FEDERAL COURT OF AUSTRALIA

Australian Energy Regulator v Australian Competition Tribunal (No 2) [2017] FCAFC 79

Application from:

Applications by Public Interest Advocacy Service Ltd and Ausgrid Distribution [2016] ACompT 1

Applications by Public Interest Advocacy Service Ltd and Endeavour Energy [2016] ACompT 2

Applications by Public Interest Advocacy Service Ltd and Essential Energy [2016] ACompT 3

Application by ActewAGL Distribution [2016] ACompT 4

File numbers:

NSD 415 of 2016

NSD 416 of 2016

NSD 418 of 2016

NSD 419 of 2016

Judges:

BESANKO, YATES AND ROBERTSON JJ

Date of judgment:

24 May 2017

Catchwords:

ADMINISTRATIVE LAW application for judicial review of decision of the Australian Competition Tribunal (Tribunal) reviewing decisions of the Australian Energy Regulator (AER) – nature and scope of review by the Tribunal – whether the Tribunal erred in its construction of provisions in the National Electricity Rules relating to the determination of the rate of return on debt, the value of imputation credits and the operating expenditure criteria whether the Tribunal allowed the distribution network service providers to raise matters that were not raised and maintained by them in submissions to the AER, contrary to s 71O(2) of the National Electricity Law

Legislation:

Administrative Decisions (Judicial Review) Act 1977 (Cth) s 3, Sch 3

Competition and Consumer Act 2010 (Cth) s 103

Fair Work Act 2009 (Cth)

National Electricity Law ss 71A, 71B, 71C, 71O, 71P, 71R

Competition and Consumer Regulations 2010 (Cth)

National Electricity Rules

Cases cited:

Applicant WAEE v Minister for Immigration and Indigenous Affairs [2003] FCAFC 184; 236 FCR 593

Application by ActewAGL Distribution [2010] ACompT 4

Application by ActewAGL Distribution [2015] ACompT 3 Application by APT Allgas Energy Limited (No 2) [2012] ACompT 5

Application by ATCO Gas Australia Pty Ltd [2016] ACompT 10

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

Application by ElectraNet Pty Limited (No 3) [2008] ACompT 3

Application by Energex Limited (No 2) [2010] ACompT 7

Application by Energex Limited (Gamma) (No 5) [2011] ACompT 9

Application by EnergyAustralia [2009] ACompT 8

Application by Envestra Pty Limited (No 2) [2012] ACompT 3

Application by United Energy Distribution Pty Limited [2012] ACompT 1

Application by WA Gas Networks (No 3) [2012] ACompT 12

Applications by Public Interest Advocacy Centre Ltd, Ausgrid, Endeavour Energy and Essential Energy [2015] ACompT 2

Australian Competition and Consumer Commission v Australian Competition Tribunal [2006] FCAFC 83; (2006) 152 FCR 33

Australian Energy Regulator v Australian Competition Tribunal [2016] FCAFC 144

Chan Yee Kin v Minister for Immigration and Ethnic Affairs [1989] HCA 62; 169 CLR 379

Collector of Customs v Pozzolanic Enterprises Pty Ltd [1993] FCA 456; 43 FCR 280

East Australian Pipeline Pty Ltd v Australian Competition and Consumer Commission [2007] HCA 44; 233 CLR 229

House v R [1936] HCA 40; 55 CLR 499

Industry Research and Development Board v Bridgestone Australia Ltd [2001] FCA 954; 109 FCR 564

Minister for Aboriginal Affairs v Peko-Wallsend Ltd [1986] HCA 40; 162 CLR 24

Minister for Immigration and Citizenship v Li [2013] HCA 18; 249 CLR 332

Minister for Immigration and Ethnic Affairs v Wu Shan Liang [1996] HCA 6; 185 CLR 259

Norbis v Norbis [1986] HCA 17; 161 CLR 513

Pfizer Pty Ltd v Birkett [1999] FCA 1778

Pilbara Infrastructure Pty Limited v Australian Competition Tribunal [2011] FCAFC 58; 193 FCR 57

Pilbara Infrastructure Pty Limited v Australian Competition Tribunal [2012] HCA 36; 246 CLR 379

SPI Electricity Pty Ltd v Australian Competition Tribunal [2012] FCAFC 186; 208 FCR 151

Dates of hearing:

17, 18, 19, 20, 21 and 24 October 2016

Registry:

New South Wales

Division:

General Division

National Practice Area:

Commercial and Corporations

Sub-area:

Economic Regulator, Competition and Access

Category:

Catchwords

Number of paragraphs:

785

Counsel for the Applicant:

Mr S Lloyd SC with Mr MH O’Bryan QC, Mr S Balafoutis, Mr J Arnott and Ms T Phillips

Solicitor for the Applicant:

Corrs Chambers Westgarth

Counsel for the First Respondent:

The First Respondent submitted save as to costs

Counsel for the Second Respondent in

NSD415/2016

NSD416/2016

NSD418/2016:

Mr CA Moore SC with Ms K Richardson SC, Mr A Hochroth and Ms C Dermody

Solicitor for the Second Respondent in

NSD415/2016

NSD416/2016

NSD418/2016:

Herbert Smith Freehills

Counsel for the Second Respondent in

NSD419/2016:

Mr N Young QC with Mr AJ McClelland QC and Mr M Borsky

Solicitor for the Second Respondent in

NSD419/2016:

Gilbert + Tobin Lawyers

Counsel for the First Intervener:

Mr T Howe QC with Mr B Lim

Solicitor for the First Intervener:

Australian Government Solicitor

Counsel for the Second Intervener in

NSD415/2016

NSD416/2016

NSD418/2016:

Mr S Horgan QC with Mr T Clarke

Solicitor for the Second Intervener in

NSD415/2016

NSD416/2016

NSD418/2016:

Public Interest Advocacy Centre Ltd

ORDERS

NSD 415 of 2016

BETWEEN:

AUSTRALIAN ENERGY REGULATOR

Applicant

AND:

AUSTRALIAN COMPETITION TRIBUNAL

First Respondent

AUSGRID

Second Respondent

NSD 416 of 2016

BETWEEN:

AUSTRALIAN ENERGY REGULATOR

Applicant

AND:

AUSTRALIAN COMPETITION TRIBUNAL

First Respondent

ESSENTIAL ENERGY

Second Respondent

NSD 418 of 2016

BETWEEN:

AUSTRALIAN ENERGY REGULATOR

Applicant

AND:

AUSTRALIAN COMPETITION TRIBUNAL

First Respondent

ENDEAVOUR ENERGY

Second Respondent

NSD 419 of 2016

BETWEEN:

AUSTRALIAN ENERGY REGULATOR

Applicant

AND:

AUSTRALIAN COMPETITION TRIBUNAL

First Respondent

ACTEWAGL DISTRIBUTION (ABN 76 670 568 688)

Second Respondent

JUDGES:

BESANKO, YATES AND ROBERTSON JJ

DATE OF ORDER:

24 MAY 2017

THE COURT ORDERS THAT:

    The parties consult and, within 21 days, file orders in an agreed form to give effect to these reasons. Failing agreement, the parties are to file, within the same period, the orders for which they contend. The proposed orders are to include orders as to costs.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

INTRODUCTION

[1]

THE CENTRAL STATUTORY PROVISIONS

[10]

THE TRIBUNAL’S JURISDICTION AND POWERS

[17]

The submissions of the parties and of the interveners

[17]

Analysis

[133]

FORECAST OPERATING EXPENDITURE

[160]

Introduction

[160]

The National Electricity Rules

[164]

The Grounds for Review

[172]

The Tribunal’s Reasons

[182]

The Amendments to Rule 6.5.6 in 2012

[184]

The Econometric Models

[188]

The Principal Issue

[193]

An Overview of the Electricity Network Respondents’ Challenges to the AER’s Estimate of the Required Opex

[194]

The AER’s Approach

[198]

Criticisms of the data used in the EI Model

[216]

Criticisms of the AER’s decision to lower the EI Model’s Comparison Point

[229]

Criticisms of the OEF Adjustments made by the AER

[232]

Criticisms of the AER’s approach to Vegetation Management Costs

[235]

The Criticisms of the AER’s approach to the Labour Costs of the NSW Service Providers

[251]

The Electricity Network Respondents’ submission that the AER used the EI Model as the Sole Determinative of Opex

[257]

The Tribunal’s consideration of the “Principal Opex Issue”

[260]

Analysis of the AER’s Grounds of Review

[272]

The Tribunal’s approach to the AER’s use of the EI Model and the Tribunal’s underlying concerns (Grounds 6, 7, 9 and 21)

[279]

The Tribunal’s approach to the Labour Costs of the Electricity Network Respondents (Ground 8)

[356]

Whether the Tribunal’s reasons and orders are uncertain and unreasonable (Grounds 10 and 11)

[373]

Conclusion

[386]

RETURN ON DEBT

[387]

Introduction

[387]

The National Electricity Rules

[389]

The AER’s final decisions

[399]

The Tribunal’s decision

[415]

The operation of s 71O(2)(a)

[420]

A regulated and common entity for all service providers?

[430]

The correct transitioning arrangement?

[439]

The grounds of review

[451]

Whether the Tribunal erred by receiving and acting upon a matter that was not raised and maintained by the service providers in their submissions to the AER

[453]

The AER’s submissions

[453]

The electricity network respondents’ submissions

[465]

The AER’s reply submissions

[478]

Analysis and conclusion

[482]

Whether the Tribunal erred in concluding that the AER was bound to address the ROR Objective on the basis that the benchmark efficient entity must be an unregulated entity

[491]

Introduction

[491]

The AER’s submissions

[497]

PIAC’s submissions

[514]

The electricity network respondents’ submissions

[517]

The AER’s reply submissions

[526]

Analysis and conclusion

[529]

Whether the Tribunal erred in its construction of r 6.5.2(k)(4) of the NER

[546]

The AER’s submissions

[546]

The electricity network respondents’ submissions

[551]

The AER’s reply submissions

[556]

Analysis and conclusion

[557]

Whether the Tribunal’s conclusions as to what is required to be done to comply with rr 6.5.2(c) and (k) reveal error

[576]

The AER’s submissions

[576]

The electricity network respondents’ submissions

[587]

Analysis and conclusion

[590]

Whether the Tribunal failed properly to undertake a review of the AER’s decision on return on debt

[597]

The AER’s submissions

[597]

The electricity network respondents’ submissions

[607]

The AER’s reply submissions

[614]

Analysis and conclusion

[615]

Conclusion

[617]

THE VALUE OF IMPUTATION CREDITS (GAMMA)

[618]

The Tribunal’s Reasons

[618]

The grounds of review relevant to gamma

[648]

The AER’s submissions

[649]

Grounds 16 and 18

[655]

Ground 17

[674]

Ground 19

[681]

Ground 20

[691]

The electricity network respondents’ submissions

[696]

The AER’s reply submissions

[732]

Analysis

[738]

Orders

[785]

THE COURT:

INTRODUCTION

1    Before the Court are five applications for judicial review brought by the Australian Energy Regulator (the AER) in relation to five determinations of the Australian Competition Tribunal (the Tribunal).

2    Those determinations, and the Tribunals accompanying reasons, were as follows: Applications by Public Interest Advocacy Centre Ltd and Ausgrid [2016] ACompT 1 (ACT 1 of 2015, ACT 4 of 2015) (Ausgrid); Applications by Public Interest Advocacy Centre Ltd and Endeavour Energy [2016] ACompT 2 (ACT 2 of 2015, ACT 6 of 2015); Applications by Public Interest Advocacy Service Ltd and Essential Energy [2016] ACompT 3 (ACT 3 of 2015); Application by ActewAGL Distribution [2016] ACompT 4 (ACT 5 of 2015) (together NSD 415 of 2016, NSD 416 of 2016, NSD 418 of 2016 and NSD 419 of 2016); and Application by Jemena Gas Networks (NSW) Ltd [2016] ACompT 5 (ACT 8 of 2015) (NSD 420 of 2016).

3    These reasons concern the first four of these matters, which involve the National Electricity Law (NEL) and the National Electricity Rules (NER), and the common issues arising in the fifth matter in this list, NSD 420 of 2016, which involves the National Gas Law (NGL) and the National Gas Rules (NGR). Separate reasons, published today, deal with the return on debt issues specific to Jemena Gas Networks (NSW) Ltd (JGN), NSD 420 of 2016.

4    The determination of the Tribunal in the first of these matters (Ausgrid), which the parties treated as the lead determination by the Tribunal, was in the following terms:

1.    Pursuant to s 71P(2)(c) of the National Electricity Law, the Final Decision Ausgrid distribution determination 2015-16 to 2018-19, April 2015, including attachments, (the Final Decision) is set aside and remitted to the Australian Energy Regulator (AER) to make the decision again in accordance with the following directions:

(a)    the AER is to make the constituent decision on opex under r 6.12.1(4) of the National Electricity Rules in accordance with these reasons for decision including assessing whether the forecast opex proposed by the applicant reasonably reflects each of the operating expenditure criteria in r 6.5.6(c) of the National Electricity Rules including using a broader range of modelling, and benchmarking against Australian businesses, and including a “bottom up” review of Ausgrid’s forecast operating expenditure;

(b)    the AER is to make the constituent decision on return on debt in relation to the introduction of the trailing average approach in accordance with these reasons for decision;

(c)    the AER is to make the constituent decision on estimated cost of corporate income tax (gamma) in accordance with these reasons for decision, including by reference to an estimated cost of corporate income tax based on a gamma of 0.25; and

(d)    the AER is to consider, and to the extent to which it considers appropriate to vary the Final Decision in such other respects as the Australian Energy Regulator considers appropriate having regard to s 16(1)(d) of the National Electricity Law in the light of such variations as are made to the Final Decision by reason of (a)-(c) hereof.

5    Each application to this Court for judicial review was made under the Administrative Decisions (Judicial Review) Act 1977 (Cth) (the ADJR Act): see paragraphs (ca) and (cb) of the definition of “enactment” in s 3(1) and items 2(d), (daa) and (da) of Sch 3 to the ADJR Act.

6    As will be seen, this Court is asked to find that the Tribunal made judicially reviewable errors in the course of reviewing, on (limited) merits review grounds, the AER’s Final Decisions.

7    These judicial review proceedings are in the original jurisdiction of the Court, although that jurisdiction has been directed by the Chief Justice to be exercised by a Full Court: s 20(1A) of the Federal Court of Australia Act 1976 (Cth).

8    As set out in the AER’s written submissions, the AER’s challenge to the Tribunal’s determinations was directed principally to the following core concerns:

(a)    The Tribunal had failed to undertake its review function lawfully by failing to properly construe and apply the grounds of review under s 71C of the NEL and s 246 of the NGL. Errors of this kind led the Tribunal to carry out reviews of a kind that were not authorised by the legislation.

(b)    In one instance, the Tribunal purported to review a decision of a type that did not and could not fall within its jurisdiction. This related only to the fifth matter, NSD 420 of 2016, which, as we have said, involves JGN and the NGL and the NGR and is the subject of separate reasons published today.

(c)    The Tribunal allowed the distribution network service providers and the covered pipeline service provider to raise, in relation to whether a ground of review existed, matters that were not raised and maintained by the service providers in submissions to the AER before the reviewable regulatory decisions were made, thus contravening the constraints imposed by s 71O(2) of the NEL and s 258(3) of the NGL.

(d)    The Tribunal erred in its construction of new provisions in the NER and the NGR relating to the determination of the rate of return on capital, the value of imputation credits and the operating expenditure criteria.

(e)    The Tribunal made other reviewable errors in making its decision, including adopting reasoning that was irrational, unreasonable and/or uncertain.

9    It should also be noted that on 17 October 2016 the Court made the following orders on applications to intervene: see Australian Energy Regulator v Australian Competition Tribunal [2016] FCAFC 144:

1.    On the application of the Minister for the Environment and Energy to intervene:

Until further order,

(a)    Pursuant to r 9.12(2) of the Federal Court Rules 2011 (Cth), the Minister have leave to intervene in each of the applications, that leave being limited to the making of the written submissions on construction filed by him on 30 September 2016.

(b)    There be no order for costs in respect of the Minister’s application to intervene or his intervention.

2.    On the application of the Public Interest Advocacy Centre Ltd (PIAC) to intervene:

Until further order,

(a)    Pursuant to r 9.12(2) of the Federal Court Rules 2011 (Cth), PIAC have leave to intervene in applications NSD 415/2016, NSD 416/2016 and NSD418/2016 that leave being limited to the making of the written submissions filed by it on 30 September 2016.

(b)    There be no order for costs in respect of PIAC’s application to intervene or its intervention.

THE CENTRAL STATUTORY PROVISIONS

10    The relevant provisions of the NEL are as follows. We do not set out the equivalent provisions of the NGL, principally ss 245, 246, 258, 258A, 259 and 261.

11    Section 71B provides:

71B—Applications for review

(1)    An affected or interested person or body, with the leave of the Tribunal, may apply to the Tribunal for a review of a reviewable regulatory decision.

(2)    An application must—

(a)    be made in the form and manner determined by the Tribunal; and

(b)     specify the grounds for review being relied on.

12    Section 71C provides:

71C—Grounds for review

(1)    An application under section 71B(1) may be made only on 1 or more of the following grounds:

(a)    the AER made an error of fact in its findings of facts, and that error of fact was material to the making of the decision;

(b)    the AER made more than 1 error of fact in its findings of facts, and that those errors of fact, in combination, were material to the making of the decision;

(c)    the exercise of the AERs discretion was incorrect, having regard to all the circumstances;

(d)    the AER’s decision was unreasonable, having regard to all the circumstances.

(1a)    An application under section 71B(1) must also specify the manner in which a determination made by the Tribunal varying the reviewable regulatory decision, or setting aside the reviewable regulatory decision and a fresh decision being made by the AER following remission of the matter to the AER by the Tribunal, on the basis of 1 or more grounds raised in the application, either separately or collectively, would, or would be likely to, result in a materially preferable NEO decision.

(2)    It is for the applicant to establish a ground listed in subsection (1) and the matter referred to in subsection (1a).

13    The term “materially preferable NEO decision” is defined in s 71A by reference to71P(2a)(c), which we set out below.

14    Section 71O provides:

71O—Matters that may and may not be raised in a review

(1)    The AER, in a review under this Subdivision, may—

(a)    respond to any matter raised by the applicant or an intervener; and

(b)    raise any other matter that relates to—

(i)    a ground for review; or

(ii)    a matter raised in support of a ground for review; or

(iii)    a matter relevant to the issues to be considered under section 71P(2a) and (2b).

(2)    In a review under this Subdivision, the following provisions apply in relation to a person or body, other than the AER (and so apply at all stages of the proceedings before the Tribunal):

(a)    a regulated network service provider to whom the reviewable regulatory decision being reviewed applies may not raise in relation to the issue of whether a ground for review exists or has been made out any matter that was not raised and maintained by the provider in submissions to the AER before the reviewable regulatory decision was made;

(b)    a regulated network service provider whose commercial interests are materially affected by the reviewable regulatory decision being reviewed may not raise in relation to the issue of whether a ground for review exists or has been made out any matter that was not raised and maintained by the provider in submissions to the AER before the reviewable regulatory decision was made;

(c)    an affected or interested person or body (other than a provider under paragraph (a) or (b)) may not raise in relation to the issue of whether a ground for review exists or has been made out any matter that was not raised by the person or body in a submission to the AER before the reviewable regulatory decision was made;

(d)    subject to paragraphs (a), (b) and (c)—

(i)    the applicant, or an intervener who has raised a new ground for review under section 71M, may raise any matter relevant to the issues to be considered under section 71P(2a) and (2b); and

(ii)    any person or body, other than the applicant or an intervener who has raised a new ground for review under section 71M, may not raise any matter relevant to the issues to be considered under section 71P(2a) and (2b) unless it is in response to a matter raised by—

(A)    the AER under subsection (1)(b)(iii); or

(B)    the applicant under subparagraph (i); or

(C)     an intervener under subparagraph (i).

(3)    For the purposes of subsection (2)(d)—

(a)    a reference to an applicant includes a reference to a person or body who has applied to the Tribunal for leave to apply for a review under this Subdivision; and

(b)    a reference to an intervener includes a reference to a person or body who has applied to the Tribunal for leave to intervene in a review under this Subdivision.

15    Section 71P relevantly provides:

71P—Tribunal must make determination

(1)    If, following an application, the Tribunal grants leave in accordance with section 71B(1), the Tribunal must make a determination in respect of the application.

Note—

See section 71Q for the time limit within which the Tribunal must make its determination.

(2)    Subject to subsection (2a), a determination under this section may—

(a)    affirm the reviewable regulatory decision; or

(b)    vary the reviewable regulatory decision; or

(c)    set aside the reviewable regulatory decision and remit the matter back to the AER to make the decision again in accordance with any direction or recommendation of the Tribunal.

(2a)    Despite subsection (2), the Tribunal may only make a determination—

(a)    to vary the reviewable regulatory decision under subsection (2)(b); or

(b)    to set aside the reviewable regulatory decision and remit the matter back to the AER under subsection (2)(c),

if—

(c)    the Tribunal is satisfied that to do so will, or is likely to, result in a decision that is materially preferable to the reviewable regulatory decision in making a contribution to the achievement of the national electricity objective (a materially preferable NEO decision) (and if the Tribunal is not so satisfied the Tribunal must affirm the decision); and

(d)    in the case of a determination to vary the reviewable regulatory decision—the Tribunal is satisfied that to do so will not require the Tribunal to undertake an assessment of such complexity that the preferable course of action would be to set aside the reviewable regulatory decision and remit the matter to the AER to make the decision again.

(2b)    In connection with the operation of subsection (2a) (and without limiting any other matter that may be relevant under this Law)—

(a)    the Tribunal must consider how the constituent components of the reviewable regulatory decision interrelate with each other and with the matters raised as a ground for review; and

(b)    without limiting paragraph (a), the Tribunal must take into account the revenue and pricing principles (in the same manner in which the AER is to take into account these principles under section 16); and

(c)    the Tribunal must, in assessing the extent of contribution to the achievement of the national electricity objective, consider the reviewable regulatory decision as a whole; and

(d)    the following matters must not, in themselves, determine the question about whether a materially preferable NEO decision exists:

(i)    the establishment of a ground for review under section 71C(1);

(ii)    consequences for, or impacts on, the average annual regulated revenue of a regulated network service provider;

(iii)    that the amount that is specified in or derived from the reviewable regulatory decision exceeds the amount specified in section 71F(2).

(2c)    If the Tribunal makes a determination under subsection (2)(b) or (c), the Tribunal must specify in its determination—

(a)    the manner in which it has taken into account the interrelationship between the constituent components of the reviewable regulatory decision and how they relate to the matters raised as a ground for review as contemplated by subsection (2b)(a); and

(b)    in the case of a determination to vary the reviewable regulatory decision—the reasons why it is proceeding to make the variation in view of the requirements of subsection (2a)(d).

(3)    For the purposes of making a determination of the kind in subsection (2)(a) or (b), the Tribunal may perform all the functions and exercise all the powers of the AER under this Law or the Rules.

(5)    A determination by the Tribunal affirming, varying or setting aside the reviewable regulatory decision is, for the purposes of this Law (other than this Part), to be taken to be a decision of the AER.

16    Section 71R provides:

71R—Matters to be considered by Tribunal in making determination

(1)    Subject to this section, the Tribunal, in acting under this Division with respect to a reviewable regulatory decision—

(a)    must not consider any matter other than review related matter (and any matter arising as a result of consultation under paragraph (b)); and

(b)    must, before making a determination, take reasonable steps to consult with (in such manner as the Tribunal thinks appropriate)—

(i)    network service users and prospective network service users of the relevant services; and

(ii)    any user or consumer associations or user or consumer interest groups,

that the Tribunal considers have an interest in the determination, other than a user or consumer association or a user or consumer interest group that is a party to the review.

