concerning the DRP, and more particularly the subset of that ground concerning debt raising costs, is not made out.

Regulatory Consistency 331 It is necessary to deal only briefly with this topic. 332 In its application for leave to seek review of the Access Arrangement Decision, DBP raised as one concern that reviewable error on the part of the ERA could be established by it having placed reliance on the decisions of other Australian regulators in its consideration of the MRP, gamma, and the DRP as inputs into the WACC. It says the views of other regulators are not relevant to prevailing conditions in the market for funds and the risks involved in providing reference services. 333 In considering the separate submissions on those three topics, the Tribunal has taken account of that contention. It is correct that the Final Decision variously refers to the desirability of regulatory consistency. Indeed, regulatory consistency would be expected on many issues because the nature of the issues are common, and the particular circumstances not idiosyncratic. The Tribunal regards regulatory consistency as a laudable objective, provided the particular regulator (in this case the ERA) independently fulfils its decision-making functions and responsibilities. Each regulator must do so in the context of the particular applicable legislation, and in the context of the particular issue and relevant material on that issue. The NGL under the NGA WA Act, the National Gas Law and the NGR are in most respects the same. It is not therefore surprising that the ERA should be aware of decisions of the AER, and vice versa, on particular provisions which have to be addressed. It is to be expected, in such circumstances, that experienced and well qualified regulators would also reach similar conclusions on such matters. It is to the benefit of providers of regulated services, the users of those services, and the community that - where appropriate - regulatory consistency should exist. 334 In this matter, the Tribunal considers that the ERA has properly fulfilled its decision-making functions and responsibilities. It independently reached its findings, exercised its discretion where required, and expressed its conclusions. 335 In the Final Decision, on the MRP issue, the ERA explained at considerable length why it reached its conclusion at [512] and [515]. It referred to the fact that other Australian regulators, including the AER, had reached a similar conclusion at [513]-[514], but its reasons indicate that those references did not dictate or direct an outcome which the ERA had not independently arrived at. 336 In the Final Decision, in deciding the value for gamma, the ERA referred to "regulatory certainty" at [533] and [537] when addressing the payout ratio (F) and theta values respectively. Again, the ERA's reasons overall indicate that it had considered the relevant material and reached its conclusion on that material. It expressly said so at [532] and in its Draft Decision at [664]-[669], noting that (not surprisingly) its views were consistent with those of the AER in relation to the value of F, and at [537] and in its Draft Decision at [678]-[682] in relation to the value of theta. As appears at [537] in the Final Decision, it adopted the expert views in a particular report as the foundation for its conclusion. 337 As to debt raising costs, the ERA at [567]-[579] explains why it reached its conclusion and did not act on certain other material before it (which it described as not persuasive). It then added in [578] that, in the absence of persuasive material to the contrary, the allowance it had decided upon of 12.5 basis points "provides regulatory certainty, given that amount has been widely used in the past by Australian regulators". That passage does not suggest that the ERA had not independently considered the material and reached its conclusion, or that its independent conclusion was different from (and should defer to) the conclusion of other regulators. The fact that the ERA applied its independent judgment in the light of other regulatory approaches is confirmed by its decision in relation to the DRP: see in particular its observations in the Draft Decision at [548]. In fact, the ERA noted in the Final Decision at [578 fn 234] that one regulator had allowed a higher figure for debt raising costs. 338 The Tribunal therefore does not accept this ground of review has been made out.

National CPI or Perth CPI 339 Required Amendment 5 of the Final Decision requires that any inflation escalation applied in the calculation and subsequent annual adjustment of reference tariffs be based on actual or forecast values (as appropriate) of the Australian Bureau of Statistics (ABS), All Groups, Eight Capital Cities CPI (the National CPI). 340 In the Final Decision at [176], the ERA said that, in the determination and application of reference tariffs, inflation is applied to shelter investors in the DBNGP, as far as is practical, from the effects of inflation on the real value of the asset. Since the objective is preservation of the value of assets (which the ERA refers to as Financial Capital Maintenance) the relevant measure of inflation is one which reflects the prices of the goods and services which those investors could purchase with distributions from the asset. 341 DBP contends that the ERA, rather than accepting DBP's proposal to use the All Groups, Perth CPI published by the ABS (the Perth CPI), incorrectly concluded that a national measure of inflation, namely the National CPI, must be applied in: (1) the roll forward calculation of the capital base so that the capital base at the commencement of the Revised Access Arrangement period is determined in real dollar values at 31 December 2010; (2) calculating the amounts to be added to total revenue under the incentive mechanism that applied for the access arrangement for the DBNGP for the period 2005 to 2010 (Prior Access Arrangement), so that the amounts are determined in real dollar values at 31 December 2010; and (3) the subsequent annual adjustment of reference tariffs over the course of the Revised Access Arrangement period. 342 In its revised access arrangement proposal, DBP stated financial information relating to the calendar years 2005 to 2010 expressed in real terms as at 31 December 2010. In doing so, DBP escalated the prior year dollar values by the Perth CPI. DBP also used the Perth CPI as the inflator in the reference tariff variation mechanism. 343 The ERA approved DBP's methodology to provide financial information in real dollar values as at 31 December 2010. However, the ERA did not agree with the use of the Perth CPI as the appropriate inflator. It substituted the National CPI. It determined that use of the National CPI complied with the requirements of the NGL and was preferable to DBP's approach. The ERA explained its reasons for preferring the National CPI when converting nominal dollar values to real values as at December 2010 for the financial model in the Final Decision at [170] to [179]. 344 The ERA stated in the Draft Decision that it was not satisfied that DBP had adopted a consistent treatment of inflation in its financial calculations, even though they satisfied the requirements of rule 73 of the NGR. It considered that: (1) the use of escalation factors based on the Perth CPI is inconsistent with the rate of return applied in the calculation of Total Revenue; and (2) the rate of return is estimated using a forecast of inflation for the Australian economy, consistent with an implicit assumption made in determination of the rate of return of the DBNGP that the DBNGP is being financed by Australian investors at [105]. 345 It therefore undertook calculations of total revenue in real terms with real values of financial information calculated by applying escalation factors derived from December quarter values of the National CPI, and given the passage of time between the submission of the proposed access arrangement and the Draft Decision, the ERA also undertook financial calculations using values of financial information expressed in dollar values at 31 December 2010. 346 The ERA therefore escalated the capital base values from the initial capital base for the previous access arrangement period for inflation so that the capital base was expressed in 31 December 2010 dollar values. In escalating values of the capital base for inflation, the ERA applied escalation factors derived from the National CPI, rather than the Perth CPI as DBP sought. Similarly, in relation to the incentive mechanism from the prior access arrangement period, the ERA re-calculated amounts under the incentive mechanism applying December quarter CPI values from the National CPI. It is common ground that the ERA made calculations of total revenue in real terms with real values of financial information calculated by escalation factors derived from December quarter values of the National CPI. By requiring Required Amendment 5 (conforming capital expenditures for 2005 to 2010 to be those indicated in Table 15 of the Draft Decision) and Required Amendment 12 (the reference tariff charges for T1 reference services for calendar year 2012 be as stated), the ERA required calculation of all real values using the National CPI. 347 In addition, in the Draft Decision, in relation to the reference tariff variation mechanism, the ERA considered that the CPI values applied in the annual variation of reference tariffs during the access arrangement period should consistently be the National CPI. Accordingly, the ERA required the reference tariff variation mechanism to be amended to base inflation escalation of reference tariffs on movements in the National CPI: Draft Decision at [971]. 348 On 18 April 2011, the DBP submitted its revised proposed access arrangement. It maintained that the calculations of total revenue and reference tariffs in dollar values at 31 December 2010 should be escalated based on the Perth CPI. 349 In a further submission of 20 May 2011, DBP contended that: (1) the capital expenditures which it had made in the past, including the expenditures made during the period 2005 to 2010, and which had been escalated so that the opening capital base for the period 2011 to 2015 is expressed in constant December 2010 prices, were expenditures largely incurred in Western Australia for the purpose of providing natural gas services to consumers of natural gas in Western Australia; (2) these capital expenditures were expenditure on labour, materials and services; (3) its operations are in Western Australia, and all of its labour costs, irrespective of whether they are capitalised or charged to operations, are incurred in Western Australia; and (4) data drawn from its procurement and materials management system indicated that, in 2010, expenditures with the fifty largest suppliers of materials and services for the DBNGP were largely (72%) incurred from Western Australian suppliers and overall almost entirely (89% including WA) from within Australia. 350 It also pointed out that during the period 2005 to 2010, prices were rising a little more rapidly in Western Australia than they were nationally, as seen by comparing the National CPI with the Perth CPI. 351 DBP also submitted that, to the extent that reference tariffs for the period 2011 to 2015 under-recover the costs which it incurs during that period, they would be an inducement for inefficient (inadequate) investment in the DBNGP, and for inefficient (excessive) use of natural gas services by consumers of natural gas. DBP submitted that the Draft Decision, if maintained on this topic, would not promote economic efficiency, and would be inconsistent with the revenue and pricing principle in s 24(3) of the NGL that the service provider be provided with effective incentives to promote economic efficiency and also would not promote efficient investment in, and efficient operation and use of, natural gas services for the long term interests of consumers of natural gas. It maintained the contention that use of the National CPI, instead of the Perth CPI, in the roll forward of the capital base would not satisfy the national gas objective. 352 The ERA in its Final Decision did not accept those arguments for use of the Perth CPI rather than the National CPI as a basis for inflation escalation in financial calculations of total revenue and reference tariffs. The ERA maintained the requirement that inflation escalation in financial calculations incorporate escalation factors based on actual and forecast inflation as measured by the National CPI. 353 As noted, the Final Decision was first delivered on 31 October 2011 and amended on 22 December 2011. Between those dates, on 13 December 2011, DBP made Submission 73 to the ERA. Part 4 of this submission responded to the reasons of the ERA to support the continued use of the National CPI. The amendments to the Final Decision did not involve any substantive change to the ERA's conclusion on this topic. The effect of the Final Decision is that reference tariffs for each reference service are calculated by using financial information which is provided in real terms: (1) with all real values expressed in dollar values of 31 December 2010; and (2) which have been calculated by applying escalation factors derived from December quarter values of the National CPI. 354 The common starting point for consideration of this issue is that one of the building blocks is a return on and depreciation of, the projected capital base for each year of the access arrangement period. The opening capital base is to be determined under rule 77(2) of the NGR. In addition, it is a common starting point that reference tariffs for the DBNGP should be determined using the real pre-tax WACC: see Final Decision at [360]. 355 There is also no issue about the basic way in which the opening capital base is reset to the start of the proposed access arrangement period, or how it is then projected for each year of that period. 356 In that light, DBP maintains for the reasons it put to the ERA (and referred to above) that the revaluation of the capital base should be done applying the Perth CPI. 357 It says at [282] of its outline of submissions: If the capital base of the DBNGP is to be a replacement cost valuation of the assets which comprise the pipeline system, the capital base which was set in 2005, the capital expenditures during the period 2005 to 2010, and the amount of depreciation, must all be re-valued at the prices at which those assets could be replaced. Those prices are prices applying in Western Australia, and not prices elsewhere. A suitable index of prices to be used for the purpose of revaluation is the Perth CPI. It is not the National CPI. A similar contention was advanced by ATCO and addressed by the Tribunal in WAGN at [189]-[197]. 358 The Tribunal maintains the view expressed in WAGN for the same reasons. The Final Decision at [176]-[178] explain the reasons for the conclusion of the ERA. In essence, the Tribunal agrees with them. 359 The fixing of the opening capital base for a later access arrangement period under rule 77(2), in the view of the Tribunal, does not require that the opening capital base be determined on replacement value. What is required is to apply an inflation measure to past capital expenditure to adjust it to real value at the commencement of the new access arrangement period so that the present day value of that capital expenditure shelters investors from the effects of inflation in the real value of their assets. Consequently, as the capital base is expressed in Australian dollars, the appropriate inflator may be one that reflects the Australian economy generally or the purchasing power of the Australian dollar generally. 360 In the view of the Tribunal, because the exercise involves putting present day value to past expenditure, it is consistent with the national gas objective and the revenue and pricing principles to account for the opportunity cost of funds invested in the past by expressing those amounts in present day terms applying the National CPI. In that way, the relevant capital base remains consistently measured, during a range of access arrangement periods, and the ERA is consistently applying an inflator to the various parameters to which it is required to adjust past expenditure or make other estimates in the course of its review. 361 The Tribunal notes that the depreciation rate required by rule 77(1)(b)(iii), for example, would not be altered because the depreciating asset was located in Western Australia rather than in some other part of Australia. That observation also applies to the forecast depreciation during the access arrangement period on the projected capital base under rule 78(c). The forecasts for future capital expenditure and operating expenditure are based upon a national inflator measure. Accordingly, once the focus is removed from simply determining the replacement cost of the then capital base, the Tribunal is of the view that the ERA's approach is not shown to have involved error of fact, or any miscarriage in the exercise of its discretion to adopt the National CPI for that purpose. 362 Contrary to the submission of DBP, the Tribunal does not consider that there is in rules 77 and 73 any indication of a different position. It is correct that, if there has been no previous access arrangement, the opening capital base under rule 77(1) starts with the cost of construction of the pipeline and associated costs. It would have been available to specify in the NGR that the opening capital base for a subsequent access arrangement period would be the cost of constructing a new capital asset. It does not do so. 363 The adjusted capital base for subsequent access arrangement periods under rule 77(2) will support the fixing of the rate of return on equity, as well as the fixing of the appropriate tariffs - balancing the interests of consumers and of the service provider on its own behalf and for its investors, in accordance with the national gas objective and the revenue and pricing principles. It is not intended that, if replacement costs, for example, were to be dramatically higher than the actual cost adjusted for inflation (the assumption underlying DBP's case on the point is that Perth CPI will better approximate replacement cost), then DBP and indirectly its investors should receive significantly higher tariffs at the expense of consumers. Historical cost adjusted for inflation is different from replacement cost, so use of a higher inflation cost to achieve a figure closer to replacement cost is not logical. The acceptance of a starting point of actual cost when the asset is first constructed would not be inconsistent with the Tribunal's view. Nor does the acceptance of the forecast conforming capital expenditure during the access arrangement period as part of the projected capital base under rule 78 lead to a different conclusion. 364 Rule 73 does not really inform resolution of the issue, except that rule 73(3) requires consistency in the making of calculations (including depreciation) so that may tend to support the Tribunal's conclusion. At a very simple level, if DBP is correct, assets would not be depreciated in a standard accounting way (as to which, see also rule 89(1)(d)) but would be revalued at their replacement (not realisable cost) for the start of each new access arrangement period. 365 For those reasons, the Tribunal does not consider this ground of review has been made out, and affirms the ERA's decision.

Project Management Retainer Fee 366 DBP contends that the ERA erred in deciding that: (1) a project management retainer fee (the PMR fee) was not conforming capital expenditure (capex) in terms of rule 79 of the NGR; and (2) the PMR fee should be removed from DBP's opening and forecast capital base. 367 In 2004 DBP was acquired by a consortium which included the Alinta Group. Also in 2004 an Operating Services Agreement (the OSA) was entered into between DBP and Alinta Asset Management Pty Ltd (AAM), part of the Alinta Group. Pursuant to the OSA, AAM provided asset management, operations maintenance and construction services to DBP. 368 In 2009, AAM changed its name to WestNet Energy Services Pty Ltd (WNES) and the OSA was amended. The amended OSA, dated 10 February 2009, introduced the PMR fee. 369 The amended OSA resulted in DBP resuming responsibility for a range of services relating to the operation and maintenance of the pipeline, with the exception of, relevantly, expansion-related project management services which were to be provided by WNES. 370 Under the amended OSA the relevant fee arrangements were that DBP would: (1) continue to pay a project management fee of 3% of the value of any additional capacity expansion and capital works (the PM Fee); (2) pay WNES an annual PMR fee of $2 million (indexed to CPI) to ensure WNES provided the necessary personnel, corporate systems and procedures to maintain an ongoing capability to provide the project management services irrespective of whether an additional capacity expansion was being planned or undertaken (clause 7.5 of the amended OSA). 371 DBP's Submission 1: Background, 14 April 2010 (Submission 1) to the ERA in support of its revised access arrangement proposal, made the following statements at paragraph 6.7 relevant to the ERA's consideration of the PMR fee in terms of rule 79 of the NGR: In relation to the Project Management Retainer Fee, the Operator considers that this is a prudent and efficient cost for the following reasons: (a) Over the last 5 years, the Operator has incurred significant costs in ensuring WNES has in place systems and procedures to ensure the expansion projects can be delivered on time and on budget and in accordance with the requirements of good corporate governance. (b) The Operator must deliver expansions to pipeline capacity in accordance with the timing requirements of its shippers and prospective shippers. It is therefore important to ensure that these systems and procedures are maintained up to date so that, if a shipper requests additional capacity, the expansion can be delivered on time. (c) If there is likely to be a delay in the commencement of the next expansion project (as is envisaged in the Proposed Revised AA), WNES needs to be incentivised to retain the "corporate knowledge" it and DBP have invested in the various staff engaged in the expansion projects over the last five years. (d) This incentive fee is likely to be significantly smaller than the additional costs that the Operator is likely to incur in bringing a new project management team up to speed with the requirements of an expansion project for the DBNGP. Even if the next expansion project does not occur until 2020, the additional costs would have to be less than $20 million for the retainer fee not to have been efficient. In the context of expansions that have been between $400 - 600 million, this up front cost could deliver significantly greater savings and benefits to the Operator and shippers. 372 Section 2 of DBP's Submission 9: Justification of Expansion Related Capital Expenditure, 14 April 2010 (Submission 9) outlined its obligations to expand the capacity of the pipeline in accordance with: (1) existing standard shipper contracts (SSC) which impose on DBP a commitment until January 2011 to expand the pipeline's capacity for any existing shipper who seeks additional Tl capacity; (2) obligations to the State of Western Australia under a Financial Assistance Agreement which requires DBP to use reasonable endeavours to enter a SSC with a shipper or prospective shipper; and (3) enforceable undertakings given to the Australian Competition and Consumer Commission in October 2004 requiring investment of up to $400m to expand the capacity of the pipeline to meet the known capacity requirements of shippers who enter into a SSC; that expansion to be completed within five years. 373 Submission 9 also referred to the following factors why, in 2009 when the OSA was being amended, DBP considered that the PMR fee was reasonable: (1) there was a likelihood that the continual expansion program that had been undertaken since 2005 would not continue beyond Stage 5B; (2) it was important to ensure that the project manager maintained the relevant personnel with the corporate knowledge and understanding of the pipeline and its operations until any transitioning issues were dealt with; and (3) particularly relevant to DBP's contractual commitments, it was also important to pay a fee that incentivised the project manager to pay certain individuals to remain on standby to commence any further project, should one materialise, in time to complete a business plan and allow an investment decision to be made. 374 DBP's Submission 18: Response to Halcrow Pacific Issues Report/Request of [sic] Information, 18 July 2010, (Submission 18) repeated the submissions it made in Submission 9 and, in response to what Submission 18 described as a report by the ERA's consultant Halcrow Pacific Pty Ltd which sought clarification of why the annual $2 million PMR fee was appropriate, submitted that: (1) an acceptable range of margins for asset managers was from 4.3% to 6.7%; and (2) while the PMR fee was not expressed as a margin per se, the quantum of the fee effectively falls within this range. It based that submission on two reports by NERA - Outsourcing by Regulated Businesses, March 2007 (Attachment 26 to Submission 9); and a report which critiqued Allen Consulting Group's Review of NERA's Benchmarking of Contractors' Margins (Attachment 22 to Submission 9). 375 Halcrow Pacific Pty Ltd in association with Zincara Pty Ltd (the ERA's consultants) provided its assessment of the PMR fee in its Dampier to Bunbury Natural Gas Pipeline Access Arrangement Review - Technical Assessment Draft Review Report, March 2011. That report advised the ERA as follows: Halcrow is not aware of other instances in which a service provider has committed to payment of a retainer fee of such quantum. Halcrow is (anecdotally) aware that retainer fees are paid in some instances, however, these are understood to be at least an order of magnitude less. Furthermore, it is understood that the more usual arrangement is that payment for actual services is initially offset against the retainer fee until the cost of services exceeds the value of the retainer fee. The basis on which the retainer fee is incurred, ie. to ensure that ongoing capability to provide services is maintained, also warrants consideration. The following comments are provided: • DBP has acknowledged that, at the time that the OSA was renegotiated in 2009, "...there was a likelihood that the continual expansion program that had been undertaken since 2005 would not continue beyond Stage 5B". This would appear to limit the need to secure access to ongoing capability; • Whilst DBP contends that it is "... important to pay a fee that incentivised the project manager to pay certain individuals to remain on 'standby' to commence any further project, should it materialise .... This is particularly relevant given the contractual obligations to deliver additional capacity in a given time ...", it is acknowledged that the labour market outlook in Western Australia for the period 2008 to 2020 showed an initial decline in demand over a few years with rapid growth set to resume in the early part of the next decade …; nonetheless, it is expected that the whilst the services secured via the retainer fee would be of a specialist nature such services could be secured from elsewhere in a timely manner should the demand arise and resources are not locally available; and • To provide some context, $2,000,000 per annum (ignoring the impact of escalation) equates to some 200 person weeks @ $250 per hour, ie. 4-4.5 full time persons. Whilst it is apparent that that DBP is contractually obligated to payment of the Project Management Retainer Fee, Halcrow does not consider this expenditure … to be prudent or efficient. (Footnotes omitted) 376 The following paragraphs in the Draft Decision concluded the ERA's consideration of the PMR fee: 235. … the Authority is not satisfied that the project management retainer fee is consistent with the prudence and efficiency requirements of rule 79(1)(a). DBP has not provided information that satisfies the Authority that the payment of the project management retainer fee is necessary to be able to contract for project management services within the required time frames for an expansion of the DBNGP. 236. Moreover, the Authority is not satisfied that the project management retainer fee is a genuine fee for a service or facility to be provided by WestNet Energy Services Pty Ltd, given: • the lack of a detailed specification in the Amended and Restated Operating Services Agreement of any relevant requirements to be met by WestNet Energy Services Pty Ltd in return for the fee; • a view of expert engineering advisors to the Authority that there is a lack of precedent to suggest that the nature and quantum of the project management retainer fee are consistent with common industry practice; and • DBP has forecast no expansion of the DBNGP for the 2010 to 2015 access arrangement period. 237. In the circumstances, the Authority is also concerned that the project management fee may represent a negotiated compensation to Alinta Asset Management and WestNet Energy Services Pty Ltd for the termination of the management fee, rather than a fee for an additional service or obligation under the Amended and Restated Operating Service Agreement. … 240. The Authority is of the view that the amount of the project management retainer fee does not satisfy the prudence and efficiency requirements of rule 79(1)(a) and requires that the amount of this fee be removed from the capital expenditure. DBP has not provided a clear statement of the value of the project management retainer fee in specified nominal or real values. The Authority has therefore estimated the amount of this fee for the three year period 2008 to 2010 as an inflation indexed amount of $2 million in 2004. The amounts deducted (in dollar values of 2010) comprise $2.375 million in each of the three years. 377 DBP responded to the Draft Decision's conclusion on the PMR fee by way of paragraphs [2.109] to [2.133] of Submission 53: Roll Forward of the Capital Base, 20 May 2011 (Submission 53). Those paragraphs may be summarised as restating the ERA's reasoning, repeating some of what had been said by DBP in its submissions leading to the Draft Decision but adding little, if anything, new that might be considered germane to the ERA's findings. 378 In July 2011, in connection with the acquisition by the ATCO Group of the interest in the pipeline formerly held by Alinta, the amended OSA was novated by ATCO Technologies Australia Pty Ltd in place of WNES. The OSA was again amended with the effect that the PMR fee ceased to be payable. DBP did not inform the ERA that the PMR fee ceased to be payable until after the Final Decision. 379 The Final Decision restated the findings that the ERA made in the Draft Decision in relation to the PMR fee and observed: 282. In the draft decision, the Authority determined that costs of the project management retainer fee in the 2005 to 2010 access arrangement period do not satisfy the prudence and efficiency requirement of rule 79(1)(a). DBP has not made any submission to the Authority on this determination. DBP has, however, addressed this matter in relation to forecast capital expenditure. … 285. DBP contends that the project management retainer fee is a genuine fee for service and is efficient, evidenced by: • in 2009 when the project management retainer fee was initiated, there was a likelihood that the expansion programme for the DBNGP would not continue beyond Stage 5B, hence it was necessary to ensure that the project manager retained the relevant personnel to commence any further project; • the Operating Services Agreement does provide that in consideration for the payment of the retainer fee, WestNet must retain the necessary personnel, corporate systems and procedures to maintain an ongoing capability to provide the Project Management Services irrespective of whether an Additional Capacity Expansion is being planned or undertaken; • the project management retainer fee covers an expansive range of services provided by WestNet under the OSA in relation to capacity expansions and capital works, including the retention of all project services, from conceptual design, through front-end engineering design (FEED) studies, planning, approvals, construction, commissioning and final delivery of the projects for operation (and all services to support these activities e.g. human resources management, and financial control); and • project management retainer fees are accepted industry practice in the construction industry. 286. DBP contends that the expense of the project management retainer fee, as an expense incurred under the Operating Services Agreement, is an expense incurred by a prudent service provider acting efficiently given: • it is prudent for the ownership consortium for the DBNGP to have relied on the resources and expertise of one of the members of that consortium to provide services relating to the operation and expansion of the pipeline; • at acquisition, negotiations took place at arm's length and all parties had experience in negotiating major construction and operating contracts; • Alcoa and DUET were commercially motivated to ensure that any fees charged by one member of the consortium were at reasonable levels; • there was no reason, and there continues to be no reason, for either DUET or Alcoa, to have any commercial or other interest in Alinta deriving non-commercial fees for performing services under the Operating Services Agreement, or for the contractual arrangements to be of a nature that are neither efficient nor in accordance with good or accepted industry practice; • the amount for the project management retainer fee is efficient because it covers an expansive range of services provided by WestNet under the OSA; and • DBP has a positive obligation to seek to minimise the capital costs of expansion of the DBNGP under the standard shipper contract. … 289. In the absence of adequate justification of the project management retainer fee, the Authority maintains the determination that the amount of this fee does not satisfy the prudence and efficiency requirements of rule 79(1)(a). 290. In its revised access arrangement proposal and supporting submissions, DBP still does not provide a clear statement of the value of the project management retainer fee during the 2005 to 2010 access arrangement period. DBP does, however, indicate a value of this fee in the forecast of capital expenditure for 2011 to 2015, with this value being a constant real value of $2.255 million per year in dollar values of 31 December 2010. The Authority has therefore deducted this value from the value of stated capital expenditure for each of the years 2008 to 2010. 380 DBP's Submission 73 provided, amongst other things, an unsolicited response to the Final Decision's conclusions on the PMR fee. Submission 73 informed the ERA that the PMR fee had ceased to be payable and repeated some of what appeared in Submission 53. Submission 73 also noted that although the ERA had re-estimated the value of the PMR fee to be deducted from DBP's capex for 2008 to 2010 as $2.255 million per annum and from the proposed forecast conforming capex for 2011 to 2015 as $2.225 million per annum, the Final Decision did not deal with Submission 53 otherwise than by summarising it. 381 Significantly, Submission 73 disputed the years in respect of which the ERA removed the amounts for the PMR fee. In this regard DBP submitted that: 61. Following the making of the Final Decision, but before the ERA made its Final Determination, the Applicant informed the ERA that the annual obligation to pay the PMR Fee, which had commenced only in 2009, had ceased in 2011 Hence, the total amount in question is in fact: (a) If the Tribunal accepts the Applicant's contention that the ERA was in error as to the issue of whether the PMR Fee complied with rule 79(1) - only $6,415,687 .44 ($2010) Year 2008 2009 2010 2011 Total ($2010) ($2010) ($2010) ($2010) ($2010) Amount ERA should have allowed as conforming capital expenditure (as to the PMR Fee) $0 $0 $4,042,921.85 $2,372,766.59 $6,415,687.44

(b) If the Tribunal accepts the ERA's contention that the ERA was not in error as to the issue of whether the PMR Fee complied with rule 79(1) - only $2,604,311.56, as shown in the following table. Year 2008 2009 2010 2011 Total ($2010) ($2010) ($2010) ($2010) ($2010) Amount ERA should have removed from proposed confirming capital expenditure if PMR Fee was not prudent and efficient $0 $0 $4,042,921.85 $2,372,766.59 $6,415,687.44 Amount ERA removed from the capital base in the Final Decision (as referrable to the PMR Fee) $2,255, 000 $2,255, 000 $2,255, 000 $2,255, 000 $9,020,000 Additional amount the ERA incorrectly removed from the proposed conforming capital expenditure (2,255, 000) (2,255, 000) 1,787,921.85 117,766.59 $2,604,311.56

(Footnote omitted) 382 To make good its contentions, DBP must show that: (1) the PMR fee is conforming capital expenditure (NGR: rule 79); and (2) the ERA erred in determining that the PMR fee is not conforming capital expenditure (NGL s 246(1)); and (3) the ERA erred in deducting the PMR fee from DBP's opening and forecast capital base (NGL s 246(1)). 383 Relevantly, rule 79 of the NGR provides: 79 New capital expenditure criteria (1) Conforming capital expenditure is capital expenditure that conforms with the following criteria: (a) the capital expenditure must be such as would be incurred by a prudent service provider acting efficiently, in accordance with accepted good industry practice, to achieve the lowest sustainable cost of providing services; (b) the capital expenditure must be justifiable on a ground stated in subrule (2). … (6) The AER's discretion under this rule is limited. 384 Rule 40 governs the exercise of the ERA's discretion. Rule 40(1) sets out when the ERA has no discretion in its decision making process, rule 40(2) when it has limited discretion and rule 40(3) when it has full discretion. Rule 40(2) provides: (2) If the Law states that the …[ERA's] … discretion under a particular provision of the Law is limited, then the …[ERA] … may not withhold its approval to an element of an access arrangement proposal that is governed by the relevant provision if the … [ERA] …is satisfied that it: (a) complies with applicable requirements of the Law; and (b) is consistent with applicable criteria (if any) prescribed by the Law. 385 Prior to the Draft Decision, DBP provided to the ERA Submissions 1, 9 and 18. Submission 1 explained, amongst other things, why DBP considered the PMR fee to be prudent and efficient. Submission 9 outlined why DBP considered the PMR fee to be reasonable. Submission 18 sought to provide, in response to a request from the ERA's consultants, clarification why the annual $2 million PMR fee was appropriate. 386 It is clear from the Draft Decision that the ERA had regard to the three submissions. The Draft Decision also makes clear the ERA's reasons for rejecting the PMR fee as meeting the prudent and efficient requirements of s 79: see [235] and [236] in the Draft Decision. 387 Submission 53 provided DBP's response to the Draft Decision's findings in relation to the PMR fee. As observed, it did not go much beyond DBP's earlier submissions. As may be seen by reference to the above-quoted passages from the Final Decision, it was considered by the ERA but was not instrumental in changing its findings on the PMR fee. 388 The Draft and Final Decisions do not evince error by the ERA in its consideration of the substance of the submissions leading to those decisions. At most they show a difference of opinion between DBP and the ERA whether the PMR fee is "… capital expenditure … such as would be incurred by a prudent service provider acting efficiently …" in terms of rule 79. 389 Nor do the decisions demonstrate that the ERA erred in exercising its discretion. DBP's submissions may be characterised as seeking to justify the existence and quantum of the PMR fee by reference to the history of the OSA. The Draft Decision should have put DBP on notice of the need for hard information to change the ERA's view of the PMR fee. DBP failed to provide it. It appears from DBP's senior counsel's concession that there was no evidence with which to benchmark the PMR fee against other like fees tends to indicate that such information does not exist. There is no error on the ERA's part in rejecting the DBP's submissions that the PMR fee is consistent with the prudence and efficiency criteria of rule 79. 390 Insofar as Submission 73 is concerned, for reasons articulated elsewhere in these reasons, the Tribunal accepts the submission by the ERA's counsel to the effect that error on its part may not be founded on Submission 73 which came after the Final Decision. 391 DBP submitted that if the Tribunal were to adopt the ERA's findings that the PMR fee does not comply with rule 79, DBP would nevertheless contend that because the PMR fee did not commence being paid until February 2009, the ERA erred in deducting $2.255 million ($2010) for 2008 because there was no amount referrable to the PMR fee included in DBP's proposed conforming capex for that year. 392 The ERA's response was to the effect that it acted on the information available to it at the time of making the Final Decision. Because the issue of it deducting $2.255 million ($2010) for 2008 was only raised by DBP in Submission 73 after the Final Decision, no error was made by the ERA in its Final Decision. However, having had the opportunity to consider the additional information in Submission 73, the ERA indicated that, subject to the Tribunal's decision, the appropriate course might be for DBP to submit a variation proposal to the ERA pursuant to rules 65 and 66, together with appropriate verification of the figures contained in Submission 73, which would enable the ERA to determine total revenue to be adjusted on the basis of DBP's figures. 393 There was some debate whether the deduction should be treated as an error and addressed by the Tribunal along the lines of the error identified by the ERA in respect of the Burrup Extension Pipeline. The Tribunal accepts the submission by the ERA's counsel that the deduction is of a different category. It also accepts his submission to the effect that where, as here, information additional to that put to the ERA before an otherwise error free Final Decision is put to the Tribunal and the information signals a mistake on the part of the ERA, sensible administration dictates that the ERA should correct it in accordance with the NGR. 394 For reasons stated in the preceding paragraphs, the Tribunal is of the opinion that DBP has failed to show error on the part of the ERA in: (1) determining that the PMR fee is not conforming capital expenditure; or (2) deducting the PMR fee from DBP's opening and forecast capital base. 395 Accordingly, in terms of s 259(2)(a) of the NGL, the Tribunal affirms the ERA's decisions in those regards.