(3)    If in a review the Tribunal is of the view that a ground for review has been made out, the Tribunal may, on application by a party to the review, allow new information or material to be submitted if the party can establish to the satisfaction of the Tribunal that the information or material—

(a)    was publicly available or known to be available to the AER when it was making the reviewable regulatory decision; or

(b)    would assist the Tribunal on any aspect of the determination to be made and was not unreasonably withheld from the AER when it was making the reviewable regulatory decision,

and was (in the opinion of the Tribunal) information or material that the AER would reasonably have been expected to have considered when it was making the reviewable regulatory decision.

(4)    Subject to this Law, for the purpose of subsection (3)(b), information or material not provided to the AER following a request for that information or material by it under this Law or the Rules is to be taken to have been unreasonably withheld.

(5)    Subsection (4) does not limit what may constitute an unreasonable withholding of information or material.

(5a)    In addition, if in a review the Tribunal is of the view—

(a)    that a ground for review has been made out; and

(b)    that it would assist the Tribunal to obtain information or material under this subsection in order to determine whether a materially preferable NEO decision exists,

the Tribunal may, on its own initiative, take steps to obtain that information or material (including by seeking evidence from such persons as it thinks fit).

(5b)    The action taken by a person acting in response to steps taken by the Tribunal under subsection (5a) must be limited to considering decision related matter under section 28ZJ.

(6)    In this section—

review related matter means—

(a)    the application for review; and

(b)    a notice raising new grounds for review filed by an intervener; and

(c)     the submissions made to the Tribunal by the parties to the review; and

(d)    decision related matter under section 28ZJ; and

(e)    any other matter properly before the Tribunal in connection with the relevant proceedings.

THE TRIBUNAL’S JURISDICTION AND POWERS

The submissions of the parties and of the interveners

17    We shall set out the competing submissions at some length since this is the first occasion on which a Full Court has been required to construe the grounds for review in the NEL and the NGL in their present form. As will be seen, there is a limit to the usefulness of construing terms such as “findings of facts”, “error of fact”, “exercise of… discretion” and “unreasonable” in the abstract.

18    The AER submitted that the legislature had confined the extent to which the merits of the AER’s determinations may be reconsidered, by reference to the grounds enumerated in s 71C(1) of the NEL (and s 246(1) of the NGL): see Application by WA Gas Networks (No 3) [2012] ACompT 12 (WA Gas Networks) at [22] and the Tribunal cases there cited. So much was common ground.

19    The AER submitted that the first two grounds of review could only be established by an applicant satisfying the Tribunal of three elements: (a) identifying the AER’s findings of fact, in so doing it is “crucial to ascertain precisely what the conclusion of fact was” (Application by Envestra Pty Limited (No 2) [2012] ACompT 3 (Envestra) at [138]); (b) determining whether, within those findings, there are one or more facts that can properly be characterised as constituting an error or errors of fact; and (c) if so, determining whether the error was material to the making of the decision.

20    As to the first element, the AER submitted it was readily apparent that the NEL and NGL were not premised upon any simple fact/law dichotomy. The reference to “discretion” in the third ground excluded such an approach. The concept of a “fact” in this legislative context did not embrace everything that was not “law”. For example, a finding of fact did not extend to the making of choices between permitted methodologies or the attaching of weight to competing regulatory considerations: (Australian Competition and Consumer Commission v Australian Competition Tribunal [2006] FCAFC 83; (2006) 152 FCR 33 (ACCC v ACT) at [172]; Envestra at [30]).

21    It was uncontroversial that the expression “findings of fact” extended to findings as to the existence of historical or present facts comprised by events or circumstances: ACCC v ACT at [171].

22    In obiter dicta, ACCC v ACT (at [171]) suggested that the expression “should be interpreted broadly” so as to extend to “an opinion about the existence of a future fact or circumstance”. This conclusion was premised upon the observation that the regulator’s functions involved not only making assessments of historical and present facts but also assessing expert opinion on various matters pertaining to likely future circumstances. It was said by the service providers that the “error of fact” ground of review should cover opinions of future facts or circumstances in order to make the review “meaningful” in relation to the regulator’s function. The AER submitted that this analysis did not arise from the language of the legislation or the intention of having limited merits review. Further, the Full Court’s approach did not address case law that distinguished between “matters of fact” and “expressions of opinion”: Pfizer Pty Ltd v Birkett [1999] FCA 1778 at [11]. In any event, these obiter statements were made in the context of provisions that were not complete analogues to s 71C. The AER submitted that opinions about the existence of future facts (in other words, forecasts) were properly reviewed under the “unreasonableness” ground of review in s 71C.

23    In Application by ActewAGL Distribution [2010] ACompT 4 (ActewAGL) referring to the ACCC v ACT decision of the Full Court, the Tribunal at [32] described the Court’s characterisation of an opinion as comprising a finding of fact as involving “a radical meaning to be given to the word ‘fact’”, observing that “the generally accepted view is that an opinion is an inference which is drawn from facts”. Although there will be many instances where it is “difficult (if not impossible) to distinguish between ‘fact’ and ‘opinion’”, the Tribunal in ActewAGL considered, by reference to examples, that the difference was “between a more concrete and specific form of descriptive statement and a less specific and concrete form”.

24    The AER contended that the Tribunal in ActewAGL was rightly sceptical of the characterisation of opinions as findings of fact for the purposes of s 71C(1)(a) and (b) of the NEL. It was submitted that this Court should not apply the Full Court’s obiter on this issue to s 71C. (It was obiter because the ACCC sought judicial review in the Full Court of the Tribunal’s conclusion that the ACCC’s exercise of discretion had been incorrect or unreasonable. The Tribunal did not find an error of fact: see ACCC v ACT, [181]; on appeal to the High Court: East Australian Pipeline Pty Ltd v Australian Competition and Consumer Commission [2007] HCA 44; 233 CLR 229, 249 [78] (EAPL).)

25    As to the second element of the first two grounds of review, the AER submitted that when the Tribunal considered whether there was an “error of fact”, the concept of “error” set a threshold that was not passed simply because there was material that could support a different finding of fact or simply because the Tribunal might, if it considered the material afresh, prefer to make a different finding of fact: Application by DBNGP (WA) Transmission Pty Ltd (No 3) [2012] ACompT 14 (DBNGP) at [326] and WA Gas Networks at [22]. Before a fact could be characterised as erroneous, more had to be established by a review applicant: such as the absence of evidence supporting the finding made. Further, “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 was not shown simply because one expert opinion or set of opinions had been preferred over another expert opinion or set of opinions.

26    As to the third element of the first two grounds of review, the AER submitted that the applicant also had to establish that the error (or the errors in combination) was (or were) “material”. In Envestra, the Tribunal (at [32]) defined an error of fact as material “if the relevant decision of the AER depends, or is based, on the error”, citing the decision of Jagot J in Ibrahim v Minister for Immigration and Citizenship [2009] FCA 1328 at [8]. The reasoning of her Honour in that case was consistent with higher authority as to the meaning of “material”, to which her Honour referred at [10]-[13].

27    The AER submitted that the third ground of review – incorrect exercise of discretion – did not concern findings of fact made by the AER; rather, it related to matters in respect of which the AER had made a choice under the NEL and NER (or the NGL and NGR): ACCC v ACT at [173]. The Full Court implicitly (and correctly) invoked the applicable principles for review of discretionary decisions in House v R [1936] HCA 40; 55 CLR 499 at 504-5 (House). On appeal in EAPL, Gummow and Hayne JJ observed that the first part of the ground of review in s 39(2)(a), that the exercise of the ACCC’s discretion was incorrect, “should be understood as encompassing the words in House, ‘[i]f the judge acts upon a wrong principle, if he allows extraneous or irrelevant matters to guide or affect him if he does not take into account some material consideration’” (at [79]).

28    The AER submitted that if the AER had exercised its discretion on correct principles and if the particular exercise of discretion was open to it, the Tribunal was not empowered to set aside that decision because it thought another decision would have been preferable: ACCC v ACT at [176]; Envestra at [34].

29    As to the ambit of the fourth ground of review, established when the AER’s decision was shown to be unreasonable, the AER submitted that in Envestra, citing Application by ElectraNet Pty Limited (No 3) [2008] ACompT 3 (ElectraNet (No 3)) at [74], the Tribunal (at [48]) considered the ambit of this ground and stated that the “unreasonableness must be of the AER’s decision itself”, in contrast to a step in the reasoning leading to that decision. In other words, “it is the AER’s decision which must be unreasonable having regard to all the circumstances before that ground is enlivened”. The Tribunal in ElectraNet (No 3) referred with approval in this context to the statements of the Full Court regarding unreasonableness in ACCC v ACT at [178], and the Tribunal’s endorsement of that statement in ElectraNet (No 3) at [65] and in Application by EnergyAustralia [2009] ACompT 8 (EnergyAustralia) at [66]–[67].

30    The AER submitted that it was necessary, and reasonable, for it to form opinions (for example, in making forecasts) and to rely upon expert opinions. Where there was conflicting expert opinion, the AER may be required to choose between the opinions. Such a decision would only be unreasonable if the choice lacked reason or the underlying opinion lacked reason. No error was established simply by reason that the Tribunal might have chosen differently: ACCC v ACT at [172]-[176].

31    The AER also made submissions on constraints on matters that may be raised in support of grounds of review. It submitted that at “all stages” of the proceedings before the Tribunal – including leave – a party (other than the AER) was restricted in the submissions it may make to the Tribunal by the operation of s 71O(2) NEL, and s 258A(3) NGL. In summary, a party was not permitted to raise any matter that was not raised and maintained in submissions to the AER when it made its decision the subject of the review. In SPI Electricity Pty Ltd v Australian Competition Tribunal [2012] FCAFC 186; 208 FCR 151 (SPI Electricity), the Court considered that, as long as a matter was raised, review was not precluded “merely because it may have been raised more forcefully or more compellingly” (at [65]). Since that decision, however, the addition of the words “and maintained” to s 71O(2) and s 258A(3) indicated a legislative intention to focus attention upon the degree to which a submission was pursued throughout the assessment process.

32    The AER submitted that an applicant must have “maintained” its position in substance, not form. Section 71O(2) of the NEL and s 258A(3) of the NGL required an applicant actively to involve itself with the regulator’s stated views throughout the decision making process. That process would not be meaningful if parties could pay lip service to a particular contention and then seek the Tribunal’s intervention. The requirement that all contentions be raised and maintained before the AER ensured that all stakeholders had a proper opportunity to make submissions and provide evidence to the AER with respect to them.

33    The AER also made submissions on the “materially preferable NEO decision” test in its legislative context. Prior to the 2013 Act, the AER submitted, the Tribunal was empowered to set aside or vary a reviewable regulatory decision or remit the matter to the AER to make the decision again in accordance with a direction or recommendation only if one or more grounds of review were established. Subsequent to the 2013 Act, after reaching a state of satisfaction in relation to the existence of grounds of review, and possibly after considering further “information or material” (see s 71R(5a) of the NEL, s 261(3b) of the NGL), the Tribunal must consider whether s 71P(2a)(c) of the NEL or s 261(4a)(c) of the NGL had been satisfied before it can set aside or vary the AER’s decision. The “only … if” formulation was unambiguous: it posited a precondition on an exercise of power.

34    The AER submitted that s 71P(2a)(c) required the Tribunal to consider the specific decision that it would be inclined to make as a result of the establishment of one or more grounds of review and to compare that proposed decision with the AER’s reviewable regulatory decision. If the correction of errors revealed by established grounds of review necessarily resulted in a materially preferable National Electricity Objective (NEO) decision, then s 71P(2a) would have no work to do.

35    The expression “materially preferable” used ordinary English words and was not a technical expression. Having regard to dictionary definitions, the combined concept referred to something being “considerably more suitable or desirable” or “more suitable or desirable to an important degree”. In its broader context, this language meant that the Tribunal was required to consider whether its proposed decision would be likely to be considerably better suited to contributing to the achievement of the NEO than the AER’s decision. The Tribunal was not to assume that correcting an error of fact or redressing a departure from the NER necessarily resulted in a better decision; it needed to be satisfied on evidence that as between the two or more decisions the proposed decision would be likely to advance the achievement of the NEO to an important degree; if not so satisfied, there could be no change to the AER’s decision (even if it contained reviewable error).

36    Section 71P(2b) of the NEL identified three mandatory relevant considerations when considering whether a proposed Tribunal decision was materially preferable (in the relevant sense) to the AER decision. First, the Tribunal was required to consider how the constituent components of the reviewable regulatory decision interrelated with each other and with the matters raised (presumably successfully) as a ground of review: s 71P(2b)(a). This consideration was facilitated by the AER’s reasons which had to address this inter-relationship: s 16(1)(c). Secondly, the Tribunal had to consider the revenue and pricing principles (RPPs) (s 7A) in the same manner in which the AER was to take them into account: s 71P(2b)(b). That was to say, they were matters that had to be taken into account in certain circumstances and that could be taken into account in other circumstances (s 16(2)). In no circumstances were the principles binding constraints in their own terms: see Rathborne v Abel (1964) 38 ALJR 293 at 295 (per Barwick CJ) and 301 (per Kitto J); Minister for Aboriginal Affairs v Peko-Wallsend Ltd [1986] HCA 40; 162 CLR 24 at 41 (per Mason J, Gibbs CJ and Dawson J agreeing); Foster v Minister for Customs and Justice [2000] HCA 38; 200 CLR 442 at [22]–[23] (per Gleeson CJ and McHugh J) and at [34], [35] and [38] (per Gaudron and Hayne JJ); A v Corruption and Crime Commissioner [2013] WASCA 288; 306 ALR 491 at [90]-[92], [245]- [246], [248] (per Martin CJ, McLure P and Murphy JA). They were considerations which were useful to the extent that they assisted in “balancing” and “informing” how the NEO may be better achieved “in the context of ensuring the service provider acts efficiently”: see DBNGP at [77]. Thirdly, the Tribunal was required to, in assessing the extent of contribution to the achievement of the NEO, consider the reviewable regulatory decision as a whole: s 71P(2b)(c).

37    The AER submitted that these provisions were designed to meet a concern about review applicants “cherry picking” parts of reviewable regulatory decisions. The Standing Council on Energy and Resources (SCER) noted that a continuation of the pre-existing merits review regime “has the potential to embed cherry picking of decisions and gaming of the regime for future determinations”. The SCER considered that “such an outcome is inconsistent with ensuring that overall regulatory determinations are in the long term interests of consumers”.

38    In addition to the (non-exhaustive) guidance provided in s 71P(2b)(a)-(c) as to the matters that could be taken into account in determining whether the Tribunal’s decision would lead to a materially preferable NEO decision, s 71P(2b)(d) provided guidance as to considerations that were not, in themselves, sufficient to justify a conclusion that a decision to be varied by the Tribunal would be materially preferable to the AER’s decision (in relation to contributing to the achievement of the NEO).

39    The AER submitted that something more than these three matters needed to be established for the Tribunal to reach the requisite state of satisfaction. This was hardly surprising. The first of them must exist in every case in which the Tribunal would be inclined to vary or set aside a decision by the AER: if no ground of review had been established, no relief would be forthcoming. Paragraph (iii) served to emphasise that the materially preferable test was not a re-statement of the financial threshold imposed on leave applications (see s 71F). Exceeding that threshold was not itself sufficient to justify varying or setting aside the AER’s decision. The materially preferable NEO decision test was doing different, separate and additional work to s 71F.

40    The AER submitted that it was also significant that the consequences for, or impacts on, the average annual regulated revenue of a service provider was a factor that was not in itself sufficient to determine the question of whether the Tribunal’s proposed decision was a materially preferable NEO decision. An applicant must go beyond observing the quantity of the revenue that was at stake. Relying only upon materials permitted under s 71R, an applicant must establish why the achievement of the NEO was more likely than not to be contributed to as a consequence of the adoption of a decision that would rectify any grounds of review that had been established.

41    The AER submitted that in the context of this legislation, the use of the language “will, or is likely to”, in connection with whether a varied or substituted decision would be a materially preferable NEO decision, required the Tribunal to reach a state of satisfaction on the balance of probabilities. It could not have been the intention of the legislature that the AER’s decision be replaced simply because there was a mere prospect of improvement that was not remote, fanciful or farfetched.

42    The respondents Ausgrid, Endeavour Energy (Endeavour), Essential Energy (Essential) (referred to by the Tribunal as Networks NSW but which we refer to in these reasons as NSW service providers) and ActewAGL, (sometimes referred to as Distribution Network Service Providers (DNSPs) but which we refer to in these reasons as the electricity network respondents), submitted that the AER’s written submissions raised a large number of contentions relating to the merits of the Tribunal’s decision. Many of the matters raised were not properly grounds of judicial review, but rather were argumentative assertions about the Tribunal’s analysis of factual matters and the contentions of the parties.

43    The electricity network respondents submitted that one theme running through the AER’s submissions was an allegation that the Tribunal had not identified the errors in the AER’s decision with sufficient specificity. This assertion resulted from significant mischaracterisations of the Tribunal’s reasoning and conclusions. Those mischaracterisations took various forms. They underlay each of the five alleged “core concerns” referred to in the AER’s submissions (see [8] above), such that none of those concerns was justified.

44    More generally, the electricity network respondents submitted that the AER’s submissions failed to do justice to the lengthy and detailed reasons of the Tribunal which carefully analysed the evidence and the extensive submissions of the parties on the issues it determined, in the context of a legislative requirement to deliver its reasons on an expedited basis: NEL s 71Q. The Tribunal was a specialist tribunal constituted by a judge and two experts, making it well suited to deal with complex questions of fact and economic regulation: Pilbara Infrastructure Pty Limited v Australian Competition Tribunal [2011] FCAFC 58; 193 FCR 57 at [15]. As noted in Pilbara Infrastructure, it was not the Court’s function to resolve the difficult and complex matters of judgment raised by the evidence and resolved by the Tribunal. The AER in the present case, like the applicants in Pilbara Infrastructure, invited the Court to engage in what the Full Court in that case referred to as the “slide into impermissible merit review” at [17].

45    In relation to the scope of merits review and the grounds of review, the electricity network respondents submitted the Tribunal’s decision formed part of the “review” that the Tribunal was required to undertake under Pt 6, Div 3A Subdiv 2 of the NEL (“Merits review for reviewable regulatory decisions”) (Subdivision 2). The powers of the Tribunal to review regulatory decisions made by the AER under the NER were derived from the NEL.

46    Of further relevance was s 103(1)(a)-(b) of the Competition and Consumer Act 2010 (Cth), which provided that the procedure of the Tribunal ‘shall be conducted with as little formality and technicality, and with as much expedition, as the requirements of this Act and a proper consideration of the matters before the Tribunal permit’. Section 44ZZR of the CCA applied s 103 of the Competition and Consumer Act to the Tribunal when it was performing functions under a State/Territory energy law or a designated Commonwealth energy law.

47    In the administrative law context, the term review “has no settled pre-determined meaning. Rather, it takes its meaning from the context in which it appears”: EAPL at [62] per Gummow and Hayne JJ (Gleeson CJ, Heydon and Crennan JJ agreeing at [13]).

48    The nature of the review by the Tribunal was, as the Tribunal repeatedly (and correctly) described in its reasons, a “limited merits review regime”: see [31], [92], [101], [699]. Specifically, the review conducted by the Tribunal was limited by: (i) the grounds for review upon which an application can be made (s 71C(1)); (ii) the matters that can be raised by a party other than the AER as part of the review (s 71O); and (iii) the material that may be considered by the Tribunal when making its determination in respect of the application for review (s 71R). However, subject to these limitations, the proper role of the Tribunal was to engage in merits review of the AER’s determination in order to determine whether a ground for review listed in s 71C was established.

49    The AER’s submissions, in section B, raised many points about the construction and scope of the grounds of review. It was somewhat difficult to discern how the balance of the submissions translated those observations into grounds of judicial review. For example, the AER made submissions (at AER [16]-[20]) about whether the Full Court was correct in ACCC v ACT to observe that a “finding of fact” could include “an opinion about the existence of a future fact or circumstance”. However, the Tribunal did not rely upon any errors of fact in that extended sense, and the AER did not anywhere suggest that it did.

50    The electricity network respondents submitted there was thus no occasion in the present case for a general review of the scope of the s 71C grounds. For completeness, they responded to the AER’s contentions as follows.

51    As to error(s) of fact – s 71C(1)(a)-(b) the electricity network respondents submitted the phrases “error of fact” and “findings of fact” in s 71C(1)(a) and (b) of the NEL were ordinary English words with a familiar meaning in law in the context of judicial and appellate review. Except to the extent the statutory scheme indicated otherwise, the words should be read consistently with that jurisprudence.

52    First, in relation to the concept of “findings of fact”, it was well accepted that this encompassed far more than ‘primary’ fact finding. Findings of fact may include the application of standards requiring normative judgments. Thus, for example, a finding that particular conduct was negligent is a finding of fact, Woods v Multi-Sport Holdings Pty Limited [2002] HCA 9; 208 CLR 460 at 502-503 [137]-[140] per Hayne J, as was a finding that conduct was misleading or deceptive, Domain Names Australia Pty Ltd v .au Domain Administration Limited [2004] FCAFC 247; 139 FCR 215 at [18] (Full Court), a finding that inadequate provision had been made for an eligible person within a family provision statute, Singer v Berghouse [1994] HCA 40; 181 CLR 201 at 210 and a finding that a contract was ‘unjust’ within the meaning of the Contracts Review Act 1980 (NSW), Perpetual Trustee Co Limited v Khoshaba [2006] NSWCA 41; 14 BPR 26,639; at [40].

53    At the level of general principle, what inferences should be drawn from the evidence as was accepted by the trier of fact, was generally a matter of fact, Brakell v Metropolitan Ice & Cold Storage Works (W Angliss & Co Aust Pty Ltd) (1946) 63 WN (NSW) 203 (Court of Appeal). The distinction was often expressed in terms of whether a question would be left for the jury in a jury trial, Australian Competition and Consumer Commission v Telstra Corp Limited [2004] FCA 987; 208 ALR 459 at [49] (Gyles J).

54    A “finding of fact” in s 71C was at least as broad as its ordinary meaning in conventional usage, such as in conventional court proceedings. Such findings of fact may be simple (‘I find that the car was travelling on the wrong side of the road at the time of the accident’) or more complicated (‘Having regard to the evidence discussed above, including the expert opinions of Professor X and Professor Y and my consideration of those opinions, I find that the prevailing required return on equity for investors in a conventional electricity business in 2015 is 7.1%’). The latter may involve the weighing of different strands of evidence, the consideration of methodologies, and the formation of ‘evaluative judgments’. It was nevertheless a finding of fact. Further, it was plain that an expert opinion may be material relevant to a fact — whether it be the speed of a motor car or the costs likely to be incurred by a regulated entity. The expert evidence was likely to be material which informed the answers to the factual issues before the Tribunal and so informed the decision as to the fact or facts, DBNGP at [325].

55    There was nothing in the statutory language or context which suggested that ‘findings of fact’ should be narrower than its conventional meaning. Indeed, the Full Court in ACCC v ACT concluded (at [171]) that the meaning of ‘fact’ in the equivalent of s 71C was broader than its conventional meaning (historical facts and present facts), and extended to an opinion about the existence of a future fact or circumstance, and an opinion formed by the AER based upon approaches to the assessment of facts or methodologies which it had chosen to apply: decision reversed on appeal in EAPL, but not on that issue; DBNGP at [323]; ElectraNet No 3 at [67]; Envestra at [27]-[29].

56    The AER’s suggestion at [18], that the Full Court’s analysis in ACCC v ACT could be distinguished on the basis that the Court was there considering a provision that was not a true analogue was misconceived (as the ground for review under consideration by the Full Court provided that there had been an “error in the relevant Regulator’s findings of facts”): EAPL at [77]. The AER placed emphasis on a suggestion by an earlier Tribunal in ActewAGL at [32]-[33] that the Full Court’s characterisation of an opinion as within the concept of a “finding of fact” was a “radical meaning” to be given to the word fact. However, later Tribunals had not adopted this observation in ActewAGL: Application by APT Allgas Energy Limited (No 2) [2012] ACompT 5 (Allgas) at [31] citing ACCC v ACT at [171], see [58] below.