Burrup Extension Pipeline Lease 396 As part of the expansion of the DBNGP, known as the Stage 5B expansion, undertaken prior to 2011, the initial part of the DBNGP required expanding by DBP to deliver additional capacity contracted for by shippers. 397 DBP considered a number of options for the design of this expansion. It could have provided the incremental contracted capacity and met its contractual pressure obligations by constructing a compressor station at Mainline Valve No.7. It could have increased the capacity of the pipeline at Compressor Station 1 via the addition of another compressor. It could have installed an additional length of pipe parallel to the then existing DBNGP (known as a "loop") for at least the first 23 kilometres of the pipeline. Finally, it could secure rights to use unutilised capacity on the Burrup Extension Pipeline (BEP), a pipeline owned by the Epic Energy group, constructed in 1998, that runs parallel to the DBNGP for the first 23 kilometres from the Northwest Shelf Domestic Gas Plant through the environmentally and culturally sensitive Burrup Peninsula to Mainline Valve No.7 at Cajaput Well before connection to the Pilbara Energy Pipeline (also owned by the Epic Energy group) which transports gas to Port Hedland. The BEP can, with minor reconfiguration, itself be used as a loop of the DBNGP - by operating valves at Cajaput Well and on the BEP itself. 398 For reasons which it is not necessary to expose, because the ERA accepts that it was an appropriate decision, DBP decided to reach a commercial arrangement with Epic Energy to utilise unused capacity in the BEP as a loop of the DBNGP. 399 In May 2008, DBP entered into a conditional lease agreement with Epic Energy (Pilbara Pipeline) Pty Ltd (Epic Energy) to lease a portion of the capacity of the BEP, namely 150TJ/Day (49% of its total capacity) for a period of not more than 20 years, with options to expand the lease to a capacity of 400TJ/Day and to extend the term of the lease by a further term of not more than 40 years (BEP Lease). 400 In December 2010, the BEP Lease commenced, when all conditions precedent were either satisfied or waived. 401 Under the BEP Lease, DBP leases capacity in the BEP for 20 years on the basis that each of Epic Energy and DBP deals with its own capacity separately, but that operating costs are shared in agreed percentages. DBP will also be the operator of the BEP under the BEP Lease. 402 The BEP Lease qualifies as a Finance Lease, resulting in an accounting practice requiring the recognition of it as an asset in the financial statement of accounts of DBP and a corresponding lease liability of an equal amount in the same year. 403 The terms of the BEP Lease provide for monthly lease payments, paid by DBP one month in arrears from the commencement of the lease term. Paragraph 4.3(a) of the BEP Lease provides that the base rent has been calculated as of 1 January 2008 at a significant monthly amount. Further, paragraph 4.3(b) provides that the base rent will be escalated annually on 1 January each year per the equation provided. 404 It is common ground that it is appropriate to treat the BEP Lease as forecast conforming capital expenditure pursuant to rule 78 of the NGR, and therefore as part of the projected capital base of the DBNGP. 405 There was, and remains a dispute between DBP and the ERA as to the value to be ascribed to that capital expenditure as at 31 December 2010. There was also, and remains a dispute between DBP and the ERA as to the correct period of depreciation. 406 In its Final Decision, the ERA made two determinations that are relevant to the matters in dispute: (1) it determined that the BEP Lease should be valued at the present value of the minimum lease payments, being $17.362 million, plus an agreed amount of initial direct costs in establishing the lease, giving a total capital value of $17.679 million. In deriving that value, the ERA applied Accounting Standard AASB 117; and (2) it also determined that the BEP capacity (which is the subject of the BEP Lease) should be treated as a pipeline asset for regulatory purposes with depreciation over the remaining asset life determined as the assumed life for new pipeline assets (70 years) less the age of the BEP to 2011 - as the BEP was constructed in 1998, the remaining asset life for the purposes of depreciation is 57 years.

Value of conforming capital expenditure 407 Rule 74 of the NGR provides: (1) Information in the nature of a forecast or estimate must be supported by a statement of the basis of the forecast or estimate. (2) A forecast or estimate: (a) must be arrived at on a reasonable basis; and (b) must represent the best forecast or estimate possible in the circumstances. Also, the determination of a value of the BEP Lease is governed by rule 100 of the NGR (requiring, inter alia, the provisions of the access arrangement to be consistent with the national gas objective in s 23 of the NGL). 408 DBP's first contention is that there is an issue between it and the ERA as to whether one only has regard to AASB 117 in determining the value of a finance lease for regulatory purposes. DBP argues that the ERA adopted the erroneous view that one only has regard to AASB 117. 409 DBP's second contention is that the ERA's calculation of the present value of the minimum lease payments under the BEP Lease contained errors. The asserted errors are: (1) calculating the present value of the lease payments using the Base Rent stated in the lease, which is an amount as at 1 January 2008, and making no allowance for the effect of inflation; and (2) not restating the lease payments as real values as at 31 December 2010; and (3) using a nominal discount rate instead of a real discount rate to determine the present value of the lease payments. 410 DBP argues that the ERA should have: (1) restated the Base Rent and the initial indirect costs of the BEP Lease in terms of prices prevailing in December 2010; (2) calculated the present value of the lease payments using a real cost of debt of 4.14%; and (3) not escalated the Base Rent for each year of the term of the lease. 411 DBP does not, in this review application, seek to support the method of valuing the BEP Lease that it had proposed to the ERA prior to the Final Decision. DBP's submissions are directed to establishing error in the ERA's calculation of the present value of the lease payments. 412 The ERA has reviewed both its calculation of the present value of the lease payments and its methodology in the light of those assertions. It has identified an error in its calculation. 413 Paragraph 20 of AASB 117 requires the lessee to recognise finance leases as assets and liabilities: … at amounts equal to the fair value of the leased property or, if lower, the present value of the minimum lease payments, each determined at the inception of the lease. The discount rate to be used in calculating the present value of the minimum lease payments is the interest rate implicit in the lease, if this is practical to determine; if not, the lessee's incremental borrowing rate shall be used. Any initial direct costs of the lessee are added to the amount recognised as an asset. 414 In the Final Decision, the ERA valued the BEP Lease on the basis of the present value of the minimum lease payments. 415 Although paragraph 20 of AASB 117 requires the determination of the present value of the minimum lease payments at the inception of the lease, paragraph 5 of AASB 117 addresses the effect of changes in the lease payments between inception and commencement. It provides that: … a lease agreement … may include a provision to adjust the lease payments for [amongst other things] changes in … general price levels … during the period between the inception of the lease and the commencement of the lease term. If so, the effect of any such changes shall be deemed to have taken place at the inception of the lease for the purposes of [AASB117]. 416 Clause 4.3 of the BEP Lease adjusts the rent payable under the lease for the effects of inflation. As such, it provides a mechanism for adjusting the rent between the inception of the lease and the commencement date. The ERA acknowledges that, in the calculation of the present value of the minimum lease payments, it should have adjusted the Base Rent ([redacted] as at 1 January 2008) by the CPI formula in clause 4.3 to derive an adjusted Base Rent as at the commencement of the lease in December 2010. The ERA did not take that step and acknowledges that the failure to do so was an error in applying AASB 117. 417 The ERA therefore has recalculated the present value of the minimum lease payments by adjusting the Base Rent in accordance with clause 4.3 to the commencement of the lease in December 2010. This present value is $18.629 million. A worksheet was produced to show that calculation; DBP has not contested its accuracy as a matter of calculation. It is also agreed that DBP's initial direct costs in establishing the lease should be added, being $0.317 million. Accordingly, correcting the calculation error identified by the ERA relating to the value of the Base Rent, results in a capital value of the BEP Lease which is forecast conforming capital expenditure of $18.946 million. 418 To that limited extent, the Access Arrangement Decision should be varied in any event. As the Tribunal proposes to remit the Access Arrangement Decision to the ERA to make the decision again in certain respects under s 259 of the NGL, and as the ERA will make that variation to the Access Arrangement Decision at the request of DBP in any event, the Tribunal will simply give a direction to secure that outcome. 419 The remaining real issue between DBP and the ERA, once that Base Rent adjustment is made, is whether the present value of the lease payments should have been calculated using a real cost of debt of 4.14%. 420 The DBP contention to the ERA was the value of the BEP Lease and the rights to operate and maintain the entire BEP - to be added to the forecast conforming capital expenditure - should be calculated in three stages. 421 First, it estimated the fair value of the asset ($19.2 million) by determining the depreciated replacement value of the BEP capacity as at the date of inception of the BEP Lease, based on the premise of constructing a similar asset. To the sum of $19.2 million the DBP added an amount of initial direct costs in establishing the BEP Lease, and the further sum of $3.6 million representing the increase in the lease liability between the date of inception of the BEP Lease (18 May 2008), and the date of commencement of the BEP Lease (17 December 2010), a period of 19 months. 422 Whilst both DBP and the ERA considered that AASB 117 providing for the valuation of finance leases was appropriate, the ERA adopted a different approach in its application. 423 In the Draft Decision, the ERA at [156] decided that the lease costs themselves should be regarded as capital expenditure in the 2005-2010 access arrangement period. It also indicated that the present value of the lease payments for the BEP Lease should be determined by a discount rate different from that proposed by DBP using the incremental cost of funds to DBP (being the nominal pre-tax WACC for the DBNGP) which it estimated at 10%: at [172]. 424 DBP responded to the Draft Decision by maintaining its claim that capital expenditure for the BEP should be included in the forecast capital expenditure for the proposed access arrangement period rather than in the then existing period. Its further information indicated that the agreement for the BEP lease became unconditional only on 17 December 2010. In the Final Decision, the ERA at [244]-[246] accepted that proposition. It is no longer an issue. 425 DBP also maintained and refined its claim about the quantum of the forecast capital expenditure. It said that AASB 117 required the determination of the fair value of the BEP Capacity at May 2008 (the inception date of the BEP Lease), the adjustment of that figure to December 2011 real value, and the additional "initial indirect costs". Those costs included an option fee payable to Epic Energy to cover the period while conditions precedent were satisfied, and for external legal, tax and regulatory consultants' fees. In the Final Decision, the ERA at [259] accepted the claim for initial direct costs. 426 Consequently, putting aside the depreciation issue, the remaining - and still contentious - matter for the ERA to determine, and now for the Tribunal to consider, is the quantification of the value of the BEP Capacity to be added to the conforming capital base. There is a significant difference between the figure nominated by DBP and that allowed by the ERA. 427 As noted above, the issue emerges from different applications of AASB 117. 428 The Final Decision, after referring to rules 77, 78 and 79 of the NGR, and the general issues about the opening capital base for the proposed access arrangement period, referred at [221] to the claim for the forecast capital expenditure for the BEP Capacity should be included, and at [235] the need for it to address that claim for the cost of the lease of the BEP Capacity as capital expenditure. It noted at [238] that, in the Draft Decision, it had accepted that the BEP Lease may constitute capital expenditure if it is in the nature of a finance lease. It noted at [240]-[241] that in the Draft Decision it had expressed concern about the forecast inflation rate and the discount rate applied in calculating the present value of the lease payments, and whether the capital expenditure allowed should be included as actual expenditure in 2010. As noted, at [246] it accepted that the expenditure is appropriately allowed in the proposed access arrangement period. 429 Ultimately, there was some $6.5 million difference at 31 December 2010 values in the proposed value of the BEP Capacity between the DBP submission and the Final Decision, allowing for the inclusion of the initial direct costs. 430 The difference reflected the ERA's view that AASB 117 required the value to be determined at the date of commencement of the BEP Lease, and accepted that value as the value DBP ascribed to it at its inception, namely May 2008. DBP's claim to add an "interest" component to fix the December 2010 value (equal to the discount rate applied of 6.78% in the present value calculation) was disallowed. The ERA at [255] said there were no grounds under AASB 117 to include that "interest" adjustment to the capital base. 431 The ERA then said at [256] that, as AASB 117 require the asset value to be determined as the lesser of the fair value of the BEP Lease or the present value of the minimum lease payments, it could not accede to DBP's claim of the "interest" adjusted amount because it had simply equated its "interest" adjusted value without providing an independent determination of fair value and the present value of the lease payments at December 2010. In other words, the comparison required by AASB 117 had not been provided, so the ERA could not make that comparison and simply selected the nominated (and accepted) fair value at May 2008. The ERA at [257] did not accept that the independent auditor of DBP, by signing off on its 31 December 2010 accounts, had thereby itself reviewed the valuation. 432 The Final Decision therefore adopted the present value of the lease payments, as nominated by DBP (but as the "inception" value), plus the initial indirect costs, as the total capital value of the BEP Capacity to be added to the capital base. It noted at [259] that the figure arrived at was less than the fair value of the BEP Capacity as determined by DBP applying a method of depreciated optimised replacement cost. 433 Under the BEP Lease, the Applicant is required to pay monthly lease payments, paid one month in arrears from the commencement of the lease term. Paragraph 4.3(a) of the BEP Lease provides that the base rent has been calculated as of 1 January 2008 at an amount of [Redacted] per month. Paragraph 4.3(b) of the BEP Lease provides that the base rent will be escalated annually on 1 January each year in accordance with a specified formula. In essence, the base rent is escalated in the current year based on the movements in CPI calculated for the 12 months to 30 September in the year before the current year. 434 In the view of the Tribunal, the ERA could and did properly take the view that the present value of the minimum lease payments was the appropriate value to adopt for the conforming capital expenditure. It is not shown to have erred in concluding that that value was not higher than, and probably lower than, the fair value of the leased property, having regard to the material available to it. 435 Paragraph 20 of AASB 117 requires the use of a discount rate in calculating that value as being "the interest rate implicit in the lease" if practically that can be determined, or otherwise the lessee's incremental borrowing rate. 436 The term "minimum lease payments" is defined in paragraph 4 of AASB 117 to be the payments over the lease term that the lessee is required to make, excluding contingent rent. The ERA asserts, and the Tribunal accepts, that because inflation adjustments produce contingent rent increases as they are not known until the CPI changes occur, the minimum lease payments are to be calculated excluding anticipated but uncertain inflation adjustments. DBP's submission includes the assertion that the Base Rent for each future year of the term of the BEP Lease should not be escalated, but that proposition is not necessarily an acceptance of the primary position of the ERA, as it is part only of DBP's approach which also requires a real cost of debt adjustment. 437 The lessee's incremental borrowing rate used by the ERA is the nominal cost of DBP's debt in the rate of return. The ERA says that that is the correct approach - rather than use the real cost of debt - to discount the present value of the future rental payments because the Base Rent itself was not recalculated for the future to allow for any inflationary increments, and the definition of "minimum lease payments" did not provide for such a recalculation. 438 In the Tribunal's view, the ERA's approach is not shown to have been in error. DBP relied upon rule 73(3) of the NGR, requiring all financial information to be provided, and calculations made, consistently on the same basis. However, it is not inconsistent with that rule that the ERA, in determining the conforming capital value for the BEP lease should apply AASB 117 and should ensure that its terms are consistently applied. In the Tribunal's view, that does not produce any conflict between the national gas objective and the way in which AASB operates. It agrees with the ERA that it would only be if the Base Rent payable in the future years under the BEP Lease were to be recalculated, and deflated for the effects of future inflation, that it would have been appropriate to apply a real discount rate to decide the minimum rental payments. 439 Accordingly, on this aspect of the BEP Lease issues, the Tribunal considers that no reviewable error has been established other than that acknowledged by the ERA. 440 The Tribunal will remit the matter to the ERA for the purposes of the ERA recalculating the present value of the minimum lease payments under the BEP Lease by adjusting the Base Rent in accordance with clause 4.3 of the BEP Lease to the commencement of the lease.