57    Secondly, the phrase “error of fact” in s 71C(1)(a)-(b) also had a familiar meaning in law in the context of judicial and appellate review. Any error in the making of the “findings of fact” as described above was an error of fact. That is, where an applicant raised an error of fact ground, the language of s 71C(1)(a) required the Tribunal to assess the impugned factual findings made by the AER to determine if they could properly be described as “erroneous”, by reference to the evidence upon which they were based: Inghams Enterprises Pty Ltd v Sok [2014] NSWCA 217; 87 NSWLR 198 at [22]. Typical errors included errors in the weighing of evidence, in failing to have regard to probative evidence, in misunderstanding the significance of certain evidence, or in making a logical error in the evaluation process: Warren v Coombes [1979] HCA 9; 142 CLR 531 at 551-552. The Tribunal did not also need to be satisfied that the factual finding of the AER was illogical or irrational. Such a restriction was contrary to the structure of s 71C(1): ElectraNet No 3 at [73].

58    There was nothing in the statutory scheme of the NEL which indicated that an impugned finding which was based on ‘evaluative judgment’ could not constitute an ‘error of fact’: cf. AER [21]. As the Full Court said in ACCC v ACT at 74 [171], the words ‘error of fact’ should be interpreted “broadly enough to be meaningful” in relation to the function of the AER under review. To remove all findings based on ‘evaluative judgments’ from the Tribunal’s review of errors of fact would be to substantially curtail the availability of merits review. In this respect, the AER quoted ACCC v ACT out of context. That case involved a different Code relating to gas which specifically set out (in s 8.4) which methodologies were permitted to be adopted in calculating total revenue (see [60]).

59    Section 8.10 of the Code then set out a list of factors that should be considered in establishing a capital base for a pipeline (at [62]). Unsurprisingly, the Full Court stated in obiter at [172] that a finding of fact is not in error because it is based upon the use of one [permitted] methodology in s 8.4 rather than another [permitted methodology]. Likewise, the Court observed that the relative weight to be given to the list of expressly-permitted considerations in s 8.10 was a matter of discretion rather than a finding of fact which could be impugned. As such, the observation of the Full Court was in fact a narrow one which related only to choices among permitted choices. In any event, the Full Court’s observations were not purporting to be an exhaustive description of the ‘error of fact’ ground.

60    At AER [21], the AER relied on the unremarkable proposition made by an earlier Tribunal in DBNGP at [326] that an error of fact was not established merely by pointing to other evidence that would have supported an alternate fact, to contend that more must be established, “such as the absence of evidence supporting the finding made”. Under the ‘limited merits review’ available pursuant to Subdiv 2, a ground based on an error of fact was clearly broader than the ‘no evidence’ ground of judicial review. As already submitted, an error of fact for the purpose of s 71C(1)(a) and (b) was at least as broad as an error of fact in ordinary judicial proceedings.

61    The conclusion that the Tribunal’s review for ‘error of fact’ bore its conventional meaning in s 71C was buttressed by the extrinsic materials which supported the conclusion that the legislative intention in relation to the merits-review provisions in Subdiv 2 was to provide for a new type of review that was broader than judicial review.

62    The electricity network respondents also made submissions on s 71C(1)(c), that the exercise of the AERs discretion was incorrect, having regard to all the circumstances.

63    The AER was mistaken to suggest that the Full Court’s decision in ACCC v ACT at [175] would support constraining the error in exercising a discretion to those available as judicial review grounds. To the contrary, the Full Court accepted that traditional grounds for judicial review, in relation to the exercise of a discretion, were clearly within s 71C(1)(c); but then continued at [175].

64    Thus the Full Court’s decision was not purporting to restrict the scope of s 71C(1)(c), as suggested by the AER, but in fact was acknowledging its breadth. It certainly was not purporting at [174]-[175] of its decision to make an exhaustive statement of the categories of error that might properly be described as an error of discretion.

65    There was no reason to construe the ground to exclude logical error or unexplained discretionary choice because of the availability of s 71C(1)(d). It may well follow that logical error or unexplained discretionary choice also resulted in an unreasonable decision (see ACCC v ACT at 75 [178]) but it was not necessarily so; hence the separate ground, and hence why logical error or unexplained discretionary choice fell within the s 71C(1)(c) ground, as recognised in a series of decisions of the Tribunal: APT Allgas at [49]; Energy Australia at [67]; Envestra at [47]; ATCO Gas at [40].

66    In MCE, Decision – Review of Decision-Making in the Gas and Electricity Regulatory Frameworks, May 2006 (MCE Decision), when commenting on a then proposed combined “exercise of the decision-maker’s discretion was incorrect or was unreasonable” ground (set out at the bottom of p21), the Ministerial Council of Energy (MCE) said:

This proposed merits review ground would instead allow the ACT to scrutinise the discretion exercised by the decision-maker, taking into account a wider range of circumstances than judicial review would allow. The ACT could decide whether, in all the circumstances, it agrees with the reasoning, judgments and choices made by the original decision-maker. If the ACT considers it made an error, or the decision was unreasonable in a less onerous sense than the judicial review Wednesbury sense, then the ACT would have the power to re-exercise the power to the extent necessary to address the error, and to this extent has all the power of the original decision-maker. But rather than being directed towards an examination at large, the merits review ground would allow the ACT to focus on those matters that it considers being of sufficient weight and importance to make out the specified ground of review.

67    This was far removed from judicial review. The MCE went on, at p24, to decide that the single “exercise of the discretion was incorrect or was unreasonable” ground should be separated out into the two grounds present in s 71C(1)(c) and (d), not so as to narrow the ground, but rather so as to avoid a difficulty if the aspect of a decision that was “unreasonable” was not strictly speaking an exercise of discretion in a legal sense. The MCE Decision was otherwise useful in explaining the policy reasons for adopting a process of merits review that was far broader than judicial review.

68    The electricity network respondents also made submissions on s 71C(1)(d) – that the decision was unreasonable. Those parties submitted that it was neither possible nor necessary to give an exhaustive definition of what is an ‘unreasonable’ decision: ActewAGL at [35]. However it was clear that the concept of unreasonableness in s 71C(1)(d) went beyond Wednesbury unreasonableness: Allgas at [51]; EnergyAustralia at [63]; ActewAGL at [35].

69    The following were given as examples of when a decision met the description of ‘unreasonable’ within s 71C(1)(d): a decision that contained logical error or irrationality: Allgas at [51], citing ACCC v ACT at [178]; a decision that contained an element of arbitrariness: Energy Australia at [67]; ActewAGL at [35]; a decision that was not determined by reference to the applicable criteria in the NEL or the NER: EnergyAustralia at [68], cited with approval in Application by United Energy Distribution Pty Limited [2012] ACompT 1 at [49]-[50], Envestra at [52], Allgas at [54] and ATCO Gas at [45]; or a decision where there had been a failure to take into account a matter which was required to be considered or consideration of a matter which was irrelevant: Allgas at [54]; ActewAGL at [35].

70    In relation to the observation at AER [26] that the unreasonableness must be in the “decision itself”, whilst in a very limited number of cases it might be possible to say that the overall decision was ‘unreasonable’ because the result was plainly unreasonable (without more), in the majority of cases it would be necessary to identify the error in logic or the irrational step which had caused the AER’s decision to miscarry, i.e. has caused the AER to make the decision. This was what the parties in the present case did before the Tribunal. They identified the error, and explained how that error led to the overall decision being a substantial departure from the correct decision. The AER’s point did not (and was not said to) go anywhere for the purposes of the present matter.

71    The AER also contended that certain errors could not be ‘unreasonable’ if the AER’s decision was merely open on the evidence. This was far from a complete statement in relation to the ground of ‘unreasonableness’ as that concept was understood, even within the limited constraints of judicial review: Minister for Immigration and Citizenship v Li [2013] HCA 18; 249 CLR 332 at [68]-[74]; see also APT Allgas at [51], [55]; ActewAGL at [35]. That the decision was open, on the evidence, may not be relevant when the Tribunal was examining an unreasonable decision that contained logical error or irrationality, or an element of arbitrariness or was not determined by reference to the applicable criteria in the NEL or the NER or if there had been a failure to take into account relevant considerations or consideration of a matter that is irrelevant.

72    Finally, the electricity network respondents noted that s 71C(1)(d) separately addressed the unreasonableness of the AER’s decision as a ground of review (as distinct from s 71C(1)(c) which referred to “incorrectness” in the AER’s exercise of its discretion). By contrast, the respective statutes at issue in ACCC v ACT, included a ground of review which linked unreasonableness to the exercise of discretion. The difference in the drafting of s 71(1)(c) and (d) of the NEL meant, as the Tribunal observed in EnergyAustralia, that the term ‘unreasonable’ did not just provide a basis for informing the presence of one or more of the established grounds which rendered a decision ‘incorrect’, in the sense of the incorrect exercise of discretion. Rather, it provided a distinct ground of review: Energy Australia at [59]. It also followed that the observations in ACCC v ACT in relation to unreasonableness in the context of an incorrect exercise of discretion were not applicable to s 71C.

73    The electricity network respondents also made submissions on s 71O(2) – the constraint on matters that may be raised.

74    Those parties submitted that this issue was only relevant to the debt topic. It was unnecessary to consider these matters in detail because on any view the relevant matter was raised and maintained by the respondents. This was readily apparent when the relevant issue was understood in its proper context, and not as mischaracterised by the AER.

75    However, the electricity network respondents submitted, the significance of the decision in SPI Electricity for the current wording of s 71O(2), and therefore to the proper construction of that wording, had been misstated by the AER. Prior to recent amendments, s 71O(2) prevented an applicant from raising any matter before the Tribunal “that was not raised in submissions to the AER”. It had been held that this did not require that the argument before the Tribunal be in precisely the same form as the submission to the AER, but rather “[i]t is only if a matter … cannot be identified as broadly arising out of a matter fairly raised before the determinations under review were made that it will not be permitted to be raised in the review”, and this would accommodate a reformulation of the argument provided the argument had been put to the AER: EnergyAustralia at [312(f)]; Application by Epic Energy South Australia Pty Limited [2002] ACompT 4; (2003) ATPR 41-932 at [24]. That authority was not affected by SPI Electricity or the subsequent amendment.

76    It was clear that the addition of “and maintained” was to deal with the specific mischief of an argument being resolved or abandoned during the process before the AER. It was not dealing with the existing authorities as to the specificity with which the issue had to be raised. Nor, contrary to the AER’s submission, was the addition of “and maintained” contrary to the Full Court’s conclusion in SPI Electricity that an issue could be “raised” (and maintained) even if it could have been pursued more forcefully or more compellingly.

77    The electricity network respondents also made submissions on the issue of the materially preferable NEO decision.

78    Those parties submitted that like the section on the s 71C grounds, the AER had advanced very general submissions about the requirements in the NEL pertaining to a “materially preferable NEO decision”. It was unnecessary to determine all of the matters raised by the AER, because the only alleged error by the Tribunal in its findings in respect of this requirement was a specific matter connected with imputation credits, and a general matter raised in connection with the AER’s overall approach to “grounds for review”. It was likewise unnecessary for the respondents to respond to all of the matters raised.

79    The AER referred to the issue of “cherry picking”, and provided an example to illustrate the point. There may be cases where the regulated business sought to correct one error, where there was a matching error or matching generosity elsewhere, such that correcting the error would not result in a materially preferable decision in terms of satisfying the NEO. However, the present case was not such a case. The errors identified by the Tribunal were errors in fundamental aspects of the revenue building blocks, which caused allowable revenue to be materially understated. As the Tribunal recorded at [1221], it was not suggested that any interrelationships or generosities were of such magnitude as to offset the potentially adverse consequences of the established grounds of review.

80    The Tribunal gave very careful consideration to the issue of a materially preferable NEO decision, at [31]-[49], [65]-[101], and [1176]-[1226] (and with other references throughout the decision). Nothing said by the Tribunal was inconsistent with any point made by the AER in its submissions and the AER did not appear to suggest otherwise. With the exception of ground 19 relating to a specific issue concerning the value of imputation credits, the AER’s application did not raise any ground concerning the Tribunal’s finding that its determination would likely result in a “materially preferable NEO decision”.

81    In reply, the AER submitted it did not ask this Court to “resolve the difficult and complex matters of judgment raised by the evidence and resolved by the Tribunal”. To the contrary, the AER challenged the Tribunal’s decision on the basis that, in purporting to resolve such matters, the Tribunal exceeded the scope of its powers under the NEL and NGL. In Pilbara Infrastructure Pty Limited v Australian Competition Tribunal [2012] HCA 36; 246 CLR 379, the High Court reversed the Full Court and set aside a decision of the Tribunal on the basis that it had formed its conclusions following a review which was not of the kind that was provided for under the relevant legislative scheme: at [120]. This Court should reach the same conclusion here. Indeed, there were even greater constraints on the Tribunal’s powers to conduct a limited merits review of regulatory decisions of the AER – which must be based on the particular grounds, the particular material and the particular subject matter specified in the NEL and NGL – than there were upon the Tribunal’s function at issue in Pilbara Infrastructure (at [35]-[37], [60]) to undertake a “re-consideration” of the relevant matter. Further, in addition to the limits imposed on the Tribunal’s powers to review the AER’s decisions by NEL ss 71C, 71O and 71R (and NGL equivalents), the Tribunal’s merits review jurisdiction was limited to one which led to a materially preferable NEO decision.

82    As to error of fact – NEL s 71C(1)(a)-(b); NGL s 246(1)(a)-(b), the AER submitted in reply that the review regime presently in issue conferred discrete powers for the Tribunal to review material errors of fact in findings of fact, unreasonable decisions, and incorrect exercises of discretion. In delineating what was required to establish a ground under each category, the articulation by courts of the meaning of a “fact” in quite different contexts – including ones concerning whether a finding warranted appellate intervention under specific provisions, was appropriately a subject of expert evidence, or was amenable to determination by a jury – could not provide any useful guidance.

83    The AER submitted that the electricity network respondents suggestion that an error of fact for the purpose of s 71C(1)(a) may arise from the formation of evaluative judgments to prefer the opinion of a particular expert, or the making of choices from permitted methodologies, found no basis in the extrinsic material, was inconsistent with authority and should be rejected: ACCC v ACT at [171]-[173]; Envestra at [30]. Such an expansive construction of the notion of an error of fact would give the “unreasonableness” and “incorrect discretion” grounds in s 71C little or no operation; and would enfeeble the objective legislative intent – evinced by the deliberate insertion of grounds of review in the merits review regime in the first place – to limit the class of errors that were susceptible to Tribunal review. Indeed, the untrammelled characterisation of an error of fact, for which the electricity network respondents contended, had been rejected even where extremely broad review powers were engaged.

84    As to s 71C(1)(c) incorrect discretion – the AER submitted that the passage from ACCC v ACT did not detract from the Court’s identification that the respects in which a discretion may be said to be exercised incorrectly mirrors the categories identified in House. If a material error of fact had been established, it may follow that a discretion informed by such a fact had also been exercised incorrectly. However, where no factual error had been established, the discretion could only be said to have been exercised incorrectly if a legal error of the kind articulated in ACCC v ACT at [174] had been made.

85    The extrinsic material relied upon by the electricity network respondents (see [66] above) gave little guidance on how to ascertain whether “the exercise of the original decision maker’s discretion was incorrect, having regard to all the circumstances”, given that the MCE commentary which was there extracted was directed to a different form of language than that which was ultimately included in each of s 71(1)(c) and s 246(1)(c).

86    As to s 71C(1)(d) – that the decision was unreasonable – the AER submitted that it had not sought to confine a decision that was “unreasonable, having regard to all the circumstances” to one which was merely not open to the AER on the evidence. The AER squarely accepted that the concept of unreasonableness extended to a decision which was not justified by reference to its stated reasons, or which was arbitrary and capricious. That said, the AER submitted in reply, as the unreasonableness ground was supplemental to grounds that related to an error of fact or incorrect exercise of discretion, there was no basis in the legislation to construe it as covering the universe of logical errors or deficient reasoning, as the electricity network respondents appeared to contend.

87    As to s 71O(2) – the constraint on matters that may be raised – the AER submitted that the respondents invited this Court to construe the provisions, in their amended form, by applying the authorities which considered what needed to be done to “raise” a matter – within the meaning of the unamended provisions – and then construing the meaning of “and maintained” by reference to the particular facts of the case (SPI Electricity) which ostensibly prompted the relevant amendments. That approach to construction should be eschewed. The provision required that a matter not be agitated before the Tribunal where the AER and not been asked to consider and reach views on it throughout its complex regulatory decision-making process.

88    The intervenors made submissions on these issues and the parties responded to those submissions as follows.

89    The Minister submitted it may be necessary when construing an amended Act to approach with caution earlier interpretations of the Act (and even its unamended provisions) which could not have accounted for the effect of the subsequent amendments. It is the text as amended which is controlling, and not “[p]araphrases of the statutory language … in cases decided under the Act or under different legislation”: Baini v The Queen [2012] HCA 59; 246 CLR 469 at [14]. An earlier version of an Act is, for this purpose, “different legislation”. Such paraphrases are “apt to mislead” and “do not, and cannot, stand in place of the words used in the statute”.

90    The Minister submitted that the relevant context included the South Australian Minister’s Second Reading Speech for the Bill for the 2013 Amending Act (South Australia, Parliamentary Debates, House of Assembly, 26 September 2013, 7171 (The Hon J R Rau) which explained that the amendments implemented aspects of COAG’s Energy Market Reform Implementation Plan. Its “most significant amendments” related to “the role of the [Tribunal] in conducting a review of a reviewable regulatory decision”.

91    The Minister referred to the review undertaken by an “independent expert panel”, being a Panel which delivered “Stage One” and “Stage Two” reports, in September and December 2012 respectively, entitled “Review of the Limited Merits Review Regime” (Panel Reports). Certain recommendations of the Panel came to be reflected in a Statement of Policy Intent by Energy Ministers in December 2012 and, in June 2013, the “Regulation Impact Statement: Limited Merits Review of Decision-Making in the Electricity and Gas Regulatory Frameworks – Decision Paper” (Decision Paper).

92    The Minister submitted that these extrinsic materials made clear that the 2013 amendments were intended (i) to substantially re-orientate the Tribunal’s role; and (ii) to “align” the decision-making objective of the Tribunal with the decision-making objective of the AER. The Minister submitted that the facts that, prior to the 2013 Amending Act, the Tribunal was not entrusted with a conventional merits review role with respect to the AER’s decisions, and the limits on its review function were further, and significantly, increased by the 2013 Amending Act, spoke powerfully in favour of a legislative intent to confer a substantial degree of evaluative latitude on the AER.

93    The Minister submitted that the extrinsic materials disclosed an unmistakeable intention to re-enliven the importance of the long term interests of consumers as the sole criterion for decision-making. The Tribunal’s task was not adequately described in terms of pursuing economic efficiency, and was not directed to the attainment of perfection in decision-making.

94    The Minister submitted that the correction of error or errors in a decision under review would not necessarily lead to a materially preferable decision. The 2013 amendments:

require the [Tribunal] to undertake an holistic assessment of whether the setting aside or varying of the reviewable regulatory decision, or remission of the matter back to the original decision maker, will or is likely to deliver a materially preferable outcome in the long term interests of consumers.

95    The Minister submitted that the legislature had not sought directly to confine judicial review with respect to AER or Tribunal decisions. However, it had clearly sought to significantly (indeed drastically) confine the role of the Tribunal when reviewing the merits of the AER’s decision by (i) expressly confining the grounds of review; (ii) imposing a strict and rigorous regime for obtaining leave to agitate grounds of review; (iii) confining the body of material and considerations to which the Tribunal could have regard; and (iv) strictly confining, in various and different ways, the Tribunal’s power to set aside the AER’s decision, even upon satisfaction of one or more grounds of review.

96    As to the NEO, set out in s 7 of the NEL, the Minister submitted that in the Ausgrid matter, the Tribunal, correctly with respect, said that the “ultimate objective” reflected in the NEO and NGO was “to direct the manner in which the national electricity market and the national natural gas market are regulated, that is, in the long term interests of consumers of electricity and natural gas respectively with respect to the matters specified” (at [77]). The Tribunal continued, however, to say (at [77]):

The provisions proceed on the legislative premise that their long term interests are served through the promotion of efficient investment in, and efficient operation and use of, electricity and natural gas services. This promotion is to be done “for” the long term interests of consumers. It does not involve a balance as between efficient investment, operation and use on the one hand and the long term interests of consumers on the other. Rather, the necessary legislative premise is that the long term interests of consumers will be served by regulation that advances economic efficiency.

97    The Minister submitted that this reasoning, if taken out of context, did not adequately reflect the relevant textual or contextual considerations. But the observations ought not to be read in isolation. See also Application by ATCO Gas Australia Pty Ltd [2016] ACompT 10 at [29] (ATCO Gas). The Tribunal should not be read as having held (or, alternatively, should not have held, if it did) that the long term interests of consumers were to be equated with efficient investment, operation and use.

98    The necessary contextual qualification of the passage in [77] was supported by a principled approach to construction. The statutory phrase “for the long term interests of consumers” was a purposive expression; it qualified the preceding purpose of promoting efficiency; it added meaningful content to the NEO and NGO and must be given real work to do such that it would be quite wrong to interpret the NEO and NGO as though the expression “for the long terms interests of consumers” did not appear (or was replaced with “for the long term interests of regulated entities”): Project Blue Sky [1998] HCA 28, 194 CLR 355 at [70]-[73]. The word “for” in this expression could not be simply equated to “with respect to”. Rather, the Minister submitted, it identified the long term interests of consumers as the sole intended targets (and beneficiaries) of efficiency gains. The ‘choice’ between outcomes must be resolved (wholly and only) by reference to the long term interests of consumers.

99    The Minister submitted it would be quite antithetical to the statutory text and the underlying objective of the NEL and NGL simplistically to conflate the long term interests of consumers with economic efficiency. Economic efficiency per se is not the objective.

100    In relation to the materially preferable NEO or NGO decision, the Minister generally adopted the submissions of the AER in relation to the construction of s 71P of the NEL and s 259 of the NGL. Contrary to the respondents’ submission that “[n]othing said by the Tribunal is inconsistent with any point made by the AER”, the Minister submitted that in those paragraphs, the Tribunal, although correctly construing s 71P and s 259, appeared ultimately to have misstated the test at its point of application in two important respects. The Court should take care not to embrace those errors (the Minister made no submission about whether the result of the Tribunal’s application of the legislation to the facts was correct or not). The errors appeared at Ausgrid [1205] and [1218], where the Tribunal stated that:

[1205] The real question … is whether, by reason of the matters where it has found grounds of review made out, the balancing exercise which the AER carried out is, or may well be, erroneous. If it is, or is likely to be, then there is a very real risk that allowing the [decision under review] to stand will, or is likely to, have … adverse consequences to the long term interests of consumers … (emphasis added)

[1218] [The AER’s] decisions have been reached on complex factual bases and/or the exercise of discretions giving rise to very significant outcomes which, by reason of the Tribunal’s conclusions on the grounds of review, are not appropriate to support the ultimate decision of the AER. (emphasis added)

101    As to [1205], the Minister submitted that the Tribunal correctly held at [101] that, in ss 71P and 259, “will, or is likely to” means “more likely than not”. The reference in [1205] to a standard of “may well be” or “very real risk” was difficult to reconcile with the correct construction reached at [101].

102    As to [1218], the Minister submitted that the Tribunal correctly held at [91]-[92] that the existence of grounds of review and consequences for revenue were not determinative of whether a materially preferable NEO decision existed. That construction could hardly be doubted in light of: (i) the subordination of the Tribunal’s power to vary or set aside a decision to the prerequisite satisfaction that to exercise that power will probably result in a materially preferable NEO or NGO decision (see s 71P(2a) of the NEL and s 259(4a) of the NGL); and (ii) the express language of s 71P(2b)(d) of the NEL and s 259(4b)(d) of the NGL. That [1218] refers only to grounds of review and “significant outcomes” was difficult to reconcile with the correct construction, unless understood as an expression of the Tribunal’s satisfaction that there was more likely than not to exist a materially preferable NEO decision.

103    The Minister submitted that the Tribunal’s powers to vary or set aside a decision were engaged “only” if the Tribunal was satisfied that to do so will probably result in a decision that is materially preferable to the reviewable regulatory decision in making a contribution to the achievement of the NEO or the NGO: s 71P(2a)(c) of the NEL and s 259(4a)(c) of the NGL. This limitation upon the Tribunal’s powers was given specific content by ss 71P(2b) and (2c) of the NEL and ss 259(4b) and (4c) of the NGL. In particular: the Tribunal must “consider the … decision as a whole” in assessing the extent of the contribution to the achievement of the NEO or the NGO; the Tribunal must consider how the “constituent components” of the decision “interrelate with each other and the matters raised as a ground for review” and specify the manner in which it has done so; and certain matters are expressed to be “in themselves” not determinative of the question about whether a materially preferable decision exists.

104    These provisions made clear, for instance, that any error (even a grave error) on the part of the AER in applying the revenue and pricing principles in s 7A of the NEL and s 24 of the NGL, or the Rules, cannot of itself establish the existence of a “materially preferable” NEO or NGO decision. The provisions made equally clear that in determining whether varying or setting aside a decision will probably result in a materially preferable NEO or NGO decision, the Tribunal must have regard not simply to the constituent components of a decision under review but, crucially, to the decision taken “as a whole”.

105    The Minister submitted that the limitations on the Tribunal’s powers were inserted, inter alia, to ensure that the limited merits review regime did not enable distortions in decision-making that were seen to have been occasioned by regulated entities “cherry picking” grounds of review. It was also clear that the limitations were introduced as additional decision-making requirements. The 2013 Amending Act made clear that there were two separate stages, and applicants bore an onus in relation to each of them: first, establishing a ground of review; and secondly, establishing, in accordance with the requirements of s 71P(2b) of the NEL and s 259(4b) of the NGL, that varying or setting aside the AER’s decision would or would be likely to result in a materially preferable decision. It was clear that: (i) the two stages cannot be conflated; (ii) the addition of the second stage by the 2013 Amending Act in no way relaxed the first stage; (iii) the steps set out in s 71P(2b) of the NEL and ss 259(4b) of the NGL must be meaningfully implemented; and (iv) the 2013 Amending Act very substantially “raised the bar” for successful review by applicants in the Tribunal.

106    In particular, the Minister submitted the intensifier, “materially”, must be given real work to do. In this context, “materially preferable” should be construed to mean something in the nature of “substantially” or “considerably” preferable. It directed attention to something of genuine importance or significance. That, in turn, could direct attention to proportionality considerations (noting, for instance, that a difference of $5 million might ordinarily seem large but, in the context of the Ausgrid Determination, would amount to less than one tenth of 1% of the amount the AER determined may be recovered over the regulatory control period).

107    The Minister submitted that when its reasons were read fairly as a whole, the Tribunal could be seen not to have erred materially in the construction of s 71P and s 259, but paragraphs [1205] and [1218], taken by themselves, must be approached with caution.

108    As to compliance with s 71O of the NEL and s 258A of the NGL, the Minister generally adopted the submissions of the AER in relation to the construction of s 71O of the NEL. He submitted that the legislative response to SPI Electricity could not be taken to be limited to the particular facts of the case. What was required was genuine engagement with the regulator on the issues for determination.

109    The electricity network respondents replied to the Minister’s submissions about construction of s 71P of the NEL (and s 259 of the NGL) as follows.

110    Insofar as the Minister’s proposed submissions went beyond the statements made by the Tribunal in Ausgrid at [90], [93] and [94] and contended that the sole criterion for determining a materially preferable NEO (or designated NGO) decision was the long term interests of consumers, independently of the requirement to promote economic efficiency (if they do go that far), the Minister was incorrect.

111    However, a debate about that issue was irrelevant to, and would be likely to distract from, the questions in issue in these proceedings. The AER did not contend that even if the grounds of review in s 71C of the NEL and s 246 of the NGL were made out, the Tribunal erred in concluding that setting aside and remitting the relevant regulatory decisions would, or would be likely to, result in a materially preferable NEO (or designated NGO) decision. The matters raised in the Minister’s submissions about the scope of s 71P of the NEL and s 259 of the NGL did not relate to any of the grounds of review in the AER’s Applications and did not otherwise arise for consideration. Accordingly, the electricity network respondents did not propose to engage with them.

112    The Minister’s submissions also dealt with passages in Ausgrid which (so it was asserted) might be read as misstating the test in s 71P and s 259, and made assertions about the meaning of the word ‘materially’ in the expression ‘materially preferable’. However, the Minister’s submissions ultimately concluded that ‘[w]hen its reasons are read fairly as a whole, the Tribunal can be seen not to have erred in the construction of s. 71P and s. 259’, but that the relevant passages ‘must be approached with caution.’ Thus, the Minister did not appear to be asserting that the Tribunal had erred. In any event, those submissions were irrelevant to the questions in issue in these proceedings. Again, therefore, the respondent parties did not propose to engage with those submissions.

113    As to the Minister’s submissions about s 71O of the NEL and s 258 of the NGL, the electricity network respondents submitted that the Minister did not cite any authority for the proposition that a matter sought to be advanced on review in the Tribunal was required to have been genuinely advanced before the AER with a degree of intensity or specificity sufficient for the AER to be truly apprised of the matter at each stage of the decision-making process. Nor was the Minister’s proposition supported by the text of s 71O or relevant extrinsic material.

114    The AER did not file specific submissions responding to the submissions on behalf of the Minister.

115    PIAC as intervener submitted that in construing the Tribunal’s powers, one must commence from a true understanding of the AER’s powers as the primary decision-maker in the statutory regime: Pilbara Infrastructure (High Court) at 398 [34]; see also EAPL at 245-246 [62], [75]-[80].

116    Prior to the 2013 amendments, the AER’s overarching obligation was cast in loosely aspirational terms, namely, that it must perform its economic regulatory functions in a manner that will, or is likely to, contribute to the achievement of the NEO: NEL, s 16(1)(a).

117    In conjunction with the amendments to the Tribunal’s powers, the Statutes Amendment (National Electricity and Gas Laws - Limited Merits Review) Act 2013 inserted a new provision in each of the NEL and the NGL that was directed to significantly increasing the AER’s accountability for advancing the NEO through its revenue determinations. When making a distribution determination, NEL s 16(1)(d) now required that:

The AER must … if the AER is making a reviewable regulatory decision and there are 2 or more possible reviewable regulatory decisions that will or are likely to contribute to the achievement of the national electricity objective—

(i)    make the decision that the AER is satisfied will or is likely to contribute to the achievement of the national electricity objective to the greatest degree (the preferable reviewable regulatory decision); and

(ii)    specify reasons as to the basis on which the AER is satisfied that the decision is the preferable reviewable regulatory decision. (emphasis added)

118    That is, to the extent that the AER was otherwise afforded any discretion, judgement or choice in making a revenue determination, it was required to make the decision that it was satisfied will promote efficiency, for the long-term interests of consumers, to the greatest degree.

119    The extrinsic material that preceded the 2013 Amending Act confirmed that the increased decision-making obligations on the AER in relation to the NEO were to be matched with the NEO-based limitation on the Tribunal’s powers under s 71P(2a)(c).

120    The origins of s 16(1)(d) may be discerned in SCER’s Statement of Policy Intent. The policy intent of establishing a clear alignment between the AER’s and the Tribunal’s obligations with regard to advancement of the NEO was then developed further in SCER’s Regulation Impact Statement.

121    Section 16(1)(d) of the NEL imposed a substantive obligation that sat atop, and confined, the discretions or judgements afforded to the AER under the NER. Unlike both the previous s 16(1)(a) obligation and the revenue and pricing principles (NEL, ss 7A, 16(2)), s 16(1)(d) it was neither a mere generalised obligation, nor a mere “consideration”. This, in turn, had a consequential effect on the scope of the available grounds of review in the Tribunal, particularly s 71C(1)(c) and (d).

122    In ATCO Gas, decided subsequently to the decisions under consideration in this Court, the Tribunal had correctly recognised that the corresponding provision inserted into the NGL in 2013 had the effect that, where the Rules purported to vest the AER with a discretion or otherwise left two or more possible decisions open to the AER, the AER was required to consider which of those possible decisions will contribute to the promotion of efficiency, for the long-term interests of consumers, to the greatest degree. In that respect, s 16(1)(d) very substantially narrowed discretions under the Rules that the AER would otherwise enjoy at large.

123    PIAC submitted the insertion of s 16(1)(d) had impliedly worked a broadening effect on the grounds of review in s 71C(1)(c) and (d) of the NEL. This was the consequence of the increased decision making obligation imposed by s16(1)(d).

124    The strong normative requirement newly added by s 16(1)(d) intersected most obviously with the “incorrect exercise of discretion” ground of review, stated in s 71C(1)(c) of the NEL. As a merits review tribunal, constituted for its specialist expertise in economics, it was open to the Tribunal to find that an exercise of discretion by the AER was “incorrect” if the Tribunal concluded that the AER’s decision was not likely to contribute to the achievement of the NEO to the greatest degree. In those circumstances, the AER’s exercise of discretion was “incorrect” because, in the Tribunal’s expert assessment, it will not, or is not likely to, promote efficiency, and thereby advance the long-term interests of consumers, optimally or most fully. That was precisely the evaluative function of the Tribunal that the 2013 amendments were designed to strengthen.

125    It followed from the insertion of s 16(1)(d), and the Tribunal’s observations in ATCO Gas, that the “incorrect exercise of discretion” ground of review had, by implied modification, broadened the scope of that ground, so as to enable effective review of the AER’s obligation to advance the long-term interests of consumers to the greatest degree: cf the Tribunal’s rejection of this proposition below, at Ausgrid [105]-[107].

126    The authorities on which the AER relied, in seeking to equate this ground of review to the House principles for appellate review of judicial discretions, all predated the 2013 amendments, and were therefore no longer apposite.

127    The s 71C(1)(d) “unreasonable decision” ground of review may also be seen to have been impliedly modified by the insertion of s 16(1)(d). The Court in Minister for Immigration and Citizenship v Li at [67] directed that the standard of review for reasonableness is inescapably informed by the proper construction of the AER’s determination-making powers:

In Klein v Domus Pty Ltd, Dixon CJ said that where discretions are ill-defined (as commonly they are) it is necessary to look to the scope and purpose of the statute conferring the discretionary power and its real object. The ordinary approach to statutory construction, reiterated in Project Blue Sky Inc v Australian Broadcasting Authority, requires nothing less. The legal standard of reasonableness must be the standard indicated by the true construction of the statute.

128    The insertion of s 16(1)(d) must be understood as having heightened the “legal standard of reasonableness” indicated by the true construction of the NEL. Or, in other words, the substantial narrowing of the AER’s discretions and judgements by s 16(1)(d) had reduced the scope of the class of “reasonable” decisions, and correspondingly broadened the class of “unreasonable” decisions, within the meaning of s 71C(1)(d). The necessary connection with the unreasonable decision ground is underscored by the s 16(1)(d)(ii) requirement that the AER must specify its reasons why its decision was likely to advance the long-term interests of consumers most effectively.

129    In response to PIAC’s submissions, the AER submitted that PIAC’s contention that s 16(1)(d) of the NEL “very substantially narrows discretions under the Rules that the AER would otherwise enjoy at large” was incorrect on a number of levels.

130    First, the AER’s discretions under Ch 6 of the NER were not enjoyed at large. Second, s 16(1)(d) did not “narrow” the AER’s discretions generally. It had a more limited field of operation and could be contrasted with ss 16(1)(a) and 16(2)(a). The requirement in s 16(1)(d) did not apply to each and every discretion that was conferred on the AER under Ch 6 of the NER in the course of making a distribution determination. The obligation applied to the overall distribution determination itself. As to PIAC’s further contention that “it is open to the Tribunal to find that an exercise of discretion by the AER is ‘incorrect’ if the Tribunal concludes that the AER’s decision is not likely to contribute to the achievement of the NEO to the greatest degree”, the AER submitted that the contention was not relevant to the present applications and was also incorrect on a number of levels. In the present matter, the Tribunal did not find that any exercise of discretion by the AER was incorrect for the reason that the AER’s decision was not likely to contribute to the achievement of the NEO to the greatest degree. Accordingly, it was unnecessary for this Court to determine whether PIAC’s contention was correct. Nevertheless, the AER submitted, the contention was wrong for at least two reasons. First, the AER’s exercise of discretion would not be incorrect merely because the Tribunal concluded that the decision was not likely to contribute to the achievement of the NEO to the greatest degree. Second, PIAC’s contention was erroneous because it purported to extend the requirement in s 16(1)(d) to each and every discretion that might be exercised by the AER, whereas the section was confined, relevantly, to the distribution determination.

131    It followed from the foregoing, the AER submitted, that PIAC was wrong to contend that the scope of the “incorrect exercise of discretion” ground of review had been broadened by s 16(1)(d) and that the House principles were no longer apposite. Section 16(1)(d) added a further requirement to be complied with by the AER in making (relevantly) a distribution determination. The section did not alter s 71C or its interpretation. The contention in PIAC [14] in respect of the unreasonableness ground was wrong for the same reason.

132    The electricity network respondents responded to PIAC’s submissions regarding s 16(1)(d) of the NEL as follows. As to the proper construction of the 2013 amendments to the NEL and the alleged importance of s 16(1)(d) of the NEL, the electricity network respondents submitted that PIAC raised the points it now sought to agitate in its submissions before the Tribunal, and the Tribunal dealt with the significance of s 16(1)(d) at [107] and [108]. The Tribunal did not accept the position propounded by PIAC on this topic. This part of the Tribunal’s decision was not the subject of any ground of judicial review. No party sought to impugn this part of the reasoning. Having not itself brought any application for judicial review, PIAC ought not be permitted, under the guise of an application to intervene, to agitate the correctness of parts of the Tribunal’s reasoning that were otherwise not the subject of the present applications, and to seek to have this Court determine legal questions that may be of interest to PIAC but which did not arise on the present applications. The electricity network respondents did not otherwise wish to make written submissions in response to the substantive submissions of PIAC on the inter-relationship between s 16(1)(d) and the scope of the grounds of review in s 71C.

Analysis

133    We begin by restating the observations made by the Full Court in Pilbara Infrastructure at [16], as follows:

It is not this court’s function to resolve the difficult and complex matters of judgment raised by the evidence and resolved by the Tribunal. This court’s role in reviewing the decision of the Tribunal is to ensure that the decision of the Tribunal accords with the law. In Re Minister for Immigration and Multicultural Affairs; Ex parte Applicant S20/2002 (2003) 198 ALR 59 at [114] it was said that it is no part of the supervisory jurisdiction of this court to

… enter upon a consideration of the factual merits of the individual decision. The grounds of judicial review ought not be used as a basis for a complete re-evaluation of the findings of fact, a reconsideration of the merits of the case or a re-litigation of the arguments that have been ventilated, and that failed, before the person designated as the repository of the decision-making power.

134    As will have been seen, the parties’ submissions did not differ widely on the proper construction of the grounds of review in s 71C so far as that construction is relevant to the present applications.

135    It will be recalled that the available grounds in s 71C are as follows:

(a)    the AER made an error of fact in its findings of facts, and that error of fact was material to the making of the decision;

(b)    the AER made more than 1 error of fact in its findings of facts, and that those errors of fact, in combination, were material to the making of the decision;

(c)    the exercise of the AER’s discretion was incorrect, having regard to all the circumstances;

(d)    the AER’s decision was unreasonable, having regard to all the circumstances.

136    These grounds may be compared to the grounds under consideration by the Full Court of the Federal Court in ACCC v ACT and by the High Court in the appeal from that judgment in EAPL. The relevant provisions in s 39(2) of Sch 1 to the Gas Pipelines Access (South Australia) Act 1997 (SA) took the following form:

(1)    An application under this section —

(a)    may be made only on the grounds, to be established by the applicant

(i)    of an error in the relevant Regulator’s findings of facts; or

(ii)    that the exercise of the relevant Regulator’s discretion was incorrect or was unreasonable having regard to all the circumstances; or

(iii)    that the occasion for exercising the discretion did not arise.

137    In was in this context that in EAPL the High Court reversed the decision of the Full Court of the Federal Court in ACCC v ACT. Gummow and Hayne JJ, with whom Gleeson CJ, Heydon and Crennan JJ agreed, said:

[78]    Paragraph (a) of s 39(2) limits to three the grounds upon which it was open to EAPL to challenge before the Tribunal the ACCC determination. The focus of the complaints by EAPL was not upon an alleged error of the ACCC in its findings of fact (sub-para (i)), nor was it said that the occasion for the ACCC to fix the ICB [Initial Capital Base] in its final determination had not arisen (sub-para (iii)).

[79]    Rather, attention was given to sub-para (ii). This distinguishes between, first, an exercise of the “discretion” of the ACCC which is “incorrect” and, secondly, an outcome which “was unreasonable having regard to all the circumstances”. In understanding this distinction assistance is provided by the well known passage in the joint judgment of Dixon, Evatt and McTiernan JJ in House v The King. The first branch of sub-para (ii) should be understood as encompassing the words in House, “[i]f the judge acts upon a wrong principle, if he allows extraneous or irrelevant matters to guide or affect him … if he does not take into account some material consideration …”. The second branch of sub-para (ii) covers the case where failure properly to exercise the discretion may be inferred from the character of the result, again in the words of House, as “unreasonable or plainly unjust”. This is the approach to sub-para (ii) of s 39(2)(a) which was taken by the Tribunal (Cooper J presiding) in Application by Epic Energy South Australia Pty Ltd.

[80]    When seen in this light, the term “unreasonable” provides the basis for inferring the presence of one or more of the well established grounds which render a decision “incorrect” in the sense of the first branch of sub-para (ii). This understanding of the notion of “unreasonableness” as founding an inference (rather than itself providing a ground of review) was developed by Dixon J in Avon Downs Pty Ltd v Federal Commissioner of Taxation. The additional use of the term “unreasonable”, in the sense of being “so unreasonable that no reasonable person could have so exercised the power”, has been developed in the case law over the last sixty years as an independent ground of judicial review and is embodied in the AD(JR) Act (s 5(1)(e), (2)(g)). Some account of that development was given by McHugh and Gummow JJ in Re Minister for Immigration and Multicultural Affairs; Ex parte Applicant S20/2002. The better view is that the limited administrative review system established by s 39 of Sch 1 to the SA Act does not include this judicial review ground by use of the word “unreasonable”. In any event, the decision of the ACCC was not treated by the Tribunal as vitiated simply on Wednesbury unreasonableness grounds.

    (Footnotes omitted.)

138    In our opinion, the administrative review conducted by the Tribunal is therefore to some extent referable to concepts familiar in judicial review. That extent is to be found in paragraphs (c) and (d) of s 71C.

139    It is clear, and was not controversial, that the Tribunal does not start again, in contrast, for example, to the review model under the Administrative Appeals Tribunal Act 1975 (Cth). Similarly, the limited form of administrative (merits) review under s 71C does not authorise the Tribunal to intervene on the basis of mere disagreement with the AER’s decision.

140    It is also clear, in our opinion, that there may be, in a particular case, overlap between the grounds on which an application may be made to the Tribunal to review a decision of the AER and, consequently, the grounds on which the Tribunal may act.

141    As to paragraph (c), we see as applicable what the High Court said at [79] in EAPL that the provision should be understood as encompassing the words in House, “[i]f the judge acts upon a wrong principle, if he allows extraneous or irrelevant matters to guide or affect him … if he does not take into account some material consideration …”. Material error of law causing an incorrect exercise of discretion would be included. Because that provision has been recast from the form it took in EAPL and because there is overlap between the paragraphs, failure properly to exercise the discretion based on an inference from the character of the result as “unreasonable or plainly unjust” is available under paragraph (c) and may also be available under paragraph (d) in circumstances where the incorrect exercise of the discretion has the consequence that the AER’s decision is unreasonable.

142    As to paragraph (d), it has also been recast from the form it took in EAPL and is no longer limited to the exercise of the regulator’s discretion. It now refers to the AER’s decision. As so recast, in our opinion the dictum of the High Court (that its previous form “covers the case where failure properly to exercise the discretion may be inferred from the character of the result, again in the words of House, as “unreasonable or plainly unjust”) is no longer applicable to paragraph (d): the provision is no longer limited to the exercise of the AER’s discretion.

143    Having said that, in our opinion paragraph (d), that the AER’s decision was unreasonable, having regard to all the circumstances, is not limited to Wednesbury unreasonableness and indeed, as made clear by Minister for Immigration and Citizenship v Li at [68], Wednesbury unreasonableness is not the starting point for the standard of reasonableness nor should it be considered the end point.

144    As to the word “discretion”, we consider that in context this term does not extend to a choice by the AER between facts or opinions, which it has authority to decide, but primarily refers to express or implied statutory choices, where those assessments call for value judgments in respect of which there is room for reasonable differences of opinion, no particular opinion being uniquely right. The contrast is with an order the making of which is dictated by the application of a fixed rule to the facts on which its operation depends: Norbis v Norbis [1986] HCA 17; 161 CLR 513 at [4]. See also ACCC v ACT at [173].

145    In relation to paragraphs (a) and (b) of s 71C, the AER made one or more material errors of fact in its findings of facts, in our opinion, there is some initial attraction in the AER`s submission that the Full Courts third category in [171] of its judgment in ACCC v ACT goes too far. The Full Court said in relation to the grounds of review as then framed, at [171]-[172]:

The first ground is error in the Relevant Regulator's finding of facts. Given the limitations imposed by s 39(5) this is a ground which can only be made out by reference to the materials which were before the ACCC. 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 ACCCs 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 the judgment of McInerney J in Morley v National Insurance Co [1967] VR 566 at 567, a question arose about what constituted a “fact” for the purposes of s 55 of the Evidence Act 1958 (Vic) which makes admissible certain documents containing statements by deceased persons tending to establish any fact. McInerney J said:

“It may be that the phrase ‘a fact’ must be given an expanded meaning equivalent to some matter in issue to be established in the proceedings. Indeed, Wigmore, in his work on Evidence, 3rd ed., vol 1, p. 1, para. 1, points out that, in one sense, ‘everything in the cosmos is a fact or a phenomenon’.

In my view, it would be contrary to the policy of the legislation to give a restricted meaning to the word ‘fact’ so as to exclude a statement of opinion by an expert.”

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.

It is important to note however that the Full Court stated that findings of fact may include a finding which was an opinion about the existence of a future fact or circumstance (our emphasis). The Full Court gave as an example assessment by the primary decision-maker of expert opinion and that the term “findings of fact” should be interpreted so as to encompass opinions formed by the primary decision-maker based upon approaches to the assessment of facts or methodologies which it had chosen to apply. The Full Court then went on to say, at [172], that the choice between permitted methodologies for the calculation of Total Revenue mentioned in s 8.4 of the then Code was not a finding of fact. Nor was a finding of fact in error because it was based upon the use of one methodology rather than another. In our opinion, an error in making a finding of fact would not always extend to an opinion about the existence of a future fact or circumstance but, for example, would include an error as to what the opinion of an expert was even where that opinion was about the existence of a future fact or circumstance.

146    As we have set out above, the AER submitted that when the Tribunal considered whether there was an “error of fact”, the concept of “error” set a threshold that was not passed simply because there was material that could support a different finding of fact or simply because the Tribunal might, if it considered the material afresh, prefer to make a different finding of fact: DBNGP at [326] and WA Gas Networks at [22]. Thus far we agree.