Depreciation 441 The second issue concerning the BEP Lease relates to the number of years over which depreciation of the value of the lease could be charged. 442 In its Final Decision the ERA agreed with DBP that the value of the BEP Lease should be added to the forecast of conforming capital expenditure in terms of rule 79 of the NGR. The Final Decision maintained the ERA's view, expressed in the Draft Decision, that the general method applied by DBP to determine depreciation allowances and the asset lives for the four main asset categories (pipelines, compression, metering and other depreciable) met the requirements of rules 88 and 89. However, having given consideration to DBP's submission on depreciation it came to the view that the reasons stated by DBP did not justify treating the BEP Lease in the category of "other depreciable assets". The ERA maintained the view expressed in the Draft Decision that the BEP Lease was in the nature of a purchase of a pipeline asset as it had the same practical outcome for the capacity of the pipeline to provide a service. In the ERA's opinion, a different regulatory treatment of a leased pipeline, on the one hand, and a self-constructed pipeline asset, on the other, would risk affecting the incentives for particular ownership and financing arrangements of assets. 443 Thus, the Final Decision found at [248] and [249] that: (1) the period of depreciation should be determined on the basis of the projected economic life of a pipeline asset (generally 70 years); and (2) the BEP capacity should be treated as a pipeline asset with a remaining asset life determined as the assumed life for a new pipeline (70 years) less the age of the BEP to 2011 (the BEP having been constructed in 1998, therefore the remaining asset life was determined by the ERA as 57 years). 444 The Tribunal notes that in the Draft Decision the ERA had with less precision specified a remaining life of 60 years and that the submissions of DBP referred to below refer to this number and not 57 years. 445 Because the lease was established in 1998 when the BEP facility was constructed, it was accepted by DBP that depreciation of the BEP Lease should be determined over a 30-year period, not the 20 years it had to run when DBP acquired it. 446 DBP submitted that the ERA made several errors in relation to depreciation, including: (1) finding that the term of the lease was 60 years, when in fact the term is no more than 20 years, with DBP having an option to extend by a further term of no more than 40 years; (2) failing to have regard to the fact that: (a) the 20-year term of the BEP Lease is most closely aligned with the regulatory life of assets (30 years) in the category of other depreciable assets; (b) the 20-year terms of each of the access contracts entered into with shippers for which the BEP capacity was required is most closely aligned with the regulatory life of assets (30 years) in the category of other depreciable assets; (c) there is no certainty that DBP will extend the term of the BEP Lease beyond the primary term of 20 years; and (3) in exercising its discretion, it was incorrect and/or unreasonable for the ERA to calculate the value of the BEP Lease using lease payments over a 20-year term, but depreciating the value of the lease over 60 years. 447 While DBP submitted that the ERA erroneously found that the acquisition of the additional capacity under the BEP Lease was in the nature of a purchase of a pipeline asset, it nevertheless sought relief in the form of the BEP Lease value being depreciated over 30 years. However, it did note that because of the long time frames involved, the effect of depreciating the asset over 60 years compared to 30 years was probably minimal in dollar terms. 448 In response, the ERA submitted that it had not erred. It had noted that the initial term of the lease was for 20 years, with an option for a 40-year renewal, and because the option to renew was exercisable at a nominal rent of $1 per month, it considered it was reasonable to conclude that the option would be exercised. Therefore the Final Decision had determined that the commercial effect of the BEP Lease was to provide DBP with the use of the pipeline for its remaining life, which the ERA estimated at 57 years (as described above). 449 In reply, DBP submitted that in taking the view that the 40-year option was at a nominal rent of $1 a month, the ERA had ignored the fact that the term of the shipper contracts that DBP held did not extend beyond 20 years. Accordingly, it submitted, that there was no evidence to indicate that it would need to exercise the option in order to deliver the contracted capacity to the shippers. 450 DBP also contended that if it were to extend the term of the lease it would incur costs beyond the nominal rent and would incur an obligation to contribute to the operation and maintenance of the BEP facility. Thus, if these costs were to be depreciated over a 60-year term and the shippers did not renew their contracts, then a 60-year depreciation period would mean that DBP would never recover its undepreciated capital. 451 In oral submissions, counsel for DBP reinforced that contention by noting that that the DBP shippers currently had 15-year contracts with two extensions of 5 years, but that there was no evidence that suggested any shipper would exercise the option to extend beyond that term. 452 Thus, DBP further contended that the ERA ignored the fact that if the capacity was not contracted, DBP would be at risk of being unable to recover the undepreciated capital after the expiry of the original lease term and that as DBP was the operator of the BEP under the lease, if the capacity were not contracted in the future, it would not wish to undertake its obligations as operator. 453 Counsel for DBP argued that under rule 89 the ERA's discretion is limited. The Tribunal observes that rule 89 is prescriptive in a broad manner that reflects conventional thinking about the purpose of depreciation, as well as allowing for a practical determination of depreciation amounts that allow for the service provider's reasonable needs for cash flow. 454 Most importantly, rule 89(1)(b) is quite specific in stating that the depreciation schedule should be designed "so that each asset ... is depreciated over the economic life of that asset" (emphasis added). 455 In the opinion of the Tribunal, what constitutes the economic life of an asset is a matter that needs careful consideration of the facts surrounding any given asset. It is important to recognise that discretion must be available to a decision-maker to gather and evaluate the pertinent facts and market circumstances. 456 Counsel for the ERA noted that it had determined that in order to arrive at the appropriate depreciation period for the BEP Lease it was proper to treat it as a pipeline asset, and then to estimate the remaining economic life of that asset - a period it ultimately determined to be 57 years. 457 The Tribunal agrees with the submission of the ERA's counsel that the issue of determining depreciation for a finance lease in this regulatory context is "a novel question". 458 Counsel for the ERA submitted that when estimating depreciation it is important to recognise that it must be specific to the conditions surrounding the asset under consideration. In particular, in the context of a finance lease like the BEP Lease, the correct depreciation method should be independent of the method of financing the particular asset involved. It would, he argued, create the wrong economic incentives if the method of financing were allowed to influence the approach to depreciation, and would create incentives to use one financing method over any other. The Tribunal agrees with this proposition. 459 The Tribunal also agrees with the ERA's submission that the method of financing an asset should have no bearing on the relevant depreciation period. Bearing in mind the purpose of depreciation (to allow for the cost of recovering an investment in a long-lived asset), it can properly be measured only by reference to an asset's economic life and its initial value. Once these two economic characteristics are objectively determined as reference points, the appropriate accounting formulae to be used may be decided depending on the purpose for determining the depreciation figure. 460 In the regulatory context under consideration here, depreciation is, without doubt, an economic concept that is given commercial form in terms of various accounting rules, formulae and practices. In that context, conceptually there is no such thing as "legal" or "financing-method" depreciation, concepts which at times seemed to be suggested by counsel for DBP. 461 In the case of an asset lease like the BEP Lease, if the lease term is shorter than the economic life of the asset, at the end of the lease the asset will still have a positive economic life, and a positive commercial value. The lease has purchased only part of the asset's total life and of the asset's value - which is brought to account by means of a relevant depreciation charge. 462 In the light of these observations, it is the opinion of the Tribunal that DBP has incorrectly submitted that the depreciation period should be linked to the length of its customer contracts. As with the incorrectness of linking depreciation to a financing method, it is equally wrong to base depreciation on the length of a business's contracts. This is readily apparent if one were to consider a firm with many different customers each operating on contracts of different lengths, or a firm offering different products that are produced from the same capital assets. 463 In the BEP Lease, DBP acquired the right to use the capacity of the pipeline for 20 years. That transaction did not alter the fact that at the time of the Final Decision, the asset had an estimated 57 years of economic life left. As rule 89 requires, the ERA accordingly, and properly, depreciated the BEP Lease over that economic life. 464 For reasons stated in the preceding paragraphs, the Tribunal is of the opinion that DBP has failed to show error on the part of the ERA in deciding that the BEP Lease should be depreciated over the remaining economic life of the asset, namely, 57 years. 465 Therefore, in terms of s 259(2)(a) of the NGL, the Tribunal affirms the ERA's decision that the depreciation allowance for the BEP Lease should be determined over a 57-year period.

OPERATING EXPENDITURE ISSUE 466 DBP contends that the ERA erred in exercising its discretion in deciding that: (1) DBP's forecast regulatory expenses did not meet the criteria governing operating expenditure (opex) in rule 97(1) of the NGR; and (2) the forecast for each of the years 2014 and 2015 should be reduced by $100,000. 467 The relevant part of Table 2, "DBNGP Forecast Operating Expenditure" in DBP's Submission 12: Justification of Operating Expenditure, 14 April 2010, (Submission 12), shows DBP's forecast of regulatory expenses in 2009 dollars (million) as follows: 2011 2012 2013 2014 2015 Regulatory expenses 0.050 0.051 0.052 0.668 0.685

468 Submission 12 noted that DBP does not maintain specialist regulatory staff other than a manager and that it relies on external advisors for the technical, economic and legal work required in producing an access arrangement revision. The nature of the regulatory expenses DBP expected to incur during the revised access arrangement period was described as preparing and submitting a revised access arrangement; participating in the subsequent approval process; and implementing the approved access arrangement. 469 Based on its previous experience with access arrangement approval, Submission 12 stated that DBP expected to spend: (1) approximately $0.8m on external advisors in 2010 to complete the current access arrangement assessment process; and (2) a similar amount in 2014 and 2015 in relation to the access arrangement to be lodged in 2015. 470 Having considered Submission 12, Halcrow Pacific Pty Ltd in association with Zincara Pty Ltd (Halcrow), the ERA's consultants, noted that the allowance for regulatory expenses was "… approximately in line with DBP's 2005 forecast of regulatory review costs, which was estimated at $0.913 million ($nominal)." Halcrow concluded that it was: … satisfied that the forecast expenditure is such as would be incurred by a prudent service provider acting efficiently, in accordance with accepted good industry practice to achieve the lowest sustainable cost of delivering pipeline services. 471 In its Draft Decision the ERA reached a conclusion different from that of its consultants. 472 The Draft Decision observed that the forecast of $1.21 million over 2014 and 2015 was greater than the $913,000 that was allowed under the 2005 access arrangement for the current review (31 December 2010 $ values), and that there was a small additional allowance for costs in each other year of the access arrangement period. 473 Noting that there were regulatory tasks that would need to be undertaken routinely throughout the access arrangement period, the Draft Decision accepted that the small annual allowance of costs was justified and consistent with the prudence and efficiency requirements. It concluded, however, that: (1) in the absence of substantiating information, the ERA was not satisfied that the increase in forecast costs for review of the access arrangement was consistent with the prudence and efficiency criteria of rule 91 of the NGR; and (2) amendment of DBP's forecast opex was required to reduce the allowance for regulatory expenses to $49,000 in each of 2011 to 2013 and to $505,000 in each of 2014 and 2015. 474 DBP's Submission 54: Operating Expenses, 20 May 2011 (Submission 54), responded to the Draft Decision by criticising the ERA's use of the current revision of the access arrangement as a benchmark because it failed to: (1) take into account the circumstances surrounding the access arrangement revisions for the 2016 to 2020 period; and (2) recognise that DBP has already incurred more than it had foreshadowed for this current access arrangement approvals process. Elaborating on sub-paragraph (1) in the preceding paragraph, Submission 54 noted that the access arrangement revisions for the 2016 to 2020 period would impact DBP's revenue and that DBP would need to ensure that all possible information was submitted to the ERA in a way that convinced it that the access arrangement should be approved. It also noted that there is likely to be a significant increase in interest from stakeholders in that process, thereby requiring DBP to spend more time reviewing and responding to submissions. 475 Submission 54 concluded its response to the Draft Decision on this issue by asserting that DBP's forecast: (1) "… is reasonable as its [sic] will be required to direct more resources to the revisions of this access arrangement relative to the current revision process as it can reasonable [sic] be expected to have larger risks and commercial consequence for its business."; and (2) was arrived at on a reasonable basis. 476 A subsequent report by Halcrow to the ERA concluded that: "… as new (qualitative) supporting information has been provided in relation to regulatory expenses, Halcrow's assessment remains unchanged …"; that is, as quoted above. 477 The Final Decision noted what was put by DBP in Submission 54 but rejected it as not providing "… evidence that justifies its forecast of regulatory costs." It concluded its consideration of the regulatory expenses issue at [727] as follows: While DBP submits that the increase in costs will be necessary in the next review of the access arrangement to "make sure that all possible information is submitted to the Authority in a way that convinces the regulator that the access arrangement should be approved", this is not, or should not be, any different to the current review of the access arrangement. The Authority therefore maintains the determination in the draft decision to reduce regulatory costs to $1.157 million over the 2011 to 2015 period for regulatory costs compared with $1.359 million proposed by DBP. This amounts to a reduction of $0.1 million in each of the years 2014 and 2015. 478 The first of DBP's submissions in support of its contentions was that it was erroneous for the ERA to adopt a forecast of regulatory costs for the next access arrangement period on a basis which had already proved to be inadequate for the 2005 access arrangement. In rejecting this submission, the Tribunal agrees with the ERA's view that the submission is mere assertion. The Draft Decision placed DBP on notice that the ERA was seeking hard information substantiating DBP's forecast of regulatory expenses. DBP did not provide the ERA with any details of the regulatory expenses it incurred in respect of the 2005 access arrangement that might have enabled the ERA to test the veracity of the assertion. It is clear that the lack of hard information substantiating DBP's forecast led the ERA to exercise its discretion to reject it and to reduce the allowable amounts for 2014 and 2015. There is no error on the part of the ERA in so exercising its discretion. 479 DBP's next submission was that the ERA did not give any weight to the importance of the revisions for the next access arrangement period to DBP's revenue, which would justify it being required to expend a greater effort in the regulatory approval process. It is clear from reading the Final Decision at [726] and [727] that the ERA did have regard to DBP's assertions in those regards. Indeed, [726] quotes the elaboration in Submission 54 of the criticism levelled at the ERA for failing to take into account the circumstances surrounding the access arrangement revisions for the 2016 to 2020 period. And, as the above quotation from [727] of the Final Decision shows, the ERA was not satisfied that the regulatory expenses in anticipation of the next review would be greater than the current review. That was a conclusion reasonably open to the ERA on the information before it. It is not open to the Tribunal to substitute a conclusion it might have reached on that information: see ACCC v ACT at [176]. 480 DBP also submitted that the ERA erred in the exercise of its discretion in that it failed to have regard to, or gave insufficient weight to, six matters. The first of those matters was rule 71(1) of the NGR. That rule provides: In determining whether capital or operating expenditure is efficient and complies with other criteria prescribed by these rules, the … [ERA] … may, without embarking on a detailed investigation, infer compliance from the operation of an incentive mechanism or on any other basis the AER considers appropriate. The Tribunal agrees with the ERA's submission that while rule 71(1) empowers the ERA to infer compliance, it does not require it to do so. Accordingly, the Tribunal rejects DBP's submission. 481 The second matter in this series of submissions was said to be rule 74(2) of the NGR. That rule provides: (2) A forecast or estimate: (a) must be arrived at on a reasonable basis; and (b) must represent the best forecast or estimate possible in the circumstances. The Tribunal accepts the ERA's submission to the effect that it did not fail to have regard to, or give insufficient weight to, rule 74(2) but rather reduced the forecast because of a lack of substantiation of the forecast. 482 The third matter DBP submitted that the ERA failed to have regard to, or gave insufficient weight to, was Submission 54. As observed above, it is clear from reading [726] and [727] of the Final Decision that the ERA did have regard to Submission 54. 483 The opinions expressed by Halcrow were the fourth matter targeted by DBP in this series of submissions. That neither the draft nor final decisions reference the Halcrow opinions suggests that there may be some substance to DBP's submission. The Tribunal is, however, persuaded by the ERA's submission that it is entitled to, and required to, make its own decisions about such matters. Clearly the ERA cannot be bound by its consultants' opinions. In accepting the ERA's submission that ultimately it was not satisfied that DBP's forecast complied with rule 91 due to a lack of substantiation, the Tribunal notes that both the draft and final decisions would have been much improved by recording Halcrow's opinions and by the ERA addressing them in terms of the ERA's submissions to the Tribunal. Indeed, but for the ERA's clear reason for rejecting the forecast, its failure to do so might, on one view, be regarded as unreasonable: see ActewAGL at [35] where after considering views on the notion of "unreasonableness" expressed by Gummow and Hayne JJ in East Australian Pipeline Pty Ltd v ACCC, the Tribunal concluded: "… if the decision maker fails to call to attention matters he/she is bound to consider or considers matters which are irrelevant, he/she will be acting unreasonably." 484 The requirement in rule 100 of the NGR that an access agreement must be consistent with the national gas objective was the fifth matter that DBP submitted the ERA failed to have regard to, or give insufficient weight to. As expressed, this was more assertion than submission and the absence of a foundation for it precludes its acceptance. 485 The sixth matter raised by DBP in this series of submissions was the fact that DBP had forecast a comparatively insubstantial increase in regulatory expenses in each of 2014 and 2015 compared to 2009 and 2010, being the corresponding years in the 2005 access arrangement. The fact that the ERA did not regard the increase as insubstantial is apparent from the Draft and Final Decisions and DBP's submission is rejected. 486 DBP's submission in reply to the effect that the ERA does not contend that the forecast was, in fact, imprudent or inefficient (rule 97(1)), nor assert that the forecast was arrived at on an unreasonable basis or was not the best forecast or estimate possible in the circumstances (rule 74(2)), fails to recognise where the onus lies. It is not for the ERA to prove the forecast is or would be imprudent or inefficient or that the forecast was arrived at on an unreasonable basis or was not the best forecast or estimate possible in the circumstances. It is for DBP to satisfy the ERA that its forecast of regulatory expenses met the criteria of rule 91(1) and the test in rule 74(2). And, if DBP fails in that regard, it must not only so satisfy the Tribunal but also satisfy it that the ERA erred in rejecting the forecast. As observed, the conclusion reached by the ERA was reasonably open to it on the information before it and it is not for the Tribunal to substitute a conclusion it might have reached on that information. 487 For reasons akin to those expressed in the preceding paragraph the Tribunal also rejects DBP's other reply submission that the ERA's reliance on DBP's failure to provide historical information was, in all the circumstances, both disproportionate and erroneous. 488 For reasons stated in the preceding paragraphs, the Tribunal is of opinion that DBP has failed to show error on the part of the ERA in exercising its discretion in relation to DBP's forecast regulatory expenses. 489 Accordingly, in terms of s 259(2)(a) of the NGL, the Tribunal affirms the ERA's decision that: (1) DBP's forecast regulatory expenses did not meet the criteria in rule 97(1) of the NGR; and (2) the forecast for each of the years 2014 and 2015 should be reduced by $100,000.