147    The AER went on to submit that before a finding of fact could be characterised as erroneous, more must be established by a review applicant: such as the absence of evidence supporting the finding made. In our opinion, although it is correct to say that more must be established than that the Tribunal prefers to make a different finding of fact, it is not to be thought that “no evidence” is any more than an example of a finding of fact which is erroneous. It was in that context that Pfizer Pty Ltd v Birkett [1999] FCA 1778 at [11], relied on by the AER, was decided in relation to the admissibility of expert opinion evidence. We do not see that decision as being of any wider assistance in the construction of s 71C of the NEL (or s 246 of the NGL).

148    For example, we accept the submission on behalf of the electricity network respondents that the concept of ‘findings of fact’ encompasses more than ‘primary’ fact finding. In our opinion, findings of facts within s 71C would generally include inferences drawn from primary facts or conclusions drawn from primary facts. We would also observe that we would not regard ‘evaluative judgment’ as an appropriate discrimen for the purposes of s 71C. In our opinion, depending on the particular circumstances, ‘findings of fact’ may include such a finding even where it is based on an ‘evaluative judgment’.

149    Considering the four available grounds in s 71C overall, and stating again that mere disagreement with the AER, whether as to facts or exercises of discretion or overall, is not a basis on which the Tribunal may intervene, it would in our opinion be strange as a matter of statutory construction if there were classes of material error on the part of the AER which the Tribunal discerned but which were beyond the Tribunal’s reach and resulted in areas of decision-making immune from (limited) review. To conclude that there were some findings of fact or exercises of discretion which were not susceptible to review by the Tribunal for material error seems to us to be unlikely as a matter of the intention of the legislature. In our view the touchstone for s 71C material error is that the Tribunal may not act where there is no error and error is not made out by choosing one available fact or opinion over others. For example, the mere availability of a different inference from facts would not establish error within s 71C: the Tribunal must find that there was error in the AER’s decision-making before it may intervene.

150    It is not, in our opinion, possible or desirable to attempt to bring any more precision to the concept of an error or errors of fact in the AER’s findings of fact. The parties will assert or deny that such an error or errors was or were made and it will then be necessary for the Tribunal to evaluate the nature of that claimed error or those claimed errors. In doing so, in our opinion, the Tribunal should recognise the permissible field of choice which the legislature has given to the AER.

151    As to the raised and maintained” question in s 71O of the NEL in relation to a matter, in our opinion whether such a matter has been raised must be addressed as an issue of substance. So also whether such a matter has been maintained must be addressed as an issue of substance. The primary meaning we would give to “maintain” is whether the matter has been continued or gone on with. We do not accept the submission on behalf of the AER that the correct approach is to ask whether a service provider has engaged with the regulator or that service providers must engage with the regulator (and other interested parties) throughout the primary decision-making process. In our opinion, that is to raise a distracting or potentially distracting test. Similarly, we do not accept the submission on behalf of the Minister that what was required was genuine engagement with the regulator on the issues for determination and that “parties must be incentivised to engage fully in the regulatory process at first instance”.

152    As to the “materially preferable NEO decision”, we accept the AER’s submission that an implicit statutory premise is that a decision rectifying errors revealed by established grounds of review may not lead to a decision that is “materially preferable” to the decision made by the AER since, if the correction of errors revealed by established grounds of review necessarily resulted in a materially preferable NEO decision, then s 71P(2a) would have no work to do.

153    We construe the word “materially” in this context to mean not only that the error must be relevant or pertinent but that the future decision would be or would be likely to be preferable to an important degree or extent.

154    In our opinion the words “would, or would be likely to, result in” a materially preferable decision do not need lengthy explanation. The first permutation of the words, “would… result in” involves a conclusion on the part of the Tribunal at a high level of certainty, while the second permutation of the words, “would be likely to… result in” involves a conclusion on the part of the Tribunal at a lower level of certainty but requiring that the conclusion be “more likely than not”.

155    We see the first permutation as more applicable where the result of the Tribunal’s review is certain because the Tribunal has specifically varied the regulatory decision and the second permutation as more applicable where the Tribunal has set aside the regulatory decision and remit the matter back to the AER.

156    The identified mischief to which the amending legislation was directed was “cherry picking”. We accept the submission on behalf of the Minister that whether there will be, or would be likely to be, a materially preferable decision to the decision in fact made by the regulator depends on an assessment of the decision as a whole, and a comparison of that decision with a putative alternative decision, either by the Tribunal or by the AER on remitter: it does not depend simply on an assessment of errors in individual components of the decision under review.

157    At a general level, we see no reviewable error on the part of the Tribunal in its consideration of this issue of a materially preferable NEO decision, at [31]-[49], [65]-[101], and [1176]-[1226]. Each of those paragraphs must be read in context, as expressly recognised in the submissions on behalf of the Minister.

158    In relation to PIAC’s submission that s 16(1)(d) of the NEL is of present significance, we do not accept that submission. The submission was that s 16(1)(d) imposed a substantive obligation that sat atop, and confined, the discretions or judgements afforded to the AER under the NER and that this, in turn, had a consequential effect on the scope of the available grounds of review in the Tribunal, particularly s 71C(1)(c) and (d).

159    In our opinion, that provision requires, where there are 2 or more possible reviewable regulatory decisions that will or are likely to contribute to the achievement of the national electricity objective, the AER must make the decision that it is satisfied will or is likely to contribute to the achievement of that objective to the greatest degree. In our opinion, in so providing, the legislature is making express, perhaps unnecessarily, the content of what is preferable. The provision then goes on to require the AER to specify reasons as to the basis on which it is satisfied that the decision is the preferable reviewable regulatory decision.

Forecast Operating Expenditure

Introduction

160    The AER’s case is that the Tribunal’s decisions and orders with respect to the forecast operating expenditure (forecast opex) of each of the electricity network respondents involved one or more of the grounds of reviewable error identified in s 5 of the ADJR Act.

161    In the case of each electricity network respondent, the Tribunal set aside the AER’s decision and made directions. The Tribunal was satisfied that there was reviewable error within s 71C(1) of the NEL and it exercised the power to remit the reviewable regulatory decision in s 71P(2)(c) of the NEL. We have set out above our views as to the proper construction of the grounds for review in s 71C(1) of the NEL (at [133]-[159]).

162    At the forefront of the AER’s submissions in relation to forecast opex is a submission that the Tribunal misconstrued or misapplied the grounds for review in s 71C(1). In other words, the Tribunal found error in the AER’s approach in circumstances in which there was either no error, or if there was an error, it was not an error of the type identified in s 71C(1). For example, the Tribunal found error in circumstances in which in fact it did no more than disagree with the AER’s decision on a particular matter or the weight it placed on a particular matter. The other grounds of review advanced by the AER included a challenge to particular conclusions of the Tribunal relating to aspects of a model used by the AER, a challenge to what was said to be the Tribunal’s interpretation of a phrase in the NEL and NER (“a regulatory obligation or requirement”) and the Tribunal’s reliance on a denial of procedural fairness as a ground of review when a denial of procedural fairness is not a ground for review within s 71C. Finally, the AER submitted that the Tribunal’s orders and directions were legally uncertain and unreasonable and did not comply with the requirements of the NEL.

163    This section of our reasons is structured in the following way. First, we will set out those provisions of the NER which are relevant to the AER’s task in making a decision on forecast opex. Secondly, we will summarise the AER’s grounds of review in their respective ADJR Act applications. Thirdly, we will examine the Tribunal’s reasons in detail. It is necessary that we do that because of the way in which the parties put their respective submissions. As we have said, Ausgrid is the lead decision and that will be the focus of our reasons, although we will also consider the Tribunal’s decisions in Endeavour, Essential and ActewAGL. Finally, we will analyse and determine the AER’s ADJR Act grounds of review. For reasons we will give, we have decided that none of the grounds of review are made out.

The National Electricity Rules

164    The rules which are relevant to forecast opex are in Chapter 6 of the NER.

165    Rule 6.4.3 provides that the annual revenue requirement for a DNSP for each regulatory year of a regulatory control period must be determined using a building block approach under which the building blocks consist of a number of components, including forecast opex for that year. The forecast opex for the year is the forecast opex as accepted or substituted by the AER in accordance with r 6.5.6.

166    Before we come to r 6.5.6, we set out r 6.12.1 which provides that a distribution determination is predicated on a number of decisions by the AER (constituent decisions) and which includes (relevantly for forecast opex) the following:

(1)    a decision in which the AER either:

(i)    acting in accordance with clause 6.5.6(c), accepts the total of the forecast operating expenditure for the regulatory control period that is included in the current building block proposal; or

(ii)    acting in accordance with clause 6.5.6(d), does not accept the total of the forecast operating expenditure for the regulatory control period that is included in the current building block proposal, in which case the AER must set out its reasons for that decision and an estimate of the total of the Distribution Network Service Provider’s required operating expenditure for the regulatory control period that the AER is satisfied reasonably reflects the operating expenditure criteria, taking into account the operating expenditure factors;

167    Rule 6.5.6(a) provides that a building block proposal must include the total forecast opex for the relevant regulatory control period which the DNSP considers is required in order to achieve four objectives which are described as the operating expenditure objectives and which are as follows:

(1)    meet or manage the expected demand for standard control services over that period;

(2)    comply with all applicable regulatory obligations or requirements associated with the provision of standard control services;

(3)    to the extent that there is no applicable regulatory obligation or requirement in relation to:

(i)    the quality, reliability or security of supply of standard control services; or

(ii)    the reliability or security of the distribution system through the supply of standard control services,

to the relevant extent:

(iii)    maintain the quality, reliability and security of supply of standard control services; and

(iv)    maintain the reliability and security of the distribution system through the supply of standard control services.

(4)    maintain the safety of the distribution system through the supply of standard control services.

168    Rule 6.5.6(b) provides that the forecast of required operating expenditure of a DNSP that is included in a building block proposal must meet certain requirements which it is not necessary for us to set out.

169    Rule 6.5.6(c) provides that the AER must accept the forecast of required operating expenditure of a DNSP that is included in a building block proposal if the AER is satisfied that the total of the forecast operating expenditure for the regulatory control period reasonably reflects each of three matters which are described as operating expenditure criteria and which are as follows:

(1)    the efficient costs of achieving the operating expenditure objectives; and

(2)    the costs that a prudent operator would require to achieve the operating expenditure objectives; and

(3)    a realistic expectation of the demand forecast and cost inputs required to achieve the operating expenditure objectives.

170    Rule 6.5.6(d) provides that if the AER is not satisfied as referred to in (c), it must not accept the forecast of required operating expenditure of a DNSP that is included in a building block proposal.

171    Rule 6.5.6(e) requires the AER in deciding whether it is satisfied as referred to in paragraph (c), to have regard to a number of matters which are described as operating expenditure factors and which are as follows:

(1)    [Deleted]

(2)    [Deleted]

(3)    [Deleted]

(4)    the most recent annual benchmarking report that has been published under rule 6.27 and the benchmark operating expenditure that would be incurred by an efficient Distribution Network Service Provider over the relevant regulatory control period;

(5)    the actual and expected operating expenditure of the Distribution Network Service Provider during any preceding regulatory control periods;

(5A)    the extent to which the operating expenditure forecast includes expenditure to address the concerns of electricity consumers as identified by the Distribution Network Service Provider in the course of its engagement with electricity consumers;

(6)    the relative prices of operating and capital inputs;

(7)    the substitution possibilities between operating and capital expenditure;

(8)    whether the operating expenditure forecast is consistent with any incentive scheme or schemes that apply to the Distribution Network Service Provider under clauses 6.5.8 or 6.6.2 to 6.6.4;

(9)    the extent the operating expenditure forecast is referable to arrangements with a person other than the Distribution Network Service Provider that, in the opinion of the AER, do not reflect arm’s length terms;

(9A)    whether the operating expenditure forecast includes an amount relating to a project that should more appropriately be included as a contingent project under clause 6.6A.1(b);

(10)    the extent the Distribution Network Service Provider has considered, and made provision for, efficient and prudent non-network alternatives; and

(11)    [Note: Clause left intentionally blank]

(12)    any other factor the AER considers relevant and which the AER has notified the Distribution Network Service Provider in writing, prior to the submission of its revised regulatory proposal under clause 6.10.3, is an operating expenditure factor.

The Grounds for Review

172    In the case of Ausgrid, the AER identifies six grounds for review in relation to the forecast opex decision in its Originating Application for Judicial Review (Grounds 6 – 11 inclusive) and a general ground that applies to forecast opex as well as other components of the AER’s decisions (Ground 21). Those grounds, among others, form the basis of its claim that the Tribunal’s decision should be set aside ab initio and that the decision be referred back to the Tribunal for further consideration, subject to such directions as the Court thinks fit. The same grounds are reflected in the AER’s challenge to the Tribunal’s decisions in relation to forecast opex in Endeavour, Essential and ActewAGL.

173    First, the AER claims that, in deciding to set aside the AER’s decision on the basis of the AER’s conclusions with respect to the allowance for forecast opex, the Tribunal misconstrued and misapplied s 71C of the NEL and, therefore, misunderstood the scope and limitations of its role. This is Ground 6 and it is also a claim made in Ground 21. The essence of the AER’s claim is that the Tribunal found that in performing its function under the NEL and the NER, the AER placed too much weight on a model known as the EI Model (defined below at [191]) and that this criticism of the AER by the Tribunal was not a ground of review within s 71C(1) of the NEL. Furthermore, in considering various matters which had formed part of the AER’s approach and reasoning, the Tribunal said that it had a “series of concerns” about the AER’s conclusions without addressing and finding that its concerns amounted to one or more grounds for review within s 71C(1) of the NEL. The AER identified the matters about which the Tribunal said it had “concerns” as relating to the following matters: the use of Australian data; the use of overseas data; the selection of a benchmark comparison point; the application and weight to be given to Operating Environment Factors (OEFs); and the assessment of vegetation management costs. The AER submitted that these errors by the Tribunal are reviewable errors within s 5 of the ADJR Act and, in particular, s 5(1)(c) (lack of jurisdiction), (d) (decision not authorised), (f) (error of law) and (j) (decision otherwise contrary to law).

174    Secondly, the AER claims that the Tribunal erred because it failed to consider or address significant aspects of the reasons for the AER’s decision in relation to opex. This is Ground 7. These errors by the Tribunal are said to have occurred in connection with the AER’s approach to the following matters: the use of Australian data; the OEFs; and the efficiency target/benchmark comparison point. The AER submitted that the Tribunal’s errors were reviewable errors within s 5(1)(c), (d), (f) and (j) of the ADJR Act.

175    Thirdly, the AER claims that the Tribunal misconstrued a phrase in the operating expenditure objective identified in r 6.5.6(a)(2) of the NER, being the phrase “applicable regulatory obligation or requirements associated with the provision of standard control services”. This is Ground 8. The meaning of “regulatory obligation or requirement” is set out in s 2D of the NEL. That section relevantly provides as follows:

2D    Meaning of regulatory obligation or requirement

(1)    A regulatory obligation or requirement is—

(b)    an obligation or requirement under—

(i)    this Law or Rules; or

(ia)    the National Energy Retail Law or the National Energy Retail Rules; or

(ii)    an Act of a participating jurisdiction, or any instrument made or issued under or for the purposes of that Act, that levies or imposes a tax or other levy that is payable by a regulated network service provider; or

(iii)    an Act of a participating jurisdiction, or any instrument made or issued under or for the purposes of that Act, that regulates the use of land in a participating jurisdiction by a regulated network service provider; or

(iv)    an Act of a participating jurisdiction or any instrument made or issued under or for the purposes of that Act that relates to the protection of the environment; or

(v)    an Act of a participating jurisdiction, or any instrument made or issued under or for the purposes of that Act (other than national electricity legislation or an Act of a participating jurisdiction or an Act or instrument referred to in subparagraphs (ii) to (iv)), that materially affects the provision, by a regulated network service provider, of electricity network services that are the subject of a distribution determination or transmission determination.

The relevant paragraph is (v) and the meaning of a participating jurisdiction is set out is s 5.

176    The essence of the AER’s challenge is that the Tribunal erroneously decided that labour costs incurred by the NSW service providers under their Enterprise Bargaining Agreements (EBAs) fell within the definition of a regulatory obligation or requirement. The AER submitted that the Tribunal’s errors were reviewable errors within s 5(1)(c), (d), (f) and (j) of the ADJR Act.

177    Fourthly, the AER claims that the Tribunal erred in reaching its decision because it took into account an impermissible consideration, namely, an alleged failure by the AER to give Ausgrid an adequate opportunity to make submissions with respect to overseas data. This is Ground 9. The AER submitted (correctly) that a denial of procedural fairness is not a ground of review within s 71C(1) of the NEL. The AER submitted that this error was a reviewable error within s 5(1)(c) (lack of jurisdiction), (d) (decision not authorised) and (e) (improper exercise of power coupled with s 5(2)(a), (f) and (j)) of the ADJR Act.

178    Fifthly, the AER claims that the Tribunal erred in that the result of its determination is uncertain. This is Ground 10. It is convenient at this point to set out again the orders made by the Tribunal in Ausgrid with respect to opex:

1.    Pursuant to s 71P(2)(c) of the National Electricity Law, the Final Decision Ausgrid distribution determination 2015-16 to 2018-19, April 2015, including attachments, (the Final Decision) is set aside and remitted to the Australian Energy Regulator (AER) to make the decision again in accordance with the following directions:

(a)    the AER is to make the constituent decision on opex under r 6.12.1(4) of the National Electricity Rules in accordance with these reasons for decision including assessing whether the forecast opex proposed by the applicant reasonably reflects each of the operating expenditure criteria in r 6.5.6(c) of the National Electricity Rules including using a broader range of modelling, and benchmarking against Australian businesses, and including a “bottom up” review of ActewAGL’s forecast operating expenditure;

(d)    the AER is to consider, and to the extent to which it considers appropriate to vary the Final Decision in such other respects as the Australian Energy Regulator considers appropriate having regard to s 16(1)(d) of the National Electricity Law in the light of such variations as are made to the Final Decision by reason of (a)-(c) hereof.

179    The Tribunal made the same orders, mutatis mutandis, in Essential and Endeavour. In ActewAGL, there is an additional paragraph dealing with the service target performance incentive scheme which follows from the decision dealing with forecast opex in the sense that the Tribunal found (and there appears to be no challenge to this) that the relationship between the two was such that setting aside the decision of the AER in relation to forecast opex meant that the decision with respect to the service target performance incentive scheme should also be set aside. The orders made by the Tribunal in ActewAGL are as follows:

1.    Pursuant to s 71P(2)(c) of the National Electricity Law, the Final Decision ActewAGL distribution determination 2015-16 to 2018-19, April 2015, including attachments (the Final Decision) is set aside and remitted to the Australian Energy Regulator (AER) to make the decision again in accordance with the following directions:

(a)    the AER is to make the constituent decision on opex under r 6.12.1(4) of the National Electricity Rules in accordance with these reasons for decision including assessing whether the forecast opex proposed by the applicant reasonably reflects each of the operating expenditure criteria in r 6.5.6(c) of the National Electricity Rules including using a broader range of modelling, and benchmarking against Australian businesses, and including a “bottom up” review of ActewAGL’s forecast operating expenditure;

(b)    the AER is to make the constituent decision on the service target performance incentive scheme in the light of such variations as are made to the Final Decision by reason of (a) hereof;

(c)    the AER is to make the constituent decision on return on debt in relation to the introduction of the trailing average approach in accordance with these reasons for decision;

The Tribunal’s decision to set aside the AER’s decisions and remit them to the AER was made under s 71P(2)(c) (which must be read with ss 71P(2a) and (2b) which are set out above (at [15]).

180    Broadly expressed at this stage, the AER contends that there is uncertainty in the orders when read together with the reasons for decision with respect to the following matters: the meaning of a “bottom up” review and a “broader range of modelling” and, by reference to the Tribunal’s reasons, the lack of direction from the Tribunal as to the following matters: the use of Australian data; the use of overseas data; the appropriate benchmark comparison point; the use and quantification of OEFs; the approach to the electricity network respondents vegetation management costs; and the approach to Ausgrid’s labour costs under the EBAs. The AER submitted that the Tribunal’s orders and directions did not comply with s 71P(2)(c) in that it was not in a position “to make the decision again” in accordance with the directions. That meant (so the AER submitted) that the Tribunal did not have jurisdiction to make the decision and this was a reviewable error within s 5(1)(c) of the ADJR Act. The AER also relied on s 5(1)(d) and (e) ((e) coupled with s 5(2) (f), (h) and (j)).

181    Finally, the AER claims that the Tribunal’s decision was so unreasonable that no reasonable person could have made it. This is Ground 11 and it overlaps with Ground 10. The essence of the challenge set out in this ground is that the Tribunal’s conclusions and orders ignored the fact that the AER had conducted a limited bottom-up review and had applied other benchmarking techniques. The AER also claimed that the Tribunal did not suggest any alternative modelling which the AER ought to employ. The AER submitted that this error was a reviewable error within s 5(1)(e) of the ADJR Act (coupled with s 5(2)(g)).

The Tribunal’s Reasons

182    As we have said, it is necessary for us to examine the Tribunal’s reasons in relation to forecast opex in detail. The Tribunal’s reasons in relation to forecast opex occupy some 112 pages of its decision in Ausgrid. By way of overview, the following may be noted.

183    A central matter to the Tribunal’s decision was its consideration of the use by the AER of an econometric model described by the Tribunal as the EI Model. The nature of this model is described below. The Tribunal found that no grounds of review within s 71C(1) of the NEL were made out in relation to the AER’s decision not to accept the forecast of required operating expenditure of each of the electricity network respondents (r 6.5.6(c) and (d), and r 6.4.3(b)(7)). The Tribunal found that grounds of review within s 71C(1) were made out in relation to the AER’s estimate of the total of the electricity network respondents’ required operating expenditure (r 6.12.1(4)(i) and r 6.4.3(b)(7)). The electricity network respondents made various criticisms of the data used in the model and of adjustments made to the results produced by the model. After outlining the AER’s approach, the Tribunal dealt with the criticisms before addressing the AER’s use of the model. Other than dealing with some general matters, we will follow a similar structure in examining the Tribunal’s reasons.

The Amendments to Rule 6.5.6 in 2012

184    Rule 6.5.6 of the NER was amended in 2012 as part of amendments to the Rules made in that year. Each party sought to support its case before the Tribunal by reference to those amendments.

185    The Tribunal summarised the effect of the 2012 amendments to r 6.5.6 as follows:

(1)    Rule 6.5.6(c)(2) was changed so that instead of referring to the costs that a prudent operator in the circumstances of the relevant DNSP would require to achieve the operating expenditure objectives, it refers to the costs of a prudent operator;

(2)    Rule 6.5.6(e)(2) was changed so that in addition to referring to a benchmark opex that would be incurred by an efficient DNSP, it also refers to the most recent annual benchmarking report published by the AER under r 6.27;

(3)    Rule 6.5.6(e)(7) was inserted and that rule refers to the substitution possibilities between opex and capital expenditure;

(4)    Rule 6.5.6(e)(12) was inserted and that rule refers to any other factor the AER considers relevant and which it has notified the DNSP in writing prior to the DNSP submitting its revised regulatory proposal;

(5)    Rule 6.12.13(f) was deleted. That rule restricted the AER’s discretion in developing a substitute estimate under r 6.12.4(ii) by providing, in effect, that if the AER refused to approve an estimate, the substitute estimate must be determined on the basis of the DNSP’s regulatory proposal and amended from that basis only to the extent necessary to enable it to be approved in accordance with the NER; and

(6)    Rule 6.2.8(a)(1) was amended and r 6.4.5 was introduced requiring the AER to develop and publish Expenditure Forecast Assessment (EFA) Guidelines specifying the approach that the AER proposes to use to assess a DNSP’s opex and capex forecasts.

186    Later in its reasons, at [473]-[480], the Tribunal made reference to the support for their arguments each party sought to obtain from the amendments made to the NER in 2012 and the extrinsic material within the NER Sch 2, s 8 of the NEL. In particular, reference was made to the Australian Electricity Market Commission (AEMC) Rule Determination issued on 29 November 2012. However, the Tribunal did not find it necessary to consider in detail the extrinsic material or the parties’ submissions on the proper interpretation of r 6.5.6.

187    The parties made submissions to this Court as to the effect of the amendments to r 6.5.6 made in 2012 in light of the AEMC Rule Determination. We will make reference to the AEMC Rule Determination later in these reasons and to some aspects of the parties’ submissions. However, like the Tribunal, we do not think it necessary to examine in detail all of the submissions made by the parties (the AER’s submissions (AS at [221]-[225]; the respondents’ submissions (RS) at [170]-[175]).

The Econometric Models

188    The Tribunal described an econometric model as a model designed to estimate a relationship between opex and output or explanatory variables, such as those used in the EI Model (defined below), being customer numbers, circuit length, maximum ratcheted demand and proportion of underground circuits, and to use the variation in costs not explained by the output or explanatory variables to derive an estimate of inefficiency for each DNSP which has been benchmarked. An econometric model will produce only an estimate of the relationship between opex and the specified output variables.

189    The Tribunal said that there are a number of reasons why an econometric model’s estimate of a DNSP’s inefficiency may differ from its actual inefficiency and it described those reasons as including the following:

(1)    the data which is used is not accurate, for example, the data may involve a measurement error or not be comparable;

(2)    the sample which is used may be too small to produce accurate estimates;

(3)    the variables included in the model may be either inappropriate or incomplete in the sense that they do not reflect all of the relevant drivers of opex; or

(4)    the assumptions used in the econometric model are inappropriate.

190    The AER engaged Economic Insights Pty Ltd (Economic Insights) to assist it with the application of economic benchmarking and to advise it on whether the AER should make adjustments to base opex for the NSW service providers (Ausgrid, Endeavour and Essential) and ActewAGL based on the results from economic benchmarking models and the productivity change to be applied to forecast opex for these electricity network respondents. Economic Insights prepared two reports. The First EI Report was dated 17 November 2014, and the Second EI Report was dated April 2015. The model which is referred to by the Tribunal as the EI Model is identified in these two reports.

191    In fact, Economic Insights developed three econometric benchmarking models which were described by the Tribunal in the following way:

(1)    A Cobb Douglas stochastic frontier analysis (CD SFA) opex cost function model. The AER described this model as its preferred econometric model and it is the model which the Tribunal and the parties referred to as the EI Model;

(2)    A Cobb Douglas least squares econometric (LSE): an econometric opex cost function using the Cobb Douglas functional form; and

(3)    A translog LSE: an econometric opex cost function using the translog functional form.

192    Economic Insights also developed two Productivity Index Number (PIN) benchmarking models which the Tribunal described as follows (at [155]):

(a)    a multilateral total factor productivity (MTFP) model which assesses the productivity of all inputs, opex and capital, relative to total output; and

(b)    a multilateral partial productivity factor (MPFP) model which assesses the productivity of opex as an input relative to total output.

The Principal Issue

193    The Tribunal described the principal issue before it as being whether the AER’s application of the EI Model discharged the AER’s obligations under rr 6.5.6 and 6.12.1(4) of the NER. It referred, at [124], to the description of the principal issue by senior counsel for the AER at the hearing. We set this out now. It will become important for reasons we will give.

It’s said that the AER has not complied with the regulatory rules; and if made out, that would involve an incorrect exercise of discretion, and it’s also said principally that the EI model that was used by the AER, both in assessing the DNSPs proposed opex and also in estimating required opex was flawed. And that argument is properly characterised as either the AER incorrectly exercising its discretion or making an unreasonable decision. Broadly, it will be necessary for the Tribunal to determine whether the AER’s approach is compliant with the regulatory framework and whether the decision had a reasonable basis.

… … …

…all econometric models are an approximation, they’re a simplification of the real world. They can never reflect absolutely all the on-the-ground features of the real world and that must be recognised. That being recognised, a regulator will take steps, or ought to take steps, acting reasonably, to make allowances for what’s not revealed by the model, and so the question then becomes as what the AER has done in its estimation, has it taken sufficient reasonable steps to make those allowances.

An Overview of the Electricity Network Respondents’ Challenges to the AER’s Estimate of the Required Opex

194    The Tribunal provided an overview of the challenges by the electricity network respondents to the AER’s estimate of the required opex.

195    The NSW service providers submitted that the estimates were too low. They submitted that the negative impact on each of them was as follows: Ausgrid $731m, Endeavour $264m, and Essential $737m. The Tribunal noted that the parties did not specify whether the figures were nominal or real figures. In broad terms, the NSW service providers submitted that the AER erred because it ignored the best source of information for the required opex which was the opex actually incurred by a DNSP, in favour of an unsound and untested econometric model to estimate opex for each of them by reference to other businesses against which they were benchmarked. Furthermore, they submitted that the issue was not whether the EI Model is better than other models, but whether (to use the Tribunal’s words at [127]):

having regard to the data limitations and other matters, any of the models are fit to be given 100 percent weight in assessing an appropriate level of opex.

196    ActewAGL submitted that the AER’s estimate of the required opex was too low. It contended that the AER’s decision not to accept ActewAGL’s forecast opex resulted in a $130.6m ($2013-14) reduction in the forecast. ActewAGL submitted that the methodology adopted by AER was inconsistent with the methodology prescribed by rr 6.5.6(c) and 6.12.1(4) of the NER and, therefore, it was contrary to law. It submitted that the benchmarking adopted by the AER had serious technical deficiencies which meant that it had no value as a means of assessing ActewAGL’s efficient costs. Finally, ActewAGL submitted that the AER erred in that its approach had regard only to exogenous considerations and not (in addition) to endogenous considerations.

197    PIAC also challenged the AER’s estimate of opex, but it did so by claiming that the estimate was too high by $365m for Ausgrid, $196m for Endeavour and $291m for Essential. It did not complain about the AER’s benchmarking methodology, but submitted that it had made adjustments to the results in favour of the electricity network respondents which were “arbitrary and illogical”.

The AER’s Approach

198    The Tribunal then described the methodology adopted by Economic Insights. A brief summary of the Tribunal’s description will suffice.

199    First, the AER obtained relevant information from Australian DNSPs by the use of regulatory information notices (RINs) (see NEL, Part 3, Division 4). Economic Insights used this information and information derived from DNSPs in New Zealand and Ontario in the EI Model. Secondly, Economic Insights applied country dummy variables for New Zealand and Ontario to pick up differences in opex coverage as well as systematic differences in OEFs, such as the harsher winter conditions in Ontario. Thirdly, the Tribunal described the Economic Insights’ outputs specification criteria, the MTFP and MPFP outputs and the EI Model’s specification.

200    The Tribunal then turned its attention to the Second EI Report. Economic Insights said that there was a case for revising the opex efficiency target and the Tribunal noted, at [172], that in this context, Economic Insights said that “this is the first time economic benchmarking is being used as the primary basis for an Australian regulatory decision”.

201    Following the EI Second Report, the AER revised its estimate by lowering the comparison point for determining the AER’s alternative estimate of base opex in the EI Model from Citipower to AusNet and made further adjustments to operating environment factors. This resulted in a more favourable allowance for the electricity network respondents, but not to the extent they desired. PIAC submitted that the revisions were overly generous.

202    The Tribunal then summarised the AER’s approach to OEFs and individually immaterial OEFs in the case of the NSW service providers and ActewAGL. The following four tables set out the AER’s approach with respect to OEF adjustments and individually immaterial OEF adjustments. The first two tables relate to the NSW service providers, and the second two tables relate to ActewAGL.

NSW Service Providers Summary of final decision on OEF adjustments

Factor

Ausgrid

Endeavour

Essential

Reasons against OEF criteria551

Subtransmission

5.2%

4.9%

3.1%

The boundary between distribution and transmission is not determined by service providers

Data from Ausgrid's regulatory accounts suggest that subtransmission assets are up to twice as costly to operate as distribution assets.

Economic Insights' SFA model does not include a variable that accounts for subtransmission assets.

Licence

conditions

1.2%

0.7%

1.2%

The network planning requirements in the NSW service providers licence conditions are not determined by service providers.

Category analysis and economic benchmarking RIN data suggest that the increased transformer capacity to meet the 2005 and 2007 change in licence conditions may lead to a material increase in maintenance expenditure.

Economic Insights' SFA model does not include a variable that accounts for changes in licence conditions.

OH&S regulations

0.5%

0.5%

0.5%

OH&S regulations are not set by service providers.

Data from the ABS and a PwC report commissioned by the Victorian Government suggest that differences in OH&S regulations may materially affect service provider's opex.

Economic Insights' SFA model does not include a variable that accounts for differences in OH&S legislation.

Termite Exposure

0.0%

0.2%

0.6%

The prevalence of termites in a geographic area is beyond service providers’ control.

Data on Powercor’s termite management costs and data from the CSIRO on the range of termites suggest that the Essential Energy may have a material cost disadvantage due to termite exposure.

Economic Insights’ SFA model does not include a variable that accounts for differences in termite exposure.

Immaterial factors

4.7%

6.7%

5.4%

There are various exogenous, individually immaterial factors not accounted for in Economic Insights' SFA model that may affect service providers' costs relative to the comparison firms. While individually these costs may not lead to material differences in opex, collectively they may.

Total

11.7%

12.9%

10.7%

The AER said in its footnote to that Table that its OEF criteria exogeneity, materiality, and duplication, are explained in detail in its section on its approach to OEFs.

NSW Service Providers Summary of individually immaterial OEF adjustment

Factor

Ausgrid

Endeavour

Essential

Asset lives

0.5%

0.5%

-0.5%

Building regulations

0.5%

0.5%

0.5%

Bushfires

-0.5%

-0.5%

-0.5%

Capitalisation Practices

-0.5%

0.5%

-0.5%

Corrosive environments

0.5%

0.5%

0.5%

Cultural heritage obligations

0.5%

0.5%

0.5%

Environmental Regulations

0.5%

0.5%

0.5%

Environmental variability

-0.5%

-0.5%

0.5%

Extreme weather events

0.5%

0.5%

0.5%

Grounding conditions

0.5%

0.5%

0.5%

Network access

-0.1%

0.5%

0.4%

Planning regulations

0.5%

0.5%

0.5%

Proportion of 11kV and 12kV lines

0.5%

0.5%

0.5%

Rainfall and humidity

0.5%

0.5%

0.5%

Specialised skills

0.5%

0.5%

0.5%

Solar uptake

-0.5%

-0.5%

-0.5%

Topography

0.5%

0.5%

0.5%

Traffic management

0.5%

0.5%

0.5%

Transformer capacity owned by customers

-0.2%

0.1%

0.0%

Division of vegetation management responsibility

0.5%

0.5%

0.5%

Total

4.7%

6.7%

5.4%

Source:    AER analysis

The totals do not reconcile entirely due to rounding

ActewAGL Summary of final decision on OEF adjustments

Factor

Adjustment

Reasons against OEF criteria

Capitalisation Practices

8.5%

Although capitalisation practices are the result of management decisions, differences in capitalisation practices can lead to material differences that are unrelated to efficiency.

ActewAGL's capitalisation practices, with regard to vehicle and IT costs, provide it with a material cost disadvantage relative to the comparison firms.

Economic Insights' SFA model does not include variables that account differences in capitalisation practices between the NEM service providers.

Backyard reticulation

5.6%

Backyard reticulation has been required by ACT planning approaches.

ActewAGL has provided evidence that backyard reticulation materially increases its costs.

Economic Insights' SFA model does not include variables that account for backyard reticulation between the NEM service providers.

Standard control services connections

4.0%

The AER determines service providers' service classifications.

Standard control services connections opex accounts for a material amount of ActewAGL's standard control services opex.

Economic Insights' SFA uses network services data. Connection services are not included in network services.

OH&S regulations

0.5%

OH&S regulations are not set by service providers.

Data from the ABS and a PwC report commissioned by the Victorian Government suggest that differences in OH&S regulations may materially affect service provider's opex.

Economic Insights' SFA model does not include a variable that accounts for differences in OH&S legislation.

Individually immaterial factors

4.4%

There are various exogenous, individually immaterial factors not accounted for in Economic Insights' SFA model that may affect service providers' costs relative to the comparison firms. While individually these costs may not lead to material differences in opex, collectively they may.

Total

23.0%

Source:    AER analysis

ActewAGL Summary of individually immaterial OEF adjustment

Factor

Adjustment

Asset lives

-0.5%

Bushfires

0.5%

Building regulations

0.5%

Corrosive environments

0.5%

Cultural heritage obligations

0.5%

Environmental regulations

0.5%

Environmental variability

-0.5%

Extreme weather events

-0.5%

Grounding conditions

0.5%

Humidity and rainfall

0.5%

Network access

-0.1%

Planning regulations

0.5%

Proportion of 11kV and 12kV lines

0.5%

Solar uptake

-0.5%

Specialised skills

0.5%

Termites

0.0%

Traffic management

0.5%

Transformer capacity owned by customer

0.1%

Topography

0.5%

Underground services

0.4%

Total

4.4%

Source:    AER analysis

203    The Tribunal set out, at [198], the AER’s five step approach to forming its alternative estimate of opex in its Ausgrid Final Decision. It is as follows:

Step 1 –Start with service provider’s opex

We typically use the service provider’s actual opex in a single year as the starting point for our assessment. We call this the base year. While categories of opex can vary year to year, total opex is relatively recurrent. We typically choose a recent year for our assessment.

Step 2 – Assess base year opex

We assess whether opex the service provider incurred in the base year reasonably reflects the opex criteria. We have a number of techniques including economic benchmarking by which we can test the efficiency of opex in the base year.

Step 3 – Add a rate of change to base opex

As the opex of an efficient service provider tends to change over time due to price changes, output and productivity we trend our estimate of base opex forward over the regulatory control period to take account of these changes. We refer to this as the rate of change.

Step 4 – Add or subtract any step changes

We then adjust base year expenditure to account for any forecast cost changes over the regulatory control period that would meet the opex criteria that are not otherwise captured in base opex or rate of change. This may be due to new regulatory obligations in the forecast period and efficient capex/opex trade-offs. We call these step changes.

Step 5 – Other opex

Finally we add any additional opex components which have not been forecast using this approach. For instance, we forecast debt raising costs based on the costs incurred by a benchmark efficient service provider.

Having established our estimate of total forecast opex we can compare our alternative opex forecast with the service provider’s total forecast opex. If we are not satisfied there is an adequate explanation for the difference between our opex forecast and the service provider's opex forecast, we will use our opex forecast.

204    The Tribunal then set out, at [199], the AER’s main techniques for testing the efficiency of Ausgrid’s base opex. The techniques are set out in Table 7.3 of the AER’s Ausgrid Final Decision which is as follows:

Technique

Description of technique

Findings

Economic

benchmarking

Economic benchmarking measures the efficiency of a service provider in the use of its inputs to produce outputs.

The economic benchmarking techniques we used to test Ausgrid's efficiency included Multilateral Total Factor Productivity, Multilateral Partial Factor Productivity and opex cost function modelling. We compared Ausgrid's efficiency to other service providers in the NEM.

Despite differences in the techniques we used, all benchmarking techniques show Ausgrid does not perform as efficiently as most other service providers in the NEM.

We consider that differences in Ausgrid's operating environment not captured in the benchmarking models do not adequately explain the different benchmarking results between Ausgrid and other service providers.

Review of

labour and

workforce

practices

Labour costs represent a large proportion of all NSW service providers' opex. We engaged Deloitte Access Economics (Deloitte) to review the NSW service providers' labour and workforce practices.

Deloitte found that because of labour and workforce management issues, Ausgrid's base year would not likely represent efficient costs.

Deloitte concludes that:

the NSW service providers have high labour costs because they have too many employees. They all engaged permanent staff in preference to contractors over the 2009–14 period for transitory capex work. Now, due to EBA restrictions on redundancies, they have stranded labour

because the NSW service providers employ a high proportion of their employees through EBAs (more than 75 percent) restrictive EBA clauses have a significant impact on workforce flexibility

the optimum level of outsourcing is likely to be higher than the level the NSW service providers outsourced at over the 2009–14 period; this is a key distinguishing factor from the Victorian service providers

while the NSW service provider have been implementing efficiency improvements, many efficiencies have not been realised until after the 2012–13 base year.

205    Subject to the following, a similar table appeared in the AER’s Final Decisions for Endeavour, Essential and ActewAGL respectively:

(1)    Other than a change in the name of the DNSP, the “Economic benchmarking” paragraphs in each table are the same.

(2)    In the case of Endeavour, the “Review of labour and workforce practices” paragraphs are the same with the following addition:

Deloitte considered that Endeavour Energy’s base year opex was likely more efficient than Ausgrid’s and Essential Energy’s because it had commenced implementing efficiency improvements earlier. However, all NSW service providers (including Endeavour Energy) had efficiencies they were yet to realise because the reforms they had implemented to date did not consider potential opportunities to improve efficiency outside of the three NSW businesses. That is, they compared efficiency among themselves, but not to businesses in other jurisdictions.

(3)    In the case of Essential, the “Review of labour and workforce practices” paragraphs are the same with the following addition:

Further, in response to submissions in its revised proposal about the adverse impact of the dispersed nature of its network on labour costs, Deloitte found that Essential Energy could potentially achieve significant cost savings by implementing a local service agent (LSA) model. Powercor achieved significant efficiencies from implementing could potentially achieve significant cost savings by implementing a local service an LSA model following privatisation.

(4)    In the case of Essential, Table 7.3 had an additional entry relating to “Vegetation management” which is as follows:

Technique

Description of technique

Findings

Vegetation

management

Essential Energy's vegetation management costs have increased significantly over the 2009–14 period. Category analysis showed Essential Energy has very high costs compared to most of its peers and Essential Energy’s regulatory proposal included a step down in vegetation management for the forecast period acknowledging that its 2009–14 practices required reform. Therefore, we decided to review Essential Energy’ vegetation management practices in detail.

Our overall findings for vegetation management remain the same as those from our draft decision. That is, Essential Energy's own documentation, including a report it commissioned from Select Solutions, provide evidence that its vegetation management practices in the base year (2012–13) were inefficient.41

Select Solutions' review found that Essential Energy must move to a "significantly more efficient" vegetation management model to reduce the impact of its expenditure on customer prices.42 Select Solutions found several causes of inefficiency, including:

attributing too much vegetation management effort to reactive spot clearing rather than proactive cyclic maintenance

primarily engaging contractors for cutting on a demonstrably less efficient hourly rate basis

less than optimal outsourcing.

We discuss our vegetation management findings in more detail in Appendix A.5.

(5)    In the case of ActewAGL, Table 7.3 contains the following paragraphs with respect to “Review of labour and workplace practices” and “Review of vegetation management”:

Technique

Description of technique

Findings

Review of

labour and

workforce

practices

Labour costs represent a large proportion of ActewAGL’s opex (approximately 80 percent). Category analysis showed ActewAGL had high labour costs relative to most of its peers and ActewAGL’s regulatory proposal suggested labour costs were a reason ActewAGL overspent its opex allowance in 2012–13.

Therefore, we decided, with the assistance of EMCa, to conduct a detailed review of ActewAGL’s labour and workforce practices.

EMCa considered that there is evidence that ActewAGL’s work practices, processes and systems in 2012–13 were ineffective. EMCa considered that this lead to inefficient use of labour in the office and field. This inefficiency is characterised by duplication of effort in work planning and scheduling, loss of field productivity through ineffective works management and through ineffective data and information management.

EMCa also considered that ActewAGL’s labour levels were not reasonably efficient in 2012–13, noting that ActewAGL has steadily increased its ASL based on assumed future growth scenarios and adopting an internal resourcing strategy.

EMCa considered that if ActewAGL had outsourced more of its work, it would likely have benefited from increased labour flexibility and reduced operating costs.

EMCa found a lack of compelling evidence to demonstrate that ActewAGL’s labour costs in 2012–13 were reflective of an efficient service provider. EMCa consider this was evident by the relatively high level of internal resources used and the extent to which work was outsourced on an hourly rate bases for the urgent clearance of vegetation..

Review of

vegetation

management

ActewAGL's vegetation management costs have increased significantly over the 2009–14 period. Category analysis showed ActewAGL has very high costs compared to most of its peers and ActewAGL's regulatory proposal suggested vegetation management was a reason ActewAGL overspent its opex allowance in 2012–13. Therefore, we decided, with the assistance of EMCa, to review ActewAGL's vegetation management practices in detail.

EMCa found that ActewAGL did not act prudently and efficiently to manage costs associated with increased vegetation growth that occurred prior to 2012–13 because its vegetation management practices and its strategic and tactical responses were inadequate.

EMCa also found evidence of inefficient vegetation management costs in 2012–13 due to the manual processes between the office and field and the extent of clearance work that was deemed to be urgent, and which was therefore undertaken with a resultant higher cost. It is EMCa’s view that a service provider acting to efficiently minimise costs would have incurred a lower level of urgent clearance work.

206    The Tribunal concluded that the AER had used the EI Model to adjust the base opex of each of Ausgrid, Essential and ActewAGL to determine a starting point for a forecast that it considered would reasonably reflect the criteria and that the AER considered that it could not use the historical opex of these electricity network respondents to arrive at an alternative forecast because such a forecast would not result in a forecast that would reasonably reflect the opex criteria. However, in the case of a forecast based on Endeavour’s actual opex in 2012-13, at [205] the AER was not satisfied that it was materially inefficient.

207    In the case of Endeavour, the Tribunal said that the AER found that its forecast opex involved two additional items which it characterised as “step changes”, being additional expenditure not incurred by Endeavour during the base year. This additional expenditure was $240.7m ($2013-14) in respect of increased vegetation management costs and $17.3m ($2013-14) in redundancy costs. The Tribunal found that the AER rejected this additional expenditure relying on the EI Model.

208    The Tribunal, at [211], then set out the steps which the AER said in its Final Decision it took to arrive at Ausgrid’s base opex. Those steps are set out in Table 7.4 of Attachment 7 to the AER’s Ausgrid Final Decision and are as follows:

Description

Output

Calculation

Step 1 – Start with Ausgrid's average opex over the 2006 to 2013 period.

Ausgrid's network services opex was, on average, $509.3 million ($2013) over the 2006 to 2013 period.

$509.3 million

($2013)

Step 2 - Calculate the raw efficiency scores using our preferred economic benchmarking model

Our preferred economic benchmarking model is Economic Insights’ Cobb Douglas SFA model. We use it to determine all service providers' raw efficiency scores.

Based on Ausgrid's customer numbers, line length, and ratcheted maximum demand over the 2006 to 2013 period, Ausgrid's raw efficiency score is 44.7 percent.

44.7 percent

Step 3 – Choose the comparison point

For the purposes of determining our alternative estimate of base opex, we did not base our estimate on the efficient opex estimated by the model.

The comparison point we used was the lowest performing service provider in the top quartile of possible scores, AusNet Services. According to this model AusNet Services' opex is 76.8 percent efficient based on its performance over the 2006 to 2013 period. Therefore to determine our substitute base we have assumed a prudent and efficient Ausgrid would be operating at an equivalent level of efficiency to AusNet Services.

76.8 percent

Step 3 – Adjust Ausgrid's raw efficiency score for operating environment factors

The economic benchmarking model does not capture all operating environment factors likely to affect opex incurred by a prudent and efficient Ausgrid.

We have estimated the effect of these factors and made a further reduction to our estimate where required. We have determined an 11.7 percent reduction to Ausgrid's comparison point based on our assessment of these factors.

A material operating environment factor we considered was not accounted for in the model is the different subtransmission configurations in NSW.

68.7 percent

= 0.768 /

(1 + 0.117)

Step 4 – Calculate the percentage reduction in opex

We then calculate the opex reduction by comparing Ausgrid's efficiency score with the adjusted comparison point score.

35.0 percent

= 1 – (0.447 /

0.687)

Step 5 – Calculate the midpoint efficient opex

We estimate efficient opex at the midpoint of the 2006 to 2013 period by applying the percentage reduction in opex to Ausgrid's average opex over the period.

This represents our estimate of efficient opex at the midpoint of the 2006 to 2013 period.