REFERENCE SERVICE ISSUES 490 DBP contends that ERA erred in terms of s 246(1) of the NGL in deciding: (1) that the proposed R1 reference service should be removed as a reference service and include in that arrangement as reference services a full haul T1 service, a part haul P1 service and a back haul B1 service as described in the 2005 access arrangement (comprising two separate grounds); (2) to include a revised definition of "part haul service" in the revised access arrangement; and (3) that amendments were required to DBP's proposed terms and conditions to include provisions substantially the same as those in the 2005 access arrangement.

Proposed R1 Reference Service v Proposed T1, P1 and B1 Reference Services 491 The following paragraphs provide the background necessary for a consideration of the first of those three matters. 492 Three reference services were specified in the 2005 access arrangement: (1) a full haul T1 reference service: that service was a firm full haul service with: (a) a receipt point (the point at which the shipper delivers the gas to DBP upstream of main line valve 31 on the pipeline located near compressor station 2 (near Onslow); and (b) a delivery point (that is, the point at which the pipeline delivers gas to the shipper) downstream of compressor station 9 on the pipeline, located near the town of Gingin. (2) a part haul P1 reference service: that service was a firm part haul service, that is, a forward haul gas transportation service which is not full haul. A forward haul gas transportation service is one where the inlet point is upstream of the outlet point. (3) a back haul B1 reference service: a back haul service means that the receipt point is downstream of the delivery point. 493 A firm service is a type of service obliging DBP to provide up to the contracted volume without interruption or curtailment, subject to certain permitted exceptions. 494 During the period of the 2005 access arrangement: (1) access to the pipeline was by way of SSCs which were substantially independent of the access terms; (2) no user obtained access to any of the 2005 reference services, nor was any capacity of the pipeline contracted under the 2005 reference services; and (3) SSCs accounted for all of the pipeline's capacity. 495 DBP's revised access arrangement proposed that there be only one reference service, namely, a full haul service it described as the R1 Service. 496 The proposed R1 reference service was described as a full haul service in which the applicant (subject to the availability of capacity): (i) takes receipt, at one or more Inlet Points on a Gas Day, of a quantity of the Shipper's Gas not exceeding: A. the sum of the Shipper's Contracted Capacity at the Inlet Points; B. plus or minus the quantity of Gas required to correct any Accumulated Imbalance on the preceding Gas Day; (ii) delivers to the Shipper at one or more Outlet Points on that Gas Day a quantity of Gas not exceeding the Shipper's Contracted Capacity at Outlet Points, without interruption or curtailment except as permitted by the Access Contract Terms and Conditions; and (iii) otherwise provides R1 Service on terms and conditions set out in the Access Contract Terms and Conditions. 497 The applicant's revised access arrangement also proposed that specified part haul and back haul services offered by the applicant would be non-reference services. The suite of non-reference services the applicant proposed to offer included: (1) a firm full haul T1 service which would, subject to availability of capacity, be provided on the terms and conditions in a T1 SSC Access Contract; (2) a forward part haul P1 service which would, subject to availability of capacity, be provided on the terms and conditions in a P1 SSC Access Contract; and (3) a back haul B1 service which would, subject to availability of capacity, be provided on the terms and conditions in a B1 SSC Access Contract. 498 As indicated above, the ERA's access arrangement proposal pursuant to s 64(1) of the NGL gave effect to the required amendments set out in its Final Decision for the removal of the proposed R1 service as a reference service. It also required the inclusion of a full haul T1 service, a part haul P1 service and a back haul B1 service as reference services. 499 As may be seen from the following paragraphs, the T1, P1 and B1 reference services should have more relevance as a touchstone for a shipper negotiating the tariffs to apply from 2016 under a SSC than during the revised access arrangement period. 500 During the period of the revised access arrangement: (1) access to the pipeline will be by way of a SSC, substantially independent of the reference services in the proposed access arrangement; and (2) the pipeline's forward haul capacity is fully accounted for under the SSCs and it is unlikely that any capacity of the pipeline will be contracted under the reference services unless a current user is no longer able to honour its obligations under its SSC, such as if it were to become insolvent. 501 The SSCs were entered into in 2004 with terms of 15 years and options of two further terms of five years, thus extending beyond the revised access arrangement period. 502 The main difference between the 2005 access arrangement's T1, P1 or B1 reference services and its equivalent under a SSC lies in the latter's mechanism for funding the pipeline's expansion. Put simply, that mechanism involved a user acquiring a service under a SSC paying a tariff higher than the then reference tariffs - the higher tariff being paid on the basis that it would revert to a regulated reference tariff in 2016. If a pipeline expansion were to have been required to provide additional capacity for a prospective user of the 2005 access arrangement's reference services, the prospective user would have had to rely on the arrangement's extension/expansion policy. The same may be said, mutatis mutandis, in relation to a service provided under a SSC in the proposed access arrangement period and a service that might be offered under the ERA's proposed T1, P1 or B1 reference services. An additional difference between the SSCs and the ERA's proposed P1 reference service lies in the definition of that service. This difference gives rise to DBP's complaint referred to in [448(2)] above and is canvassed below. 503 The SSCs were negotiated to enable the pipeline to be purchased out of receivership and in order to fund expansion of the pipeline. It was Alinta and Verve's submission that absent the higher tariff under the SSCs, not only would the pipeline not have been expanded after 2004, it is most likely that the pipeline would not have been purchased out of receivership at all. They also submitted that the offering of the services under the SSCs was therefore fundamental to the continued operation of the pipeline and what they described as the very significant expansion and investment in it that has occurred since 2004. 504 The APA Group submitted that the fact that a reference service may not be supplied at its tariff during the access arrangement period is not of direct relevance to the ERA's determination because the reference service, reference tariff and other reference terms of access will still have a significant role in the regulatory system established under the NGL and the NGR. It noted in this regard that: (1) the terms provide a basis for negotiating access to the pipeline and, if there is an access dispute, they may effectively be enforced through arbitration; and (2) the part haul P1 reference tariff would provide a basis from which non-reference services might be negotiated and, ultimately, something the access seeker may revert to in the event it is unable to reach agreement. 505 The APA Group also submitted that the specification of the reference services and tariffs will have even more significance than they had previously. That is because the SSCs tariffs revert to arrangements tied to the access arrangement applying to the pipeline in 2016, which reversion will occur during the proposed access arrangement period. It submitted in this regard that: (1) it may be anticipated that the parties to the SSCs will review their arrangements which apply beyond 2015 and, perhaps, seek to enter into negotiations with the applicant as to the arrangements that should apply following the date that their tariffs are governed by the access arrangement; and (2) such reviews of the SSCs and any negotiations would be properly informed by an access arrangement which specifies reference services that those shippers are likely to demand over the access arrangement period. 506 If approved, DBP's proposed R1 reference service would have fewer features than an existing SSC T1 service, in particular, less flexibility in relation to imbalances, peaking and overrun. 507 As counsel for DBP explained it, the proposed R1 reference service would, if approved, be a "plain vanilla" service and its T1 non-reference service "one with the bells and whistles". For example, the SSC T1 service allows peaking and imbalance rights that would give a user an ability to effectively reserve more capacity in the pipeline than might actually be necessary. Thus, she submitted, a user might be attracted to a plain vanilla R1 reference service rather than paying extra for the bells and whistles of the SSC T1 service which may never be used. 508 On the other hand, Verve and Alinta submitted that the R1 reference service was an inferior and less valuable service than the T1 reference service. Counsel for these companies supported that submission by reference to paragraph [46] of the Draft Decision which notes that a majority of submissions made to the ERA claimed that there was no evidence that the proposed R1 reference service would be sought by a significant part of the market because the service is of a lower quality than the existing T1 service. He reinforced his submission by noting that the eight shippers listed by the ERA in a footnote to the paragraph gave a resounding "No" in answer to the question whether existing shippers had an interest in the proposed R1 reference service. 509 The Draft Decision also noted that: (1) the proposed Rl reference service had limited attractiveness to existing and prospective shippers which now have access to the T1 service on more attractive terms and conditions at the same price (citing in support of that observation submissions from Alinta, BHP Billiton, Rio Tinto, and Verve); and (2) several parties (Alinta, Verve and the APA Group) had submitted that reference services additional to DBP's proposed R1 reference service are necessary to accommodate gas storage facilities; in particular, the APA Group submitted that a range of additional reference services should be included to support development and use of the Mondarra Gas Storage Facility (MGSF). 510 Another criticism of the proposed R1 reference service levelled by Verve and Alinta relates to DBP allocating costs to the services provided to shippers with SSCs as if those shippers had been provided with the R1 reference service and that costs allocated for recovery by the R1 reference service match the total forecast revenue for the pipeline. As their counsel explained (using the T1 service as a shorthand for the three reference services under the 2005 access arrangement): … costs were allocated to determine the tariff to all the T1 shippers as if they were taking R1. In other words, the costs for the lesser R1 service are equated to the costs for the superior T1 service …. … So the T1 costs, what would otherwise be the T1 revenue, was proposed by the applicant to be allocated to R1 [sic]. 511 Thus, as seen by the protagonists (and as the ERA observed in its Final Decision), while the existence of the SSCs mean that the 2011-2015 reference services and reference tariffs under the revised access arrangement may not significantly affect users, the parameters of the revised access arrangement will have a significant effect on the starting point for the subsequent (2016-2021) access arrangement, including the approved building-block components that will determine the total revenue requirement and reference tariffs in 2016. 512 Acknowledging that the proposed reference services/tariffs may not have an immediate application during the proposed access arrangement period, Verve and Alinta submitted that: (1) ensuring that the agreed relationship between the SSCs and the regulatory regime is maintained is fundamental to ensuring that the operation of, and investment in, the pipeline is consistent with the national gas objective in s 23 of the NGL; (2) any undermining of that relationship would be inconsistent with achieving the objective and potentially contrary to s 321 of the NGL as it would have the effect of depriving SSCs shippers of protected contractual rights, namely, the rights set out in clause 20.5 and the reversion to a regulated tariff from 2016; and (3) clause 20.5 is to have effect as a contractual right for the purposes of ss 188 and 321 of the NGL. Accordingly, Verve and Alinta submitted that the ERA erred in deciding that DBP's proposal would not contravene s 321. 513 Expanding on that submission, counsel for Verve and Alinta contended that clause 20.5 commits DBP to do what it can to seek to preserve the T1 service as a reference service in the proposed access arrangement period because in 2016 the tariff under the SSC reverts to a regulated tariff, being the tariff for the reference service which is closest to the T1 service under the contract. Thus, he submitted a reference service needs to be identified which is closest to the service provided under the SSC, and the tariff for that service then, in broad terms, becomes the tariff in 2016. Accordingly, the interest of Verve and Alinta in preserving the T1 service as a reference service is that the further the reference services under the proposed access arrangement depart from the SSC regime, the harder it is going to be to make them apply in the way that it is intended. 514 In reply to Verve's and Alinta's contractual/ss 188 and 321 submission DBP submitted that the only contractual right created by clause 20.5, for the purposes of ss 188 and 321 of the NGL, is the right of a shipper to have DBP include a T1 service as a reference service when DBP considers it appropriate to do so. 515 The Verve and Alinta contractual/ss 188 and 321 submission was squarely addressed by the ERA in the following paragraphs in the Final Decision: 22. In submissions to the Authority on the proposed revised access arrangement, some parties … [Alinta and Verve] … contend that the link between the standard shipper contract and the access arrangement is explicit and needs to be maintained to ensure the transition in 2016 to reference tariffs. Some users submitted that the link is critical to the re-commercialisation and ongoing investment in the DBNGP and users have paid a premium over and above the reference tariff to ensure this. Further, it was submitted that the link needs to be maintained and to do otherwise would be inconsistent with section 23 (the national gas objective) and section 321 (protection of certain pre-existing contractual rights) of the NGL. … 24 The Authority considers that the existence and terms of the standard shipper contract (and any other contract for services that DBNGP may have) do not have a direct bearing on the Authority's assessment of the access arrangement proposal except that, under section 321 of the NGL, an access arrangement must not have the effect of depriving a person of a relevant protected contractual right. 25. The Authority has considered the terms of clause 20.5(f)(iii) of the standard shipper contract (relating to obligations of the operator in respect of a reference service for the access arrangement and the tariff for that service) in light of the requirements of section 321. The Authority is of the view, however, that whether or not this clause creates contractual obligations for DBP to make certain inclusions in the access arrangement is a matter for DBP and its contracted shippers to resolve and does not affect the Authority's assessment of the access arrangement proposal. 26. Indeed, the parties themselves appear to have recognised this, as clause 20.5(f)(iii) required no more from the Operator than "to endeavour ... to have the Regulator approve amendments" to the access arrangement that would have specified outcomes. This is implicit acknowledgement that any submissions made to the Authority would have, at best, persuasive value and would not be binding on the Authority. 516 In advancing its case in relation to these matters, DBP contended that the ERA erred in: (1) its construction of rules 48(1)(c) and (d) and 101(2) of the NGR and, in particular, made an error of fact in finding that the SSCs were an important indicator of the relevant market for pipeline services; and (2) failing to give sufficient weight to the national gas objective in s 23 of the NGL and, in particular, DBP's submission that its proposed R1 reference service would offer a more efficient utilisation of the pipeline's capacity. Each contention is considered seriatim below. 517 Rules 48 and 101 of the NGR relevantly provide: 48 Requirements for full access arrangement (and full access arrangement proposal) (1) A full access arrangement must: … (b) describe the pipeline services the service provider proposes to offer to provide by means of the pipeline; and (c) specify the reference services; and (d) specify for each reference service: (i) the reference tariff; and (ii) the other terms and conditions on which the reference service will be provided; and … 101 Full access arrangement to contain statement of reference services (1) A full access arrangement must specify all reference services. (2) A reference service is a pipeline service that is likely to be sought by a significant part of the market. 518 To make its case that the ERA erred in its construction of rules 48 and 101, DBP also relied on the following definition of "pipeline service" in s 2 of the NGL: pipeline service means- (a) a service provided by means of a pipeline, including - (i) a haulage service (such as firm haulage, interruptible haulage, spot haulage and backhaul); and (ii) a service providing for, or facilitating, the interconnection of pipelines; and (b) a service ancillary to the provision of a service referred to in paragraph (a), but does not include the production, sale or purchase of natural gas or processable gas; 519 DBP asserted two construction errors on the ERA's part. The first concerned the relevant market for determining the demand for reference services. The second concerned the specification of a reference service in the access arrangement. 520 The first difference between DBP and the ERA turns on whether, in determining if a service is "… likely to be sought by a significant part of the market" (rule 102(2)), regard should be had to: (1) only a narrow class of users or prospective users who are likely to wish to enter contracts for further services during the access arrangement period; or (2) a broader class of existing users and prospective users who are likely to be supplied with existing or further services during the access arrangement period. DBP opted for the former, the ERA the latter. 521 The second difference between DBP and the ERA turns on whether in specifying a reference service for the purpose of rule 48 DBP needs to: (1) only specify it by reference to the availability, length and direction of the haulage provided (eg: firm haulage, interruptible haulage, spot haulage, full haul or back haul); or (2) specify it by reference to all the particular terms and conditions upon which it may be provided. DBP opted for the former, the ERA the latter. 522 The Draft Decision required amendments to remove from DBP's revised access arrangement its proposed R1 reference service and include as reference services the T1, P1 and B1 services as described in the 2005 access arrangement. In deciding on the amendments the ERA concluded, amongst other things, that the services provided under the SSCs were evidence of the relevant market and the services that should be specified reference services. 523 Relying on its construction of rules 48 and 101, DBP submitted in response to the Draft Decision that the SSCs are not, of themselves, evidence of: (1) the relevant market for pipeline services; and (2) whether the pipeline services to be provided under the SSCs are services likely to be sought by a significant part of the market. 524 DBP also challenged the Draft Decision's conclusion that whether a service is likely to be sought by a significant part of the market (rule 101(2)) requires consideration of the nature of services demanded by users and prospective users, unconstrained by the availability of pipeline capacity to expand the provision of services during the course of the access arrangement period. 525 A further challenge to the ERA's construction of rules 48 and 101 in the Draft Decision was based on DBP's interpretation of the NGL's definition of "pipeline services". As observed, it opted for a broad interpretation suggesting that the reference to pipeline service in rule 101 should be characterised by its general nature rather than its specific terms and conditions; that is, a firm haulage service or an interruptible haulage service, without regard to the differences in the terms and conditions on which the service is provided. According to this view, DBP submitted, its R1 reference service, the negotiated T1 SSC service and the T1 reference service are merely different forms of firm haulage services and it is immaterial that the proposed R1 reference service differed, even in significant respects, from the existing T1 service. 526 Submissions in response to the Draft Decision from Alinta, Verve and BHP supported the ERA's amendments to remove from DBP's revised access arrangement its proposed R1 reference service and include as reference services the T1, P1 and B1 services as described in the 2005 access arrangement. 527 Alinta agreed with the ERA's interpretation of rule 48 and 101 and stated: As the T1, P1 and B1 services continue to be the primary services required by shippers on the … [pipeline] … during the period 2011 lo 2015, it is clear that those services are likely to be sought by a significant part of the market and are therefore required to be Reference Services under rule 101 of the NGR. 528 In its submission, Verve noted that: DBP has not provided any evidence that the proposed R1 service is one that would be sought by a significant part of the market and is therefore required to be a Reference Service under the NGRs. Third party submissions overwhelmingly disagree with DBP's Proposed Revisions, and in so doing reject the introduction of the R1 service and require that the Tl, P1 and B1 services are included as Reference Services. Verve agrees with the ERA's interpretation of rules 48(1)(b) and 101 of the NGRs as to the relevant services (including Reference Services) to be included in the Access Arrangement. Verve agrees with the ERA that the question, under rule 101(2) of the NGRs, whether a pipeline service is likely to be sought by a significant part of the market requires consideration of the nature of services sought by users and prospective users, unconstrained by the availability of pipeline capacity to expand the provision of services during the course of the relevant access arrangement period. As the T1, P1 and B1 services continue to be the primary services required by shippers on the DBNGP during the period 2011 to 2015, it is clear that those services are likely to be sought by a significant part of the market and are therefore required to be Reference Services under rule 101 of the NGR. 529 BHP submitted that along with other shippers it maintained that: … the proposed Rl Service is unlikely to be sought by a significant part of the market. Therefore, in light of rule 101(2) of the NGR, such service which is not demanded by users and prospective users should not be a reference service. BHPB concurs with the Regulator's Draft Decision that T1, Pl and B1 Services are all pipeline services that will be sought by a significant part of the market during the period covered by the access arrangement. Consequently, BHPB agrees that these services should be specified as reference services in the DBNGP Access Arrangement. Based on the above, and the submissions made in BHPB's First Submission, BHPB submits that DBP's proposal in the Amended Proposed Access Arrangement to retain the Rl Service and not include the T1, Pl and Bl Services as reference services be rejected. 530 In the Final Decision the ERA noted that the users referred to in footnote 12 to paragraph [46] of its Draft Decision consistently submitted that the proposed R1 service is of an inferior quality to the existing T1 Service and that they would not be seeking to use the R1 Service. 531 It is clear from the following paragraphs in the Final Decision that the ERA had regard to DBP's submissions summarised above and rejected them: 84. In its draft decision, the Authority took the view that, under rule 101(2) of the NGR, the question of whether a pipeline service is likely to be sought by a significant part of the market requires consideration of the nature of services sought by users and prospective users, unconstrained by the availability of pipeline capacity to expand the provision of services during the course of the relevant access arrangement period. That is, the Authority took the view that the question of whether a pipeline service is likely to be sought by a significant part of the market requires consideration of the totality of demand for services and should not be limited to consideration of only incremental demand over and above the quantum of services already contracted for under pre-existing contracts. … 132. The Authority disagrees with DBP's contention that a consideration of reference services should only occur according to the "general character" of the service and without regard to the terms and conditions for the service. The character of the service is in large part determined by the principal terms on which the service is provided. In the case of the R1 Service, this service is characterised by features established in the terms and conditions, including priority of curtailment relative to other services; provisions for overrun, imbalances and peaking; and provisions for aggregation across inlet and outlet points. Users regard the R1 service as different from the T1 Service as a result of these terms and conditions. 133. On the matter of the relevant market for services that forms the basis of consideration of which services should be reference services, the Authority maintains the view expressed in the draft decision that the relevant market is the total market for pipeline services provided by the DBNGP, which will include any expected increase in provision of services during the access arrangement period for which the approved access arrangement will apply. 134. On the matter of the relevance of existing contracts for services, the Authority maintains the view that existing contracts comprise one source of evidence of the nature of services demanded by users, which indicates a demand for services in the nature of the T1, P1 and B1 services. A second source of evidence is submissions from users of the DBNGP that clearly indicate a demand for the T1, P1 and B1 Services included in the current access arrangement, and indicate that there is no demand for a service in the nature of the proposed Rl Service. Together, these two sources of evidence are the only substantive evidence available to the Authority on the nature of services sought by users. In contrast to this evidence, DBP has not provided any supporting evidence of demand for the proposed R1 Service. 532 DBP contends that in exercising its discretion to remove the proposed Rl reference service the ERA failed to consider, or give sufficient weight to, the following matters: (1) DBP's proposed Rl reference service would better achieve the national gas objective and, by reason of rule 40(3), it was not open to the ERA to prefer its T1 reference service; and (2) the terms and conditions attaching to the proposed Rl reference service would offer a more efficient utilisation of the pipeline's capacity for the long-term benefit of all shippers and prospective users. 533 Rule 40 governs the exercise of the ERA's discretion. Rule 40(1) sets out when the ERA has no discretion in its decision making process, rule 40(2) when it has limited discretion and rule 40(3) when it has full discretion by providing that in all other cases: … [ERA] has a discretion to withhold its approval to an element of an access arrangement proposal if in the [ERA's] opinion, a preferable alternative exists that: (a) complies with applicable requirements of the Law; and (b) is consistent with applicable criteria (if any) prescribed by the Law,. 534 In its written submissions to the Tribunal, DBP contended that its proposed R1 reference service satisfied the requirements of the NGL and NGR and should have been accepted as the relevant reference service because its terms and conditions would: (1) modify the behaviour of shippers and prospective users by imposing stricter limits (relating to accumulation of imbalances and peaking and overrun limits) upon their utilisation of pipeline capacity; and (2) offer more efficient utilisation of the pipeline for the long-term benefit of all shippers and prospective users than the T1 service. 535 Responding to DBP's written submissions concerning the practical implications of how imposing stricter limits relating to accumulation of imbalances and peaking and overrun limits, the ERA noted that such submissions were: (1) not advanced prior to the Final Decision; and (2) only advanced by DBP in its Submission 73, submitted 13 December 2011. 536 While conceding that prior to the Final Decision, DBP's Submission 50 contained assertions that its proposed R1 service was: (1) likely to encourage competition and growth leading to efficient operation and use of natural gas services consistent with the national gas objective; and (2) was a service that would lead to better utilisation of pipeline capacity, the ERA submitted that DBP's Submission 50 contained no explanation or evidence to substantiate those assertions. 537 The Final Decision concluded that DBP had not substantiated its submissions that its proposals dealing with imbalances, peaking and overrun limits would benefit the integrity of the pipeline or efficient use of pipeline capacity. 538 The ERA submitted that in the circumstances its failure to expressly address DBP's Submission 73 does not give rise to reviewable error because: (1) it was not obliged to undertake consultations with respect to that aspect of the Final Decision prior to proposing its access arrangement decision pursuant to rule 64(2); and (2) it would have been improper for it to take account of Submission 73 after making the Final Decision without giving other interested parties an opportunity to respond to those submissions. 539 Having regard to the Tribunal's finding set out below that there is no merit in these grounds of review, it is not necessary for the Tribunal to give detailed consideration to Verve and Alinta's contractual/ss 188 and 321 submission. Suffice it to say that parties that are subject to the regulatory scheme under consideration here may neither contract out of their regulatory obligations nor may they contract to govern the ERA's determination in relation to an access arrangement proposal. The Tribunal agrees with the ERA's observations in those regards in paragraph [26] of its Final Decision. 540 The Tribunal also agrees with the submission by counsel for Verve and Alinta to the effect that the terms and conditions on which a reference service is to be offered are inseparable from the nature of the service. That is clear from rule 48(1)(b),(c) and (d). That was the ERA's conclusion in paragraph [132] of its Final Decision. The Tribunal agrees with the conclusion. To support his submission that the terms and conditions cannot be separated from the nature of the service as contended by DBP, counsel referred to the revenue and pricing principles in s 24 of the NGL. 541 Thus, if there were to be bifurcation of the nature of a reference service, on the one hand, and its terms and conditions, on the other, it would not be possible to make the assessment required by s 24(2) whether a reference service provided a reasonable opportunity to recover at least the efficient costs of providing the service. Likewise with s 24(5), which requires an assessment whether a reference tariff allows for a return commensurate with the regulatory and commercial risks involved in providing the reference service to which the tariff relates. 542 As counsel submitted: … the whole point of specifying a reference service is to understand what is being provided and allocate costs and revenue to it efficiently, and … that can't be done … if you separate the terms and conditions from the specification of the reference service." … … nothing is to be read into the fact that …[rule 48(1)] …(c) talks about specifying a reference service and … (d) talks about specifying the terms and conditions. They are inseparable as a concept. The Tribunal is of the same mind. 543 The Tribunal also agrees with the submission by counsel for the APA Group to the effect that DBP's construction of rule 48 conflates two different requirements of rule 48(1), namely sub-rules (b) and (c). As he submitted, each sub-rule requires a different thing. Rule 48(1)(b) requires a description of the pipeline services which the pipeline service provider proposes to offer. Rule 48(c) requires a specification of the reference services. They are different things, yet by DBP's construction of the definition of "pipeline services" it conflates the services it might offer with the services that are to be reference services. The Tribunal adopts this submission that that is not the scheme of rule 48. Further, the suggestion that a reference service might be described in some short form way by reference to the definition is incorrect. The reference service is, by rule 48(1)(d), to be described together with its tariff and its terms and conditions. 544 Turning now to the construction of rule 101. It follows from what is said in the preceding paragraphs concerning the construction of rule 48 that rule 101(1) requires more than just specification of the reference services in terms of the definition of pipeline services in s 5 of the NGA. Specification in terms of rule 101(1) requires the applicant to specify the material terms and conditions. As counsel for Verve and Alinta submitted: A description such as "firm, forward, full haul" is not enough. Indeed, even the description "firm", for example "firm service", is meaningless unless the terms and conditions accompany it. Just saying it's firm provides no real guidance on how firm it is, how interruptible it is, what the curtailment rates are, where it stands in the curtailment plan; all of these are important and describing or specifying a service in any meaningful way requires that terms such as those and many others be specified. Again, the Tribunal is of the same mind. 545 The Tribunal rejects DBP's submission that the assessment required by rule 101(2) of whether a pipeline service is likely to be sought by a significant part of the market is confined to a narrow class of users or prospective users who are likely to wish to enter contracts for further services during the access arrangement period. As observed by the ERA in paragraphs [84] and [134] of its Final Decision, the SSCs are but one source of evidence about the nature of the services demanded by users. It is not just the incremental demand above the SSCs users which is to be taken into account in determining whether a pipeline service is likely to be sought by a significant part of the market. 546 The word "market" used in rule 101(2) is not a word defined in the NGL or the NGR. Absent such a definition, it is "back to basics". That is to apply the ordinary meaning of the word in the context of rule 101(2) and the regulatory scheme as a whole. The ordinary meaning of "market" in that context is "… a demand for a commodity or service" (The Australian Concise Oxford Dictionary). Applying that ordinary meaning, the ERA was correct in having regard to the totality of the market and not just a segment of it artificially conceived by DBP to suit its construction. 547 Contrary to DBP's submission, there is no an error of fact in the ERA finding that the SSCs were an important indicator of the relevant market for pipeline services. As is clear from paragraph [134] of the Final Decision, the SSCs were but one of two indicators taken into account by the ERA. The second was the submissions from users that indicated a demand for T1, P1 and B1 services and no demand for DBP's proposed R1 reference service. Leaving to one side Submission 73, those indicators were the only hard information available to the ERA at the time of its Final Decision and DBP did not provide any supporting hard information of a demand for its proposed R1 reference service. 548 Again leaving to one side Submission 73, prior to the Final Decision there was no explanation, no material or hard information advanced by DBP to support its assertions that its proposed R1 reference service would offer a more efficient utilisation of the pipeline's capacity. All that was advanced by DBP were assertions. 549 An explanation came only after the Final Decision. It took the form of Submission 73 purporting to be in response to an invitation issued by the ERA after its Final Decision to make submissions on amendments to DBP's proposed revised access arrangement additional to those in the Final Decision. The Tribunal accepts the ERA's submission that: (1) only Sections 10 and 11 of Submission 73 were responsive its invitation; and (2) those parts of Submission 73 that might explain and/or support DBP's assertions were not responsive to the invitation. 550 The Tribunal has previously made observations to the effect that a regulated entity subject to a limited merits review regime, like the one governing this application, must ensure that material it wants considered by the Tribunal is put fairly and squarely to the regulator in accord with the regime. 551 In this matter the explanatory material that DBP sought to have the Tribunal consider in support of its assertions was put to the ERA only after its Final Decision. 552 The regulatory regime under consideration here envisages consultations with DBP and interested parties following: (1) submission of DBP's revised access proposal; and (2) publication of the ERA's Draft Decision, but not after the publication of the ERA's Final Decision and prior to formulation of the access arrangement that s 64(1)of the NGL requires it must propose if it refuses to approve DBP's proposal. 553 Section 64(3) of the NGL is very clear in that regard - it provides the ERA may (but is not obliged to) consult on its s 64(1) proposal. There are sound reasons for not imposing a post-Final Decision requirement to consult. First, consultation, indeed usually extensive consultation, will have occurred prior to the Final Decision. Secondly, if, as here, the ERA refuses to approve DBP's proposal, s 64(1) imposes a time constraint upon the ERA which may well preclude a proper consideration of additional post-Final Decision material. A fortiori as procedural justice dictates that any such material submitted by DBP would have to be the subject of consultation with interested third parties, then, more likely than not, it would provoke a further round of submissions to which the applicant would have to be given the right of reply. Thus, the regulatory scheme draws a line leaving it to the ERA's unfettered discretion whether to proceed down that path. The APA Group's submission in this regard made in respect of the ground of review under this heading about the definition of "part-haul" service and canvassed below is particularly pertinent. 554 Here, the Draft Decision and post Draft Decision third party submissions should have left no doubt in DBP's mind that it would take more than mere assertions (indeed that it would take hard information) to convince the ERA that: (1) the proposed R1 reference service was in fact a superior service in terms of national gas objective; and (2) would offer a more efficient utilisation of the pipeline's capacity. 555 DBP's post-Draft Decision submission failed to provide that hard information. Parts of its post-Final Decision Submission 73 may or may not have provided such hard information. The ERA exercised its discretion not to consider those parts of Submission 73. There was no error by the ERA in so exercising its discretion. 556 Nor was there error on the part of the ERA in the paragraphs of its Final Decision where it made observations to the effect that: (1) it expected DBP's proposal would be supported by a clear demonstration that the benefits to the integrity of the pipeline (and hence to all pipeline users) justified it and any additional cost to users; and (2) despite the opportunity to do so, the applicant had not provided such a demonstration.