330.9 million

($2013)

= (1 – 0.350)*

509.3 million

Step 6 – Trend midpoint efficient opex forward to 2012–13

Our forecasting approach is to use a 2012–13 base year. We have trended the midpoint efficient opex forward to a 2012–13 base year based on Economic Insights’ opex partial factor productivity growth model. It estimates the growth in efficient opex based on growth in customer numbers, line length, ratcheted maximum demand and share of undergrounding.

It estimated the growth in efficient opex based on Ausgrid’s growth in these inputs in this period to be 8.48 percent.

359.0 million

($2013)

= 330.9 x (1 =

0.0848)

Step 7 – Adjust our estimate of 2012–13 base year opex for CPI

The output in step 6 is in real 2013 dollars. We need to convert it to real 2013–14 dollars for the purposes of forming our substitute estimate of base opex. This reflects one and a half years of inflation. This is our estimate of base opex.

374.2 million

($2013-14)

= 359.0 x (1 +

0.042)

There is a mistake in the numbering in this Table in that there are two steps identified as 3. We will adjust the numbering for the purposes of the observations which follow.

209    The following matters may be seen from this Table. In step 1, the AER started with Ausgrid’s average opex over the period from 2006 to 2013, namely, $509.3m. In step 2, the AER used the EI Model to calculate a raw efficiency score of 44.7 percent. In step 3, the AER chose a comparison point of 76.8 percent, based on AusNet which is at the bottom of the top quartile of possible scores using the EI Model. In step 4, there is an adjustment to the raw efficiency score for the OEFs, that adjustment being made to the comparison point on the assumption that the comparator (AusNet) would have to contend with the same environment factors as Ausgrid.

210    The adjustment made by the AER once the model had been run lowers the comparison point to 68.7 percent. The mathematical consequence of this adjustment is that the efficiency score of Ausgrid becomes 44.7 percent and, therefore, in step 5, it is calculated that Ausgrid is, in effect, 35 percent below the comparison point. In step 6, the AER calculated a mid-point efficient opex, that is, the average efficient opex over the 2006-13 period based on the EI Model. In step 7, it adjusted that mid-point forwards to 2012-13 following that adjustment in step 8 with an adjustment of its estimate of Ausgrid’s 2012-13 base year opex for consumer price index (CPI) to arrive at a figure of $374.2m ($2013-14).

211    The Tribunal then summarised the AER’s approach to the same matters in the case of Essential and ActewAGL. It is not necessary for us to set out those details.

212    The Tribunal then turned to consider how the AER had applied the operating expenditure factors in r 6.5.6(e) in making the decision required by r 6.5.6 (d).

213    The Tribunal said, at [225], that Table 7.7 in Attachment 7 to the Ausgrid Final Decision provided a convenient summary of the AER’s application of the benchmarking opex factor in r 6.5.6(e)(4) and the opex factors in rr 6.5.6(e)(5), (5A), (6), (7), (8), (9A) and (10). The Table is as follows:

Opex factor

Consideration

The most recent annual benchmarking report that has been published under rule 6.27 and the benchmark operating expenditure that would be incurred by an efficient Distribution Network Service Provider over the relevant regulatory control period.

There are two elements to this factor. First, we must have regard to the most recent annual benchmarking report. Second, we must have regard to the benchmark operating expenditure that would be incurred by an efficient distribution network service provider over the period. The annual benchmarking report is intended to provide an annual snapshot of the relative efficiency of each service provider.

The second element, that is, the benchmark operating expenditure that would be incurred an efficient provider during the forecast period, necessarily provides a different focus. This is because this second element requires us to construct the benchmark opex that would be incurred by a hypothetically efficient provider for that particular network over the relevant period.

We have used several assessment techniques that enable us to estimate the benchmark opex that an efficient service provider would require over the forecast period. These techniques include economic benchmarking, opex cost function modelling, category analysis and a detailed review of Ausgrid's labour and workforce practices. We have used our judgment based on the results from all of these techniques to holistically form a view on the efficiency of Ausgrid's proposed total forecast opex compared to the benchmark efficient opex that would be incurred over the relevant regulatory control period.

The actual and expected operating expenditure of the distribution network service provider during any proceeding regulatory control periods.

Our forecasting approach uses the service provider's actual opex as the starting point. We have compared several years of Ausgrid's actual past opex with that of other service providers to form a view about whether or not its revealed expenditure is sufficiently efficient to rely on it as the basis for forecasting required opex in the forthcoming period.

The extent to which the operating expenditure forecast includes expenditure to address the concerns of electricity consumers as identified by the distribution network service provider in the course of its engagement with electricity consumers.

We understand the intention of this particular factor is to require us to have regard to the extent to which service providers have engaged with consumers in preparing their regulatory proposals, such that they factor in the needs of consumers.103 We have considered the concerns of electricity consumers as identified by Ausgrid in assessing its proposal – particularly those expressed in the consumer-focussed overview provided as an attachment to its regulatory proposal. For example, a clear theme present in this document is that customers consider electricity prices are too high.104

The relative prices of capital and operating inputs

We have considered capex/opex trade-offs in considering step changes for Ausgrid's head office building and for demand management expenditure. We considered the relative expense of capex and opex solutions in considering these step changes.

We have had regard to multilateral total factor productivity benchmarking when deciding whether or not forecast opex reflects the opex criteria. Our multilateral total factor productivity analysis considers the overall efficiency of networks in the use of both capital and operating inputs with respect to the prices of capital and operating inputs.

The substitution possibilities between operating and capital expenditure.

As noted above we considered capex/opex trade-offs in considering step changes for Ausgrid's head office building and for demand management expenditure. We considered the substitution possibilities in considering these step changes.

Some of our assessment techniques examine opex in isolation – either at the total level or by category. Other techniques consider service providers' overall efficiency, including their capital efficiency. We have relied on several metrics when assessing efficiency to ensure we appropriately capture capex and opex substitutability.

In developing our benchmarking models we have had regard to the relationship between capital, opex and outputs.

We also had regard to multilateral total factor productivity benchmarking when deciding whether or not forecast opex reflects the opex criteria. Our multilateral total factor productivity analysis considers the overall efficiency of networks with in the use of both capital and operating inputs.

Further, we considered the different capitalisation policies of the service providers and how this may affect opex performance under benchmarking.

Whether the operating expenditure forecast is consistent with any incentive scheme or schemes that apply to the distribution network service provider under clauses 6.5.8 or 6.6.2 to 6.6.4.

The incentive scheme that applied to Ausgrid's opex in the 2009–14 regulatory control period, the EBSS, was intended to work in conjunction with a revealed cost forecasting approach.

In this instance, we have forecast efficient opex based on benchmark efficient service provider. We have considered this in deciding how the EBSS should apply to Ausgrid in the 2009–14 regulatory control period and the 2014–19 period.

The extent the operating expenditure forecast is referable to arrangements with a person other than the distribution network service provider that, in our opinion, do not reflect arm's length terms.

Some of our techniques assess the total expenditure efficiency of service providers and some assess the total opex efficiency.

Given this, we are not necessarily concerned whether arrangements do or do not reflect arm's length terms. A service provider which uses related party providers could be efficient or it could be inefficient. Likewise, for a service provider who does not use related party providers. If a service provider is inefficient, we adjust their total forecast opex proposal, regardless of their arrangements with related providers.

Whether the operating expenditure forecast includes an amount relating to a project that should more appropriately be included as a contingent project under clause 6.6A.1(b).

This factor is only relevant in the context of assessing proposed step changes (which may be explicit projects or programs). We did not identify any contingent projects in reaching our final decision.

The extent the distribution network service provider has considered, and made provision for, efficient and prudent non-network alternatives.

We have not found this factor to be significant in reaching our final decision.

214    Finally, at [226], the Tribunal reproduced Table 7.8 which sets out the other factors the AER had regard to, being factors it considered relevant and pursuant to r 6.5.6(e)(12), notified each DNSP of prior to the DNSP submitting its revised regulatory proposal. The Table is as follows:

Opex factor

Consideration

Our benchmarking data sets, including, but not necessarily limited to:

1. data contained in any economic benchmarking RIN, category analysis RIN, reset RIN or annual reporting RIN

2. any relevant data from international sources

3. data sets that support econometric modelling and other assessment techniques consistent with the approach set out in the Guideline

as updated from time to time.

This information may potentially fall within opex factor (4). However, for absolute clarity, we are using data we gather from NEM service providers, and data from service providers in other countries to provide insight into the benchmark operating expenditure that would be incurred by an efficient and prudent distribution network service provider over the relevant regulatory period.

Economic benchmarking techniques for assessing benchmark efficient expenditure including stochastic frontier analysis and regressions utilising functional forms such as Cobb Douglas and Translog.

This information may potentially fall within opex factor (4). For clarity, and consistent with our approach to assessment set out in the Guideline, we are have regard to a range of assessment techniques to provide insight into the benchmark operating expenditure that an efficient and prudent service provider would incur over the relevant regulatory control period.

215    Having summarised the AER’s approach to the determination of opex, the Tribunal turned to consider the principal issue before it which it said was whether the AER’s application of the EI Model discharged its obligations under rr 6.5.6 and 6.12.1(4) of the NER. Those obligations were that the AER either accepted the forecast opex in the building block proposal or, acting in accordance with r 6.5.6(d), did not accept it in which case the AER was required to provide reasons for its decision and an estimate of forecast opex that the AER was satisfied reasonably reflected the operating expenditure criteria, taking into account the operating expenditure factors. As we have said in this case, the AER did not accept the forecast opex in the building block proposal and arrived at its own estimate of forecast opex. The NSW service providers and ActewAGL submitted that the AER made a number of errors in its approach. The Tribunal then turned to consider each of those alleged errors.

Criticisms of the data used in the EI Model

216    The data used in the EI Model consisted of the RIN data and overseas data.

217    With respect to the RIN data, we do not think it is necessary for us to set out the Tribunal’s summary of the competing submissions and the expert and other evidence. A reference to the key points and conclusions will be sufficient.

218    The NSW service providers submitted that the AER failed to collect data pursuant to its RINs which was sufficiently rigorous for use in benchmarking. In particular, they submitted that the RINs were unclear and the requests for eight years of data resulted in different DNSPs recording data differently and some DNSPs estimating and backcasting data. The act of backcasting data was described by the NSW service providers as involving the DNSPs creating estimates based on a set of assumptions for data points related to the past where actual results were not classified or categorised in the way the RIN required the information. The Tribunal accepted that description of backcasting.

219    The Tribunal, at [259], concluded that the AER’s reliance on the RIN data pointed to a weakness in its benchmarking and, in reaching that view, it relied on opinions of consultants engaged by the electricity network respondents. The Tribunal said at [260] that “in this point of its evolution” the RIN data was not data upon which the AER might rely on to the extent that it did in its application of r 6.5.6(e)(4) to determine the benchmark opex that would be incurred by an efficient DNSP or to corroborate the AER’s use of the EI Model.

220    With respect to overseas data, the data consisted of data relating to service providers in New Zealand and Ontario. In preparing the EI Model, the AER augmented the Australian RIN data with this overseas data.

221    The Tribunal, at [262], found that Economic Insights chose to incorporate the overseas data in the EI Model because the data was available and in a similar form to the data in the RIN database and not because the DNSPs in New Zealand and Ontario were comparable to the Australian DNSPs. There were significant disparities between the size of the Australian DNSPs and the overseas DNSPs. The Tribunal, at [265], also concluded that Economic Insights choice of output variables in the EI Model, being ratcheted maximum demand, customer numbers and circuit length have been “largely” determined “in a somewhat circular manner” by the data that was available between all three countries. The Tribunal drew the following further conclusions from these two conclusions. First, the construction of the EI models and the data used within those models was dictated by what was available to Economic Insights and not by what was likely to give the most reliable indicator of efficient costs. Secondly, the first conclusion ought to cause the Tribunal significant concern about the reliability of the output of the EI’s models. Read in context, we think this second conclusion means that the Tribunal did have significant concern about the reliability of the output of the EI models. We should add that on occasion the Tribunal made its finding by stating that a submission made was rightly made. We see no difficulty with that as long as it is clear from the context that the Tribunal is adopting that as its finding or conclusion.

222    The Tribunal then addressed the issues raised by the parties about the use of the overseas data. A number of matters were advanced, but we will address only those matters where the Tribunal found error or, to use the language the Tribunal sometimes used, there was a matter of concern.

223    The first matter raised by the electricity network respondents with respect to the overseas data related to whether the parties had an adequate opportunity to verify the accuracy of the overseas data. The Tribunal found that the electricity network respondents had been afforded only a limited opportunity, and PIAC and other interested parties, no opportunity, to provide comments on the AER’s application of a benchmarking methodology which relied on overseas data. The Tribunal acknowledged that this was not a ground of review. However, at [288] the Tribunal said that the lack of opportunity told strongly against the acceptance of the methodology and the resulting estimates of the electricity network respondents’ required opex that the AER derived from it. We are of the opinion that, read in context, what the Tribunal was saying was as follows. The AER was using an ambitious modelling technique for the first time as the sole basis on which to reach an assessment of the quantum of the electricity network respondents’ opex allowance. The electricity network respondents had raised a number of issues about the appropriateness of using the overseas data. The elucidation and determination of those issues would have benefited from a robust exchange of views. That process did not take place and that was a reason to doubt the methodology and the resulting estimates.

224    The second matter raised by the electricity network respondents was that the overseas data contained errors. We do not think that the Tribunal made a finding to that effect, but it did make a finding, at [291], that if the AER’s conclusion that changes in the overseas data were movements attributable to the small size of the overseas DNSPs was correct, then that fact supported a conclusion that the relationship between cost drivers and opex is not consistent across Australian, New Zealand and Ontario DNSPs and that it was not appropriate to compare small New Zealand and Canadian DNSPs which may experience volatile changes to large DNSPs in Australia which do not.

225    The third matter raised by the electricity network respondents was whether the overseas DNSPs adopted different definitions of opex and maximum ratcheted demand for the purpose of collecting data. In the Tribunal’s view, this turned on whether the country dummy variables operated to correct systemic reporting differences. The Tribunal concluded at [296] that they do not because the type of country dummy variables used by Economic Insights assumed that the relevant relationships between cost drivers and opex was the same across the three jurisdictions and, therefore, it did not control for the situation where one of the relevant cost drivers had been defined differently in one jurisdiction, thereby altering the relationship between the cost driver and opex in that jurisdiction.

226    The Tribunal made another important finding in this context. It was in connection with the AER’s response to the electricity network respondents’ arguments that the use of country dummy variables while controlling for level differences between databases do not account for cost relationship differences and each service provider’s costs are influenced by factors not captured by the explanatory variables in the EI Model. In response to this, the AER referred to Economic Insights’ opinion that for such differences to have a material impact on the model results, significant differences in the technology to distribute electricity would need to exist. Economic Insights did not think such a fundamental difference existed because the international service providers deliver the same services using poles, wires and transformers. The Tribunal was not impressed with this explanation. The Tribunal said that there were a multitude of differences between the poles, wires and transformers of the Australian DNSPs and their overseas counterparts, including differences of topography, climate and regulations relating to the standard of the poles, wires and transformers. The Tribunal found that the AER’s explanation glossed over these differences. Furthermore, the AER’s explanation failed to address the fact that the EI Model’s use of country dummy variables impugns the robustness claimed for the model by the AER. The Tribunal considered, at [301]-[302], that the EI Model’s use of country dummy variables revealed a significant weakness in the model.

227    The fourth matter raised by the electricity network respondents was whether the AER or Economic Insights conducted and, if not, should have conducted, sensitivity testing in order to quantify potential errors in the overseas data, or inconsistencies between the overseas data and the Australian data. The way the Tribunal dealt with this criticism of the electricity network respondents was to say, at [306], that there was merit in the opinions for sensitivity testing suggested by the NSW service providers and the fact that they were not followed, does not “increase confidence in the AER’s reliance on overseas data to arrive at its estimate of a DNSP’s required opex”.

228    The fifth matter raised by the electricity network respondents was whether the AER and Economic Insights were correct to ignore other potential sources of overseas data which yielded different results, particularly USA data which is preferable to the Ontario data. Although the Tribunal found, at [308], that this criticism was not made out, it reiterated in this context the significance (although acknowledging it was not a ground of review) of the AER’s failure to consult more fully about the overseas data.

Criticisms of the AER’s decision to lower the EI Model’s Comparison Point

229    The AER lowered the efficiency target comparison point from the 0.86 in its Draft Decisions to 0.77 in its Final Decisions. The AER had used an “average” approach in their Draft Decisions, whereas in its Final Decisions it used the lowest of the efficiency scores in the top quartile of possible scores (represented by AusNet). The AER advanced a number of reasons for making this adjustment which may be briefly summarised as follows: the AER’s averaging approach produced unusual results; a cautious approach was called for in the circumstances where benchmarking was being applied for the first time; and the adjustment was more likely to achieve the National Electricity Objective and the Revenue and Pricing Principles (ss 7 and 7A of the NEL).

230    The lowering of the efficiency target comparison point was attacked by both PIAC on the one hand, and the NSW service providers and ActewAGL on the other, but on different grounds. PIAC attacked the reasons advanced by the AER and the Tribunal said that there was merit in PIAC’s challenge. However, the Tribunal said, at [311]-[312], that there was also merit in the submission made by ActewAGL that the lowering of the efficiency target comparison point was arbitrary and indicated a lack of confidence in the results of the EI Model.

231    The Tribunal said, at [316], that the AER’s reasons for lowering the comparison point contained a significant acknowledgement of the general limitations in its models with respect to the specification of outputs and inputs, data imperfections and other uncertainties and furthermore, did nothing to assuage concern about the use of such a sophisticated econometric model as the EI Model where economic benchmarking is being used for the first time in Australia. The Tribunal, at [317], described the submissions of PIAC and ActewAGL as “cogent”.

Criticisms of the OEF Adjustments made by the AER

232    The Tribunal said that the AER recognised that the EI Model did not account for many OEFs relevant to estimating opex, and the AER made ex post adjustments for them. The adjustment process it adopted was as follows. The first step involved the AER deciding whether an OEF was material or immaterial. If AER decided that the OEF was material, then it would assess its quantum for the particular DNSP in question. The second step involved the OEFs which the AER had decided were individually immaterial. In those cases and where the AER considered the OEF was likely to provide a cost disadvantage, it made a uniform +0.5 percent adjustment, and in those cases where the AER considered the OEF was likely to provide a cost advantage, it made a uniform -0.5 percent adjustment. Where there was a doubt about whether a particular immaterial OEF was likely to provide a cost advantage or disadvantage, the AER made a uniform +0.5 percent adjustment. PIAC described this last category of OEFs as “directionally ambiguous” and this description was adopted by the Tribunal.

233    In considering the criticisms of the AER’s approach to OEF adjustments, the Tribunal adopted the approach we previously referred to of identifying the parties’ respective submissions and indicating its finding or conclusion by stating that the particular party (or expert) rightly submits or states.

234    We have considered the reasons carefully and we think that the Tribunal did make the following findings or drew the following conclusions. First, at [330], the 23 percent OEF adjustments that the AER made to ActewAGL’s efficiency score which was generated by the EI Model demonstrated that even the AER considered there were serious comparability issues that could not be addressed by the specification of the EI Model. Secondly, at [331], for all the consideration of environmental variables individually in the Draft and Final Decisions, the adjustment amount allowed for the collective influence of these variables was merely a subjective estimate. Thirdly, at [333]-[335], the adjustments should be made before modelling, by normalising the data set, rather than ex post modelling. The skewed cost relationship (where adjustments are not made pre-modelling) cannot be corrected by post modelling OEF adjustments made to some only of the DNSPs. Fourthly, at [342] and [344], the AER’s approach to directionally ambiguous OEFs was an indication of the “immaturity” of the data it used. Fifthly, at [347], the directionally ambiguous OEFs will, on average, have a zero opex impact and the AER gave no explanation why it applied adjustments at the outer bounds of the immaterial range for immaterial OEFs that would, by definition, have absolute magnitudes only less than (or equal to) those outer bounds. Finally, the Tribunal did not accept the submission that the AER’s approach that the immaterial OEFs provided the electricity network respondents with the opportunity to recoup at least their efficient costs consistent with the Revenue and Pricing Principles (RPP). It described, at [353], that approach as “tenuous”, which ordinarily means insignificant or insubstantial.

Criticisms of the AER’s approach to Vegetation Management Costs

235    The Tribunal addressed the challenges of Endeavour, Essential and ActewAGL respectively to the AER’s assessment that the vegetation management costs in their respective 2012-13 base year opex were not efficient.

236    Endeavour challenged AER’s refusal to provide an allowance for increased vegetation management costs caused by its retendering of outsourced contracts for that purpose. It claimed that the AER’s refusal resulted in a reduction of some $240.7 million in Endeavour’s forecast of its vegetation management opex.

237    Endeavour’s case was that its forecast of its vegetation management opex was required to enable it to comply with a NSW Industry Standard known as Industry Safety Steering Committee 3 Standard – Guidelines for Managing Vegetation Near Powerlines, December 2005 (ISSC 3) and directed at meeting the opex objectives in r 6.5.6(a)(2), (3) and (4). It sought new contracts by competitive tender for trimming vegetation commencing in 2011-12 because of previous problems. The new contracts increased opex costs, but resulted in increased compliance with ISSC 3. Endeavour relied on the competitive tendering process as evidence of efficiency in contractor costs and provided evidence of its increased vegetation management costs to the AER. The AER’s EFA Guideline addressed increases in opex forecasts. The guideline referred to “step changes” which are costs not captured in base opex or the rate of change that is required for forecast opex to meet the opex criteria. The AER was not prepared to accept the forecast opex as a step change.

238    The Tribunal found, at [372], that the AER’s approach based on whether the forecast opex was a step change was at the forefront of its analysis and that the AER did not consider the fact that Endeavour’s approach in outsourcing its vegetation management contracts by way of competitive tender was sufficient to demonstrate that Endeavour’s vegetation management opex forecast reasonably reflects the r 6.5.6(c) opex criteria or the opex objectives.

239    The Tribunal described the AER’s reasons for rejecting Endeavour’s forecast of its vegetation management opex as “tenuous” in light of the evidence Endeavour had provided to the AER and the fact that Endeavour was committed to paying contractors retained through a competitive tender process so that it may comply with an applicable regulatory obligation. The Tribunal, at [373], said that this conclusion was reinforced when regard was had to “the significant adverse consequences that may flow from a failure to comply with regulatory vegetation management requirements as demonstrated by the Victorian bushfires”.

240    There was another issue in relation to Endeavour’s forecast of its vegetation management opex. The AER contended that its assessment was influenced in part by a report prepared by Deloitte entitled NSW Distribution Network Service Providers Labour Analysis, Final Addendum to 2014 Report, 28 April 2015 (the 2015 Deloitte Labour Report). The AER had previously engaged Deloitte to conduct an analysis of the NSW DNSP’s labour costs in the 2009-14 regulatory period and it prepared a report entitled NSW Distribution Networks Service Providers Labour Analysis, 17 November 2014 (the 2014 Deloitte Labour Report). The Tribunal found that the report informed the AER’s assessment of the DNSP’s 2015-2019 capex and opex forecasts, and was referred to in the Draft Decisions for each of the electricity network respondents.

241    The Tribunal noted that in the 2015 Deloitte Report, Deloitte expressed the following conclusion:

Our view remains that the NSW DNSPs have higher labour costs than their peers (driven by the number of employees rather than costs per employee) due to in [sic] restrictive EBA provisions, a high degree of unionisation and inefficient labour practices, which means that their base year opex was not efficient.

242    The Tribunal noted that the NSW service providers did not have the opportunity to be heard in relation to the 2015 Deloitte Report and addressed the significance of that fact. It again noted that a failure to accord procedural fairness was not a ground of review in s 71C(1) of the NEL. It acknowledged the administrative challenges faced by the AER in light of the decisions it had to make and the time constraints within which it was required to do so. The NSW service providers argued that their lack of opportunity to be heard was part of the context in which its grounds of review were to be assessed. The Tribunal’s approach to the issue was as set out in the following passage in its reasons at [378]:

Certainly, it is a matter of common sense that a report such as the 2015 Deloitte Labour Report might carry greater weight if it had been the subject of any response from Endeavour, depending of course on the terms of that response.

243    In the Endeavour separate reasons, the Tribunal repeated some of the points Endeavour made about its forecast of vegetation management opex and redundancy costs in the Ausgrid decision. The Tribunal said, at [44], that it did not need to address Endeavour’s assertions “fully”. That was because the AER’s rejection of the vegetation management expenditure was infected by the same error which affected its benchmarking exercise and the EI Model, in particular. The reason for that, at [45], was that the AER relied upon its benchmarking exercise to find that Endeavour was not materially inefficient, but only marginally so with the consequence that any additional expenditure could not be added without jeopardising the finding of efficiency, and because the AER relied upon its benchmarking exercise to support the proposition that comparator networks operate safe and reliable networks. The Tribunal said that insofar as this aspect of the AER’s decision included redundancy expenditure, the matter was common to the NSW service providers and was dealt with in the Ausgrid decision. The cogency of these findings depends upon whether the Tribunal’s conclusions with respect to the AER’s reliance on the EI Model are upheld.

244    Essential’s case was that the AER wrongly concluded that Essential’s proposed step down in its vegetation management opex supported its reliance on the EI Model. Essential submitted that the AER did not address the critical questions which were whether its forecast vegetation management opex over the 2014-19 period was efficient or prudent, made no attempt to quantify any inefficiencies in terms of the effect on opex of the vegetation management issues identified and did not ascribe a figure quantifying inefficiencies in Essential’s vegetation management.

245    The Tribunal reached the following conclusions. First, neither the NEL nor the NER mandated a line-by-line bottom-up review of each category of forecast opex. Secondly, nevertheless, in circumstances where benchmarking in Australia is in its infancy, “sensible administration” dictated that the AER should not have cast aside its previous practice of conducting bottom-up reviews in favour of the emphasis it placed on benchmarking. This conclusion was reinforced by the fact that the AER’s preferred EI Model’s reliance on overseas data and the AER’s final OEF adjustments could not have the benefit of full exposure to the consultation process mandated by the NEL and the NER. The Tribunal said, at [390]:

Viewed in that context, the AER’s apparently untested conclusion that the recommendations of the Select Solutions Report could not have been implemented in the 2012–13 base year and its preference for its assessment of Essential’s overall opex based on the EI model are unconvincing. Likewise, the Tribunal is not convinced by its dismissal of Essential’s submission that the AER should have quantified its vegetation management inefficiencies and should not have ignored Essential’s comparative vegetation management opex per vegetation management performance.

246    In the Essential separate reasons, the Tribunal described Essential as an “outlier” in terms of network length (particularly route length) and sparsity of customers. It noted that Essential has longer distances to travel to carry out asset inspection and maintenance and that some of its network was in remote and challenging country, even to the point, as the Tribunal put it, of “impaired telephone access”. The Tribunal noted that Essential maintained a larger number of local depots across rural New South Wales so that it is able to respond in a timely way to network difficulties. It noted that Essential and the other electricity network respondents submitted that these factors supported its contentions about the validity of the EI Model. With respect to forecast opex, the Tribunal in the Essential separate reasons reached the following conclusions, at [26]-[30]:

Those features of Essential’s network, apart from illustrating the contentions about the appropriateness of the EI Model (addressed in the PIAC-Ausgrid Decision), were said to support Networks NSW’s criticism of the AER for regarding a Local Service Agent model for service delivery as available, and so Essential’s service delivery structure and costs as inefficient. In part, they also were said to support the criticism of the AER in concluding that Essential’s Revised Regulatory Proposal concerning its existing and projected vegetation management costs were inefficient.

All of those matters are addressed in some detail in the opex section of the PIAC-Ausgrid Decision. They are features of Essential’s particular circumstances which were relevant to the reasoning of the Tribunal about why the inputs into the structure of the EI Model led to grounds of review of the AER’s decision that each of Ausgrid and Essential had, and proposed, inefficient opex for the current regulatory period and that it should adopt its own assessment of the efficient opex for both Essential and Ausgrid based upon the EI Model and the benchmarking it referred to.

For the reasons explained in detail in the PIAC-Ausgrid Decision, the Tribunal is satisfied that common grounds of review as asserted are made out. The AER’s reasons for substituting the opex allowance at the figure it did, having found that Essential’s opex was not efficient, expose the same flaws as the Tribunal referred to in the PIAC-Ausgrid Decision. It is not necessary to repeat those reasons. At the conclusion of these reasons, the Tribunal addresses the question of whether any determination should be made as a result of that conclusion.

PIAC’s contentions with respect to opex, if successful, would have led to a further significant reduction in the opex allowance for each of the Networks NSW entities. It sought to restore the opex allowance to that proposed by the AER in its respective Draft Decisions. Maintaining in broad terms the methodology adopted by the AER, PIAC contended that the AER erred by reducing the benchmark comparison point below the weighted average of the upper quartile (0.86) of the AER’s analysis and, secondly by making unreasonable OEF adjustments for unquantifiable, immaterial and directionally ambiguous OEFs.

As explained in the PIAC-Ausgrid decision, those contentions were dependent on the starting premise about the broad correctness of the AER’s methodology. The Tribunal has not accepted that premise.

247    With respect to ActewAGL, the issues can be dealt with relatively briefly.

248    In addressing ActewAGL’s opex for labour and vegetation management, the AER relied on, among other matters, a review carried out by EMCa. The Tribunal noted the AER found that the EMCa Report identified that there were significant issues in those categories of ActewAGL’s opex which the AER considered evidence of base year efficiency, supporting its benchmarking results. The Tribunal, at [400]-[401], considered the EMCa Report and other evidence in the course of which it identified errors by the EMCa and AER respectively.

249    The Tribunal found that the EMCa Report was no more than a “limited scope review” or, put another way, but a “desk top” qualitative review. The Tribunal found that the EMCa Report relied on, but misconstrued, earlier reports commissioned by ActewAGL. The Tribunal said, at [408]:

Where, as here, the application of a new untested benchmarking model is applied to arrive at a total opex figure, sensible administration suggests that the regulator responsible for its application would apply some form of quantitative “reasonableness check bottom-up analysis to at least some, if not all, of the opex components. That is, however, not the case here.

250    In the ActewAGL separate reasons, the Tribunal first made the point that it did not accept that the AER had not taken the step of critically considering ActewAGL’s forecast opex. However, it was satisfied that grounds of review within s 71C(1) were made out in relation to the AER’s benchmarking methodology. The Tribunal identified matters particular to ActewAGL’s network which fortified its conclusions with respect to the benchmarking methodology. It said that it was not necessary for it to determine the issues relating to labour costs or vegetation management costs.

The Criticisms of the AER’s approach to the Labour Costs of the NSW Service Providers

251    The NSW service providers challenged the AER’s findings that inefficiencies in their labour management practices were, in part, responsible for the gap between them and the frontier electricity network respondents identified in the AER’s economic benchmarking analysis.

252    We have set out above Table 7.3 in the Ausgrid Final Decision which identifies the AER’s main techniques for testing the efficiency of Ausgrid’s base opex (at [204]). That table identifies the AER’s findings with respect to labour and workforce practices and the conclusions of Deloitte. We have already referred to the 2014 Deloitte Report and the 2015 Deloitte Report in the context of the electricity network respondents’ vegetation management practices. The position with respect to Endeavour and Essential is also set out above (at [205]).

253    In the 2015 Deloitte Report, Deloitte referred to the relatively high number of employees of the NSW service providers compared with private DNSPs in the National Electricity Market and expressed the opinion that the current high number of employees was likely being sustained “by restrictive EBA provisions relating to no forced redundancies and a relatively high proportion of employees employed under EBAs”.

254    The Tribunal referred to a submission put to the AER by the NSW service providers that compliance with an EBA is a “regulatory obligation or requirement” within r 6.5.6(a)(2). They submitted that, in those circumstances, the AER’s analysis of the NSW service providers' labour costs did not establish that the electricity network respondents’ revised regulatory proposals included any labour costs which did not reasonably reflect the operating expenditure criteria.

255    The Tribunal set out the terms of r 6.5.6(a)(2) and the relevant part of the definition of “regulatory obligations or requirements” in s 2D(b)(v) of the NEL and it referred to the fact that the AER rejected the submission that the EBAs were regulatory obligations or requirements. The Tribunal did not find it necessary to decide whether the AER’s decision was correct.

256    The Tribunal reasoned in the following way. First, the EBAs may reasonably be regarded as otherwise required to achieve an opex objective, namely, the r 6.5.6(a)(4) objective to “maintain the safety of the distribution system through the supply of standard control services” and reasonably reflecting the opex criteria in r 6.5.6(c)(3) of “a realistic expectation of the demand forecast and cost inputs required to achieve the operating expenditure objectives”. Secondly, the AER’s focus on benchmarking, in particular, the EI Model and the total opex outcomes, led it to treat the EBAs as an endogenous factor to be ignored in the AER’s estimate of the total required opex made pursuant to r 6.12.1(4)(ii). Thirdly, the EBAs may not be a regulatory obligation or requirement, but nevertheless the NSW service providers are bound by them as a matter of law. Fourthly, although the 2014 Deloitte Report and the 2015 Deloitte Report suggested inefficiencies in the NSW service providers’ labour practices, the reports do not quantify those inefficiencies and provide no corroboration of inefficiencies of the scale identified by the AER in reliance on the EI Model. Fifthly, the EBAs should not be viewed as an endogenous managerial choice at least in circumstances where the AER has shifted from an itemised bottom-up approach to assessing opex to benchmarking total opex per se, particularly where the benchmarking had not been exposed to the rigors of the consultation the NEL and NER envisage for such a radical change. The NSW service providers’ were shackled with the EBAs that effectively restricted their ability to reduce the size of their workforce to achieve efficiency and that restriction was due to an exogenous factor, being the Fair Work Act 2009 (Cth). Sixthly, the policy of the legislature was that if the EBAs resulted in inefficiency, the cost of that inefficiency was to be borne by consumers of electricity. Finally, the AER used the EI Model to select the measurement of efficiency. It did so without regard to the obligations under the EBAs as they presently existed.

The Electricity Network Respondents’ submission that the AER used the EI Model as the Sole Determinative of Opex

257    The Tribunal said that it accepted the NSW service providers’ submission that the AER had used an experimental model as the sole determinant of opex, contrary to sensible regulatory practice, including significant experience of modelling in other jurisdictions. The Tribunal referred at some length to how the regulator in the United Kingdom, the Office of Gas and Electricity Markets Ofgem (Ofgem), proceeded in similar circumstances. It noted that Ofgem had over a decade’s experience in benchmarking. The Tribunal referred to the differences between Ofgem’s approach to benchmarking and that adopted by the AER, for example, Ofgem uses top-down models and bottom-up models, adjusts data prior to modelling and has had many years to develop its data collection techniques.

258    The Tribunal also accepted ActewAGL’s submission that the AER arrived at its own estimate of opex based on the EI Model that was lower than the electricity network respondents’ and, as it was not satisfied that there was an explanation for the difference, rejected the electricity network respondents’ forecast and deemed its own estimate as the appropriate estimate. The electricity network respondents’ forecast did not otherwise play a role in the AER’s decision-making process, whether as a “starting point” or otherwise: at [458].

259    The Tribunal found, at [461], (quoting from the NSW service providers’ expert, Frontier, in the Frontier Report at p 105) that “… the AER had put undue faith in the ability of it and its advisers to develop a single benchmarking model (or suite of very closely related models, all derived from the same data and missing the same wider review of factors and sense checks) that can capture very well relative inefficiency”.

The Tribunal’s consideration of the “Principal Opex Issue”

260    The Tribunal described the way in which the parties in their submissions approached the principal issue. The principal issue was whether the AER’s application of the EI Model discharged its obligations under rr 6.5.6 and 6.12.1(4) of the NER. The Tribunal said that the first stage or level involved a consideration of the effect of the 2012 Rule Amendments, particularly the changes to r 6.5.6 and other rules relevant to its interpretation and application. The second stage or level arose if it was established that the AER’s application of the EI Model failed to discharge its obligations. In that event, it was necessary to consider the effect of the 2013 Legislative Amendments, particularly the introduction of s 71P(2a) and (2b) in the NEL.

261    The Tribunal said that it had already noted in its reasons that there were a number of issues with the EI Model and the AER’s application of the EI Model. The Tribunal identified those issues as follows, at [467]:

(1)    Inadequacies in the EI Model’s data set and comparability issues (i.e., the RIN data, the overseas data and the country dummy variables);

(2)    The AER’s decision to lower the EI Model’s comparison point;

(3)    The OEF adjustments made by the AER;

(4)    The AER’s reliance on qualitative analysis rather than bottom-up quantitative assessment to test issues such as those raised by the electricity network respondents in relation to vegetation management opex in circumstances where economic benchmarking is in its infancy in Australia; and

(5)    The AER’s use of the EI Model as the sole or principal determinative of opex

262    The Tribunal referred to the first step in the decision-making process, namely, whether the AER should accept the forecast opex of a DNSP under r 6.5.6(c) and found that, having regard to the AER’s concerns about the Revised Regulatory Proposals submitted by the NSW service providers and ActewAGL, it was not persuaded that the AER’s lack of satisfaction with respect to that matter exposed a ground of review. The Tribunal said that there was material upon which the AER could decide not to accept the Revised Regulatory Proposals. The AER’s decision under r 6.5.6(d) that it was not satisfied of the matter identified in r 6.5.6(c) did not, according to the Tribunal, at [468], “enliven any grounds of review under s 71C of the NEL”.

263    The AER was then obliged under r 6.12.1(4)(ii) to provide an estimate of the total of the DNSP’s required operating expenditure for the regulatory control period that “the AER is satisfied reasonably reflects the operating expenditure criteria, taking into account the operating expenditure factors”. It was in this respect that the Tribunal found error by the AER falling within the grounds for review in s 71C(1) of the NEL. The Tribunal explained its reasons for that conclusion in the following passages at [471], [472]:

As is apparent from the above, there are a number of respects in which, or reasons for, the Tribunal on these applications being of the view that one or more of the grounds of review under s 71C(1) are made out. At a general level, that is because the AER placed too much weight on the outcome of the EI model. That, in the Tribunal’s view represents an exercise of the AER’s discretion about the use to which the EI model should have been put which was incorrect.

Underlying that view are a series of concerns about the inputs to the EI model, and the OEF adjustments (including those of concern to PIAC), and including the AER’s treatment of the vegetation management costs of Essential, Endeavour and ActewAGL, and further including the AER’s treatment of the labour costs of the Networks NSW DNSPs. Those concerns can generally be described as errors of fact by the AER in its findings of fact, as discussed in detail above. Those errors do not simply reflect the AER’s choice of competing expert views. There are underlying elements to the EI model which mean that the AER at this point (accepting that the available Australian data is not sufficiently extensive for appropriate modelling) should not have placed the weight it did on the output of the EI model. As the earlier Introduction to these reasons discuss, there may be room for debate about whether a particular step shows an error of fact in a finding of fact, or is an incorrect exercise of a discretion. It would be possible, in a number of the specific instances (in particular in relation to the OEFs) to use either description by the use of different semantics. The line between the two is often hard to draw. The Tribunal, having regard to its conclusion in the preceding paragraph, does not think it is helpful to embark on that exercise.

264    It is necessary to analyse carefully the points the Tribunal was making in these passages. First, at a general level, the AER committed an error within s 71C(1) described as an incorrect exercise of discretion by placing too much weight on the outcome of the EI Model. We will need to consider later in these reasons what the Tribunal meant by “too much weight”. Secondly, underlying that conclusion were a number of concerns which were those matters summarised above (at [263]), and labour costs. Those concerns can generally be described as errors of fact and were not simply a choice between the competing views of experts. There was room for debate as to whether a particular step could be described as an error of fact or an incorrect exercise of discretion, and the line was often difficult to draw. The Tribunal was not going to attempt the task of drawing the distinction in view of its conclusion that the AER had placed too much weight on the outcome of the EI Model.

265    In the longer of the two passages set out above, the Tribunal refers to a discussion earlier in the Introduction to its reasons. The only section in the Introduction which may answer this description is [102]-[108]. In [104], the Tribunal said:

As will be seen, the Tribunal is well alive to the terms of s 71C of the NEL and s 246 of the NGL. The onus is on a particular applicant to satisfy the Tribunal that the AER determination, in the respect being debated, is in error within one or more of the available grounds: Application by DBNGP (WA) Transmission Pty Ltd (No 3) [2012] ACompT 14 at [483]-[486]. See also the Tribunal’s remarks in Application by EnergyAustralia [2009] ACompT 8 at [70]; Application by WA Gas Networks (No 3) [2012] ACompT 12 at [22] (WA Gas Networks).

We have considered each of the Tribunal decisions referred to in this passage. They make the point that the Tribunal may intervene only where one of the grounds of review is made out and it may not intervene simply because it preferred a different decision from that of the primary decision-maker.

266    The Tribunal may have been referring to, or also referring to, its discussion of the differences between factual error, opinion and discretionary judgement in its reasons on the leave applications (Applications by Public Interest Advocacy Centre Ltd, Ausgrid, Endeavour Energy and Essential Energy [2015] ACompT 2 at [33]-[59], especially at [57]; Application by ActewAGL Distribution [2015] ACompT 3 at [21]).

267    In those decisions, the Tribunal said that the line between the several available grounds of review is not necessarily always clear cut and noted there is no prescription in s 71C that, in particular facts and circumstances, there can be only one ground of review established. Although it is not sufficient to establish error merely to show that the Tribunal might have preferred to have made a different finding of fact, nevertheless where a material factual error is established, then a ground of review is made out. In addition, the factual error may lead the Tribunal to conclude that the AER’s decision is, in all the circumstances, unreasonable. The Tribunal also observed that there is no clear line between factual error, opinion and discretionary judgement, and it said:

One may feed into the other.

268    The Tribunal discussed the different ways in which Ausgrid’s challenge to the data used in the EI Model and the use of the model may be analysed in terms of factual error, opinion and exercise of discretion. In the course of this discussion, the Tribunal acknowledged that factual error is not shown by “the legitimate making of choices between permitted methodologies” or “by the weight legitimately given to competing factual data or to competing regulatory decisions”, or “by legitimately formed and maintained opinion” (Applications by Public Interest Advocacy Centre Ltd, Ausgrid, Endeavour Energy and Essential Energy [2015] ACompT 2 at [57]). The Tribunal said that the errors identified by ActewAGL may, if accepted, be a combination of error or errors of fact, wrongful exercise of discretion, and/or the outcome of an unreasonable decision (Application by ActewAGL Distribution [2015] ACompT 3 at [21]).

269    A little later in its reasons when considering the effect of the 2012 Rule Amendments, the Tribunal expanded upon its reasons for finding errors within s 71C(1). It said that because of inherent weaknesses in the EI Model and the ex post adjustments to its outcomes, the AER was wrong to rely on the EI Model to estimate the DNSP’s required opex. The Tribunal said, at [479], that the AER, in considering the operating expenditure factors in r 6.5.6(e), gave discordant weight to r 6.5.6(e)(4) in a context where the AER was applying benchmarking for the first time.

270    The Tribunal then addressed two issues briefly. The first was the application of s 71P(2a) and (2b) and the Tribunal said that it was premature to consider these subsections at that stage in its reasons. It was necessary to consider first the challenges to the other elements of the decisions. The second was the electricity network respondents’ submission that the AER should have allowed a transition path for the reduction of opex. The Tribunal said in view of its findings on opex, it was not necessary for it to consider that submission.

271    The Tribunal then summarised its conclusions on opex, other than the application of s 71P(2a) and (2b) of the NEL. The Tribunal said that the AER’s reliance on the EI Model failed to discharge its obligations under rr 6.5.6 and 6.12.1(4) of the NER. The AER treated the EI Model as the determinative factor in estimating each DNSP’s required opex pursuant to r 6.12.1(4)(ii). That was in a context where the model had limitations and benchmarking was being used for the first time to set opex allowances. The Tribunal also referred, at [496], to the restricted opportunity afforded to the parties to test the veracity of the EI Model and the AER’s restricted opportunity to carry out the required consultation in relation to the EI Model.

Analysis of the AER’s Grounds of Review

272    Some of the AER’s grounds of review under the ADJR Act in relation to forecast opex overlap in terms of the error alleged, or even though the errors alleged are different, they relate to the same subject matter. We find it convenient to deal with Grounds 6, 7, 9 and 21 together, Ground 8 separately, and Grounds 10 and 11 together.

273    Before turning to the grounds of review, the following general matters should be noted.

274    The Tribunal must keep its determinations and the statement of the reasons for the determination in the register established under regulation 9C of the Competition and Consumer Regulations 2010 (Cth). Under the NEL where it decides to vary a reviewable regulatory decision or to set it aside and remit it back to the AER, the Tribunal must “specify” in its determination certain matters. In the case of an order of the latter type, it must specify the manner in which it has taken into account the inter-relationship between the constituent components of the reviewable regulatory decision and how they relate to the matters raised as a ground of review which is a matter it is required to consider (s 71P(2b) and (2c)). There are no other sections which direct the Tribunal to include matters in its determination or reasons.

275    The procedure of the Tribunal is within its own discretion. It is not bound by the rules of evidence and the proceedings before the Tribunal shall be conducted with as little formality and technicality, and with as much expedition, as the requirements of the Act and a proper consideration of the matter before the Tribunal permit (Competition and Consumer Act, s 103(1)).

276    In addition to that requirement for expedition, s 71Q of the NEL provides tight target times (with a power to extend) for the Tribunal to make a determination. This is in a context where from time to time the Tribunal deals with complex matters. We were told that in this case the hearing before the Tribunal on all issues involved 15 parties, 21 counsel, 20 days of hearing, 2 days of community consultation, over 2,800 pages of submissions, and a transcript totalling over 1,360 pages.

277    The Tribunal is an administrative body and it is well-settled that the reasons of such a body “are not to be construed minutely and finely with an eye keenly attuned to the perception of error”: Collector of Customs v Pozzolanic Enterprises Pty Ltd [1993] FCA 456; 43 FCR 280 at 287; Minister for Immigration and Ethnic Affairs v Wu Shan Liang [1996] HCA 6; 185 CLR 259 at 272.

278    As we have said, Ground 7 of the applications under the ADJR Act alleges that the Tribunal did not take into account reasons given by the AER or arguments raised by the AER. We will come to deal with the particular matters raised by the AER. At this stage, and at a general level, we note the observation of the Full Court of this Court in Applicant WAEE v Minister for Immigration and Indigenous Affairs [2003] FCAFC 184; 236 FCR 593 at [47]):

The inference that the Tribunal has failed to consider an issue may be drawn from its failure to expressly deal with that issue in its reasons. But that is an inference not too readily to be drawn where the reasons are otherwise comprehensive and the issue has at least been identified at some point. It may be that it is unnecessary to make a finding on a particular matter because it is subsumed in findings of greater generality or because there is a factual premise upon which a contention rests which has been rejected. Where however there is an issue raised by the evidence advanced on behalf of an applicant and contentions made by the applicant and that issue, if resolved one way, would be dispositive of the Tribunal’s review of the delegate’s decision, a failure to deal with it in the published reasons may raise a strong inference that it has been overlooked.

The Tribunal’s approach to the AER’s use of the EI Model and the Tribunal’s underlying concerns (Grounds 6, 7, 9 and 21)

279    The AER submitted that the Tribunal’s conclusion that the AER placed too much weight on the EI Model in estimating the relevant DNSP’s forecast opex for the purposes of r 6.12.4(ii) of the NER and that this represented an incorrect exercise of discretion was based on a series of concerns (about the use of Australian data and overseas data, the selection of the benchmark comparison point, the use of the OEFs and the assessment of vegetation