Definition of part haul services 557 DBP alleges the ERA erred in terms of s 246(1) of the NGL in its: (1) decision to require amendment of DBP's revised access arrangement proposal to include a definition of "part haul service" and the P1 reference service to make it consistent with the terms of the 2005 access arrangement; and (2) conclusion that the P1 service definition in the 2005 access arrangement does not allow for the transportation to outlet points downstream of compressor station 9, regardless of the location of the inlet point. 558 The background necessary to a consideration of this ground is to be found in the preceding section of these reasons and in the following paragraphs which outline the sequence of submissions to the ERA by DBP and interested third parties in the process leading to the ERA's Final Decision. 559 Prior to the Draft Decision the APA Group, Alinta and Verve made submissions to the ERA on the definition of the P1 reference service. Those submissions argued additional reference services were necessary to accommodate gas storage facilities. In particular, the APA Group submitted that a range of additional reference services should be included to support development and use of the MGSF near Dongari. 560 DBP's Submission 26 Response to 3rd Party Submissions, 6 August 2010 (Submission 26) observed that the gas production opportunities that might exist south of the MGSF were potential opportunities at best and were not likely to materialise in the next five years and that: (1) it was not in receipt of any request for access to capacity at the Mondarra outlet point or any other outlet point for delivery in the MGSF; (2) the pipeline did not directly connect to the MGSF and there was a need for shippers to access some of APA Group's facilities to access the MGSF; and (3) while DBP was eager to support the growth of additional gas fields to create competition in the upstream market, there was no evidence to justify that these new fields would be able to be commercialised. 561 The Draft Decision disagreed with DBP and accepted that there was a reasonable prospect of: (1) increased use of the MGSF during the course of the revised access arrangement period; and (2) this facility being used by a significant part of the market. 562 The ERA took the view that the Pl and the B1 reference services specified under the 2005 access arrangement and required by the ERA to be specified in the revised access arrangement period, supported the use of the MGSF. 563 In response to the Draft Decision, DBP submitted a further revision to its revised access arrangement proposal which, as outlined above, contained a single full haul reference service, namely, the R1 service and which, most relevantly to the contentions under consideration here , did not include part haul and back haul services as reference services. 564 On 20 May 2011, the APA Group responded to DBP's further revision by submitting that the definitions of the services under the current access arrangement are open to an interpretation that would result in the P1 reference service not supporting use of the MGSF. The submission included a detailed interpretation of the terms and clauses. 565 Paragraph [145] of the Final Decision observed that while the ERA did not necessarily agree with the APA Group's detailed interpretation, the definition of the term "part haul service" in the revised proposed access arrangement should be revised to achieve clarity in the nature of the P1 reference service (consistent with the intent for this service in the 2005 access arrangement) and for inclusion of this service in the access arrangement for the proposed access arrangement period. The ERA's observations concluded with a Required Amendment 4 in the following terms: The definition of "part haul service" in the revised proposed access arrangement and the terms and conditions for reference services should be amended to: Part Haul Service means a service to provide Forward Haul on the DBNGP which is not a full haul service and which includes, without limitation, Services where the Inlet Point is upstream of main line valve 31 on the DBNGP and the Outlet Point is upstream of Compressor Station 9 on the DBNGP, Services where the Inlet Point is downstream of main line valve 31 on the DBNGP and the Outlet Point is downstream of Compressor Station 9 on the DBNGP, and Services where the Inlet Point is downstream of main line valve 31 on the DBNGP and the Outlet Point is upstream of Compressor Station 9 on the DBNGP. The specification of the P1 Service as a reference service in the access arrangement should be consistent with this definition of part haul service. 566 The final submission in the sequence of submissions relevant to a consideration of this contention was DBP's Submission 73 which, as observed above, followed the Final Decision. 567 DBP submits that in redefining the P1 reference service the ERA failed to appreciate that: (1) it was fundamentally changing the previous definition of this reference service; and (2) the consequence of so doing would be to alter the capacity of the pipeline and the ability of DBP to earn the total revenue prescribed for the purposes of the revised access arrangement proposal. The ERA's failures in those regards constitute, DBP submitted, an incorrect exercise of discretion or the making of an unreasonable decision: NGL s 246. 568 In support of its submissions. DBP contended that the fact that the ERA did not appreciate the effect of the changes which it made to the definition of the P1 reference service is illustrated by contrasting the definitions of "part haul service" and "full haul service" in the SSC prescribed under the access arrangement. 569 The definition of "part haul service", it contended, is consistent with the Required Amendment 4 definition, but the definition of "full haul service" is a gas transportation service on the pipeline "where the Outlet Point is downstream of Compressor Station 9 on the DBNGP, regardless of the location of the Inlet Point, but does not include Back Haul". 570 It further contended that applying this definition of a full haul service would cover a "part haul service" where "the Inlet Point is downstream of main line valve 31 on the DBNGP and the Outlet Point is downstream of Compressor Station 9 on the DBNGP". 571 The overlapping definitions on "part haul service" and "full haul service" in the standard shipper contract, DBP submitted, clearly demonstrate an incorrect exercise of discretion at some point in the process. Further, the ERA's decision to include the T1, P1 and B1 reference services (in lieu of DBP's proposed R1 reference service) does not reflect the ERA's reasoning contained in paragraph [145] of the Final Decision because Required Amendment 4 is not consistent with the intent of the P1 reference service in the 2005 access arrangement. The Pl reference service definition specified under the 2005 access arrangement, it submitted, does not allow for transportation to outlet points downstream of compressor station 9, regardless of where the inlet point is. Thus, Required Amendment 4 fundamentally changes the nature of a Pl reference service that has been in place since the 1990s - that is, that the outlet point is upstream of compressor station 9. In support of its submissions DBP referred the Tribunal to Submission 73. 572 Also for reasons supported by reference to its Submission 73, DBP contended that the change in the definition of the Pl reference service is inconsistent with the national gas objective in s 23 of the NGL and would make an access arrangement containing it contrary to rule 100 of the NGR. That is because, in its submission, the definitional change does not promote the efficient operation and use of gas services and therefore could not be construed to be in the long-term interests of consumers of natural gas. 573 The APA Group submitted that the ERA's Required Amendment 4 would allow shippers to efficiently utilise the MGSF. That is, it submitted, the part haul reference service is defined by the ERA to not be a full haul service and to include, without limitation: (1) services where the inlet point is upstream of main line valve 31 and the outlet point is upstream of compressor station 9; (2) services where the inlet point is downstream of main line valve 31 and the outlet point is downstream of compressor station 9; and (3) services where the inlet point is downstream of main line valve 31 and the outlet point is upstream of compressor station 9. 574 On the other hand, if accepted, DBP's contentions would, the APA Group submitted, deny the efficient utilisation of the MGSF because its proposed definition for a part haul service would not explicitly provide for transportation to outlet points downstream of compressor station 9 (to the wider Perth area). Rather, the DBP's proposal would have the definition remain unchanged from the definition in the 2005 access arrangement's P1 service. The APA Group contended that the definition should not be so limited and should explicitly provide for transportation to outlet points downstream of compressor station 9 to the wider Perth area as well as receipt points below main line valve 31. 575 The APA Group, noting the following statement in DBP's Submission 64: Response to Third Party Submissions, submitted 20 July 2011 (Submission 64), in response to the Draft Decision: … While DBP is still not in receipt of formal access requests for services for the proposed MGSF facility, DBP [has] been approached by 3 parties that have made initial enquiries regarding the interconnection between the MGSF and the DBNGP, which if they were to result in in [sic] excess of 100 TJ/day being delivered to the relevant outlet point. submitted that it, along with another statement that DBP was "… currently discussing appropriate commercial service options with prospective customers", provided evidence that pipeline services that would provide for the delivery of gas into and out of the MGSF were likely to be sought by a significant part of the market. 576 Notwithstanding that the first statement was somewhat qualified by subsequent paragraphs in Submission 64 and the second was in response to a submission by the APA Group that customers who wish to use the MGSF in combination with the pipeline would face twice the transport costs if there were but a full haul service alone, the Tribunal is of opinion that the statements do provide hard information pointing to the likelihood of a part haul service being sought by a significant part of the market: cf rule 101(2). 577 The APA Group challenged DBP's submission that the ERA redefined the Pl reference service in a manner that fundamentally changed the previous definition of the service. On the contrary, it submitted that the part haul service as defined in the reviewable regulatory decision operates to clarify and remove any ambiguity that was otherwise apparent in the definition in the 2005 access arrangement. 578 The APA Group also submitted that Required Amendment 4: (1) makes clear that the Pl reference service will facilitate delivery of gas from any number of inlet points along the pipeline to any outlet point along it and from any outlet point to any number of delivery points along the pipeline that are downstream from the inlet point; and (2) is consistent with the ERA's decision to specify the P1 service as a reference service on the basis that it is likely to be sought by a significant part of the market pursuant to rule 101(2). 579 The APA Group attacked DBP's reliance on its Submission 73 because it: …. was submitted significantly after the ERA's final decision and only shortly before the Reviewable Regulatory Decision, no party, including APA Group was able to respond to that material (and is not permitted to introduce new material seeking to respond to that material in this merits review process). In order to properly respond to that material, APA Group and other parties, including the ERA, would require further information to test the data relied upon and the corresponding assertions made. 580 As counsel for the APA Group put it: … the applicant had every opportunity to make submissions about the clarification of the definition of the P1 part haul service - the clarification of the definition so as to avoid the possibility of shippers into and out of Mondarra having to pay twice the full haul tariff. 581 Submissions leading to the Draft Decision, the Draft Decision itself and subsequent third party submissions may be fairly summarised as raising the issue of the efficient use of the MGSF and whether a shipper into and out of the facility might have to pay the full haul tariff twice and whether the definition of the P1 service should be clarified. 582 Prior to the Final Decision, DBP had the opportunity to respond to the Draft Decisions and to the third party submissions made in response to the Draft Decision. 583 Counsel for the APA Group contends that DBP's Submission 64, which followed the Draft Decision, failed to address the issue of the efficient use of the MGSF and the related questions summarised above. The APA Group's written submissions say, in effect, that DBP should not be allowed to rely on its Submission 73 which, DBP submits, addresses the efficient use of the MGSF and Required Amendment 3. 584 For reasons outlined above, the Tribunal accepts the APA Group's submissions. As observed above, the ERA chose to exercise its discretion not to consider Submission 73 insofar as it was not responsive to its invitation and there was no error on its part in its exercise of its discretion in the present circumstances. That leaves the questions whether counsel for the APA Group is right in his contention that Submission 64 failed to address the efficient use of the MGSF and Required Amendment 3. 585 The Tribunal's has made a full considered assessment of Submission 64. While Submission 64 did address the use of the MGSF and the questions whether a shipper into and out of the facility might have to pay the full haul tariff twice and whether the definition of the P1 service should be clarified, Submission 64 did not advance hard information sufficient for the Tribunal to conclude that the ERA erred in incorporating in its s 64(1) access arrangement the definition in Required Amendment 4.

Issues in Standard Shipper Contracts 586 By letter dated 15 May 2012, DBP's solicitors informed the Tribunal and the parties that if it were unsuccessful in respect of having its proposed R1 reference service restored as the only reference service, it would not pursue its contention in [448(3)] above. The letter suggested that if DBP were successful in having the proposed R1 reference tariff restored, issues in relation to the terms and conditions might be resolved by way of agreement between the parties or, failing that, might be addressed at a further hearing. 587 For reasons appearing above, DBP has not succeeded in having the proposed R1 reference tariff restored. Accordingly, it is not necessary to decide whether the terms and conditions issues might be resolved as suggested by DBP's solicitors. The Tribunal notes that, for the following reasons advanced by counsel for Verve and Alinta, they had reservations about adopting the suggestion: (1) DBP and interveners have no power to agree what should be in the terms of a regulated access arrangement; and (2) while the ERA might be able to determine the terms and conditions at a later date as a result of a variation application under rule 65 of the NGR or remittal under s 259 of the NGL, the ERA's power to determine the terms and conditions was exhausted when it made its Final Decision; and (3) s 259 of the NGL does not readily admit to an interpretation that would allow the Tribunal to, in effect, make a staged determination of the issues raised by Ground 6 and Ground 9. 588 The Tribunal does not need to consider the correctness of those concerns. 589 For reasons outlined above, the Tribunal : (1) finds no merit in DBP's grounds of review; and (2) affirms the ERA's access arrangement proposed pursuant to rule 64(1) insofar as it was put in issue by DBP on those grounds..

COVERAGE OF EXTENSIONS AND EXPANSIONS 590 One of the requirements for the Access Arrangement Decision is that it set out the extension and expansion requirements: rule 48(1)(g) of the NGR. 591 Section 2 of the NGL defines "extension and expansion requirements" to include the circumstances when an extension or expansion of a covered pipeline is to be treated as forming part of that pipeline and to include whether the extension or expansion will be subject to the applicable access arrangement applying to the pipeline services. 592 The Access Arrangement Decision provided that its terms would apply to incremental services to be provided as a result of any expansion in the capacity of the DBNGP, except in instances where DBP could demonstrate to the ERA's reasonable satisfaction that application of the access arrangement terms to such services is inconsistent with the national gas objective. The position which DBP put to the ERA, and which the ERA declined to adopt, was that any expansion of the DBNGP should not be covered by the Access Arrangement Decision at the election of DBP. 593 An expansion is the enhancement of the capacity to deliver gas within the geographic range of the DBNGP, such as by the addition of a loop or a compressor. 594 It should be noted that the issue raised on this review by DBP does not apply to extensions (as distinct from expansions) of the DBNGP, that is, where the geographical range of the DBNGP has been extended. The Final Decision of the ERA in relation to extensions is not challenged. It remains a matter at the option of DBP whether any extension to the DBNGP during the Access Arrangement Decision period becomes part of the DBNGP and so subject to the access terms. 595 At the time of the Access Arrangement Decision there was no evidence of any predicted expansion of the DBNGP during the period covered by the Access Arrangement Decision. There was also no allowance for any forecast capital expenditure for any expansion in the period of the Access Arrangement Decision. 596 The Revised Access Arrangement Proposal that DBP submitted to the ERA on 8 September 2011 suggested that it should give notice to the ERA of any proposal to expand or extend the DBNGP, and the extension or expansion should become part of the covered pipeline (and so subject to the Access Arrangement Decision) unless DBP elected that it would not become part of the covered pipeline. It also specified some factors to which DBP may have regard in considering whether to treat an extension or expansion as part of the covered pipeline. 597 In the first version of the Final Decision of 31 October 2011, the ERA accepted that proposal. However, by notice of 1 December 2011, the ERA indicated that it was then proposing to amend that decision in relation to expansions to the position that it had adopted in the Final Decision of 22 December 2011 and confirmed in the Access Arrangement Decision, and invited consultation on that issue. The notice of 1 December 2011 gave reasons for that changed view. 598 DBP, as invited, made a further detailed submission on the topic on 13 December 2011. As its contentions in that submission are reflected to a significant degree in its present contentions, it is not necessary to record them in detail at this point. 599 The revised Final Decision and the Access Arrangement Decision adhered to the ERA's foreshadowed ruling. It did not accept that, in respect of any expansions to the DBNGP, it was appropriate to allow DBP in effect an unfettered discretion (Final Decision at [1627]) to elect whether the expansion should be covered by the Access Arrangement Decision. It observed that the result would entitle DBP to take account only of its own commercial interests. 600 The ERA then referred in the Final Decision to the two decisions of the Western Australian Electricity Review Board on proposed revisions to the access arrangement for the Goldfields Gas Pipeline of 22 November 2011: see BHP Billiton Nickel Wet Pty Ltd v Southern Cross Pipelines Australia Pty Ltd & Ors (Application No 1 of 2010) and Southern Cross Pipelines Australia Pty Ltd & Ors v BHP Billiton Nickel West Pty Ltd & Anor (Application No 2 of 2010). That decision was based upon somewhat different provisions in the Gas Code. The ERA did not follow that decision unthinkingly, but said at [1630] that aspects of the reasoning, including those rejecting the service provider's proposed extensions/expansions policy "may have implications" for its approach. 601 Thus, the ERA's approach did derive some underpinning from those decisions. But, at least as a starting point, it cannot be shown to have erred in that regard. The underpinning first caused the ERA to remind itself of the unexceptional proposition that it should have regard to the national gas objective in making a decision on DBP's proposed extensions and expansions provisions when addressing the requirements of rules 48(1)(g) and 104 of the NGR: Final Decision at [1633]. 602 The real issue between DBP and the ERA is whether error in terms of s 246 of the NGL is shown by the way in which the ERA applied the national gas objective, in the context of the overall provisions of the NGL. 603 The ERA's reasons for its decision are best identified in [1634]-[1636] of the Final Decision. They provide: 1634. The Authority also considers that, in the current circumstances of the DBNGP, an election by DBP not to include an expansion of capacity as part of the covered pipeline is likely to result in outcomes that are contrary to the National Gas Objective and the coverage criteria under section 15 of the NGL(WA). The next significant expansion in capacity of the DBNGP is likely to be achieved by the completion of looping of the pipeline between compressor stations. The result of this is likely to be a decrease in the average cost of gas transmission when the increment to capacity becomes fully utilised. In the event that the expansion in capacity does not form part of the covered pipeline, there is a risk that the benefits of the expansion (in a reduced average cost of gas transmission) will not be passed on to all pipeline users with adverse consequences for competition in energy markets in Western Australia. 1635. The Authority is therefore concerned that the treatment of expansions under the proposed extension and expansion requirements is inconsistent with the National Gas Objective. 1636. The Authority considers that it would be more appropriate for the extension and expansion requirements to provide that the access arrangement will apply to incremental services to be provided as a result of any expansion in capacity of the DBNGP, except in instances where DBP can demonstrate to the Authority's reasonable satisfaction that application of the access arrangement to such services is inconsistent with the National Gas Objective. If DBP were to take the view at any time that an expansion of capacity should not form part of the covered pipeline, it is open to DBP to seek revocation of coverage of the relevant part of the DBNGP under the coverage provisions of the NGL(WA). 604 As noted above, there are no expansions forecast to take place during the operative period of the Access Arrangement Decision. DBP says that there was, therefore, no basis for presuming that the particular circumstances of a particular expansion should be treated as part of the covered pipeline. Allied with that proposition, it says that because the National Competition Council first has the role of recommending to the Minister whether a particular pipeline should be declared as a covered pipeline (in accordance with the criteria in s 15 of the NGL), the criteria in s 15 should by inference also be considered by the ERA in deciding whether to reject the DBP's proposed expansions proposal. 605 The fact that no particular extensions were planned for the regulatory period in issue does not relieve DBP or the ERA from proposing and providing respectively for how any such extensions should be addressed. It involved no error on the ERA's part to do so, and subject to the other contentions of DBP, to do so in a manner which conformed with its discretion under rule 40(3) of the NGR. 606 The Tribunal also does not accept that each of the pipeline coverage criteria in s 15 of the NGL must be considered by the ERA when deciding the terms for extensions and expansions in a revised access arrangement. Their context is different. The declaration of a pipeline as a covered pipeline obviously requires the existence of a pipeline, and the criteria in s 15 then specify a series of criteria which include the effect on competition of it being declared, the relative cost of an alternative pipeline being constructed (rather than access being granted) and the public interest. Section 18 of the NGL then applies in the event that a covered pipeline is extended or expanded: it directs attention to the "extension and expansion requirements" under the applicable access arrangement. They are defined in s 2 of the NGL. They are decided by the relevant regulator. There is no provision directing the regulator to consider the criteria in s 15. 607 The wider considerations concerning the public interest (beyond those expressed in the national gas objective) are not necessarily within the compass of the role of a regulator. Nor are, necessarily, the broader issues relating to competition, including overseas competition, or the economic capacity of a potential user of the expanded capacity of a covered pipeline to develop another pipeline to secure that capacity. Such matters extend beyond the role of the regulator, including the regulatory discretions under rule 40, and the regulatory limits on what a regulator may require under rule 104. Moreover, there is no basis to support the proposition, which is at least implicit in the DBP argument, that the ERA could not make a decision in the terms it did because it assumes that coverage should apply to an expansion without regard to the particular circumstances of the expansion. Rule 104(1) appears to contemplate such a determination, otherwise it would require that where no particular expansion is in contemplation, the requirements could not include coverage of the incremental services covered by an expansion, but (at least) must allow for later determination of that question. Rule 104(1) provides: Extension and expansion requirements may state whether the applicable access arrangement will apply to incremental services to be provided as a result of a particular extension to, or expansion of the capacity of, the pipeline or may allow for later resolution of that question on a basis stated in the requirements. 608 Rule 104(2) would also be otiose in those circumstances, or rather not apply until the later determination. 609 Alternative submissions by DBP are based on the use of the word "particular" in rule 104(1). It is said firstly that the decision of the ERA could not have been made because it does not provide for the circumstances of any particular (presently uncontemplated) expansion, and secondly that it effectively reverses the onus which otherwise would apply so that DBP must show that it is consistent with the national gas objective that the Access Arrangement Decision should not apply to that particular expansion. The onus is reversed because, it is said, the starting point is for the ERA to decide whether coverage should extend to an expansion, rather than for DBP to persuade the ERA that it should not. That submission, if accepted, would not necessarily lead to the Tribunal substituting the proposed extensions and expansions term put forward by DBP. 610 As to the first of those contentions, for similar reasons to those already given, the Tribunal considers that the Access Arrangement Decision could have addressed in the extensions and expansions requirements how expansions presently unidentified should be dealt with. It is clear that those requirements must be addressed, even though no particular extension or expansion is contemplated at the time. Otherwise, rules 48(1)(g) and 104 would have been expressed differently, and the definition of the "extensions and expansions" requirements in s 2 of the NGL would also have been expressed differently, to distinguish between extensions and expansions which are planned to be made during an access arrangement period, and those which occur during that period but were not planned at the start of that period. To state that is really to demonstrate that no such intention existed. The word "particular" in rule 104(1) will accommodate (as happens in this instance) the different treatment of extensions on the one hand and expansions on the other, as well as permitting differential provisions for different geographical areas or different types of expansions or in other respects. In the Tribunal's view, rule 104(1) does not preclude the ERA from making a provision which in fact applies equally to any and all extensions during a regulatory period. 611 Having taken that step, it is still necessary to determine whether error is made out in the manner in which the ERA prescribed how each expansion during the regulatory period would be addressed, that is by including it within the operation of the Access Arrangement Decision unless DBP satisfies the ERA that its coverage in that way is not consistent with the national gas objective. That stipulation, placing that onus on DBP, is said to be unreasonable and also contrary to the policy and intention of the NGL and the NGR, so that the discretion under s 40(3) miscarried. 612 In the view of the Tribunal, the ERA's decision does not involve reviewable error on that basis. Rule 104(1) of the NGR contemplates a provision in the Access Arrangement Decision which enables the ERA to prescribe whether and how the access terms will apply to any particular expansion - when and if one is undertaken - or to provide for the later resolution of such questions. In the event of a later resolution, DBP would be required to present material to the ERA in the light of which (and other material received by the ERA) the ERA would make a decision on those questions. The present ruling achieves no more than that. It is expressed in terms of DBP having the opportunity to claim that the particular expansion should not be included within the access arrangement terms, and then the onus of satisfying the ERA that its inclusion within those terms would be inconsistent with the national gas objective. 613 For the reasons the ERA gave, it could properly conclude that - as a starting point - the application of those terms to an expansion of the DBNGP would be consistent with the national gas objective. That is because the overall terms imposed by the Access Arrangement Decision (subject to other issues raised on this review) can reasonably be taken as consistent with that objective and the revenue and pricing principles. There is no reason why the ERA, in that context, should regard its starting point as a disincentive to efficient investment in, and efficient operation and use of, any expansion of the DBNGP. It has, however, preserved to DBP the opportunity to claim to the contrary in respect of any particular expansion, and to show to the ERA that in the particular circumstances the subjection of that expansion to the terms of the access arrangement would not be consistent with the national gas objective. So understood, the ERA has simply left the door open to DBP, if it contemplates an expansion during the access arrangement period, to present material to it which it may then review and decide whether, on the whole of the material, the national gas objective is not served by the inclusion of expansions in the scope of operation of the Access Arrangement Decision. As the ERA noted, DBP may also seek a coverage revocation determination under Part 4 Division 2 of the NGL in any event. Given the starting point of the ERA, the Tribunal does not consider that the ERA's decision in that respect is unreasonable or that it involves an incorrect exercise of its discretion. 614 Finally, the Tribunal notes the specific matters relied on by DBP which DBP argues, whether alone or collectively, demonstrate that the decision on the extensions and expansions requirements is unreasonable or involved incorrect exercise of the ERA's discretion. 615 The Tribunal has already rejected the proposition that the ERA could not have made its determination in the terms it did because it started with a "coverage" ruling, and then allowed itself to reconsider that starting point. That is first a matter of construction of the relevant provisions. It does not turn on any undue weight being given to decisions in the other cases referred to, although it may be observed that a similar result was reached in those cases on somewhat different regulatory provisions. 616 Nor does the Tribunal accept that it is erroneous for the ERA to have observed what is really self-evident, namely that the acceptance of the DBP alternative proposed would give it an unfettered discretion to include or exclude an expansion from the scope of the Access Arrangement Decision, with the risk that DBP's decision may result in an outcome not consistent with the national gas objective. That is not to attribute to DBP any sinister motives, but to point to a simple fact. Under the national gas objective and the NGL and NGR, the regulator is given specified responsibilities as an independent entity which, if DBP's proposal were accepted, it would be excluded from fulfilling in relation to any expansion during the regulatory period. 617 The observation of the ERA at [1634] of the Final Decision about the likely next significant expansion of the DBNGP also does not demonstrate reviewable error on its part. The ERA left open the consideration of any actual expansion when and if it occurs. If the circumstances of an expansion warrant it, then the use of that expansion and the capacity it offers may be excluded from the scope of the Access Arrangement Decision. 618 In the Tribunal's view, for those reasons, the ERA's decision on this topic does not involve reviewable error.

CONCLUSIONS 619 For the reasons given, DBP has succeeded in satisfying the Tribunal of error only in one respect, other than the error which it has acknowledged in the course of submissions. 620 The Tribunal found that the ERA was correct in using a bond-yield approach to determine the DRP, but erred because its choice of a value for the DRP based on its averaging procedure constitutes an incorrect exercise of its discretion, or was unreasonable: at [311]. For the reasons given, the Tribunal remits the Access Arrangement Decision to the ERA, to determine a value for the DRP using its bond-yield approach confined to its averaging procedure to conform to the reasons for decision of the Tribunal. The ERA and DBP are each given liberty to apply to the Tribunal for further directions on that issue. Naturally, the ERA will make such consequential alterations to the Access Arrangement Decision as are necessary as a consequence of any changed DRP input into the modelling required by rule 87(2) of the NGR. 621 The Tribunal accepts that the ERA, as it acknowledged, failed to determine correctly the present value of the minimum lease payments by adjusting the Base Rent in accordance with clause 4.3 of the BEP Lease to commencement of that lease. The Tribunal remits the Access Arrangement Decision to the ERA to correctly determine the present value of those minimum lease payments to give effect to that adjustment. Again, the remittal of the matter extends to the ERA making such consequential alterations to the Access Arrangement Decision as are necessary. 622 In all other respects the Tribunal affirms the decision of the ERA. I certify that the preceding six hundred and twenty-two (622) numbered paragraphs are a true copy of the Reasons for Decision herein of the Honourable Justice Mansfield (President) and Mr R Davey (Member) and Professor D Round (Member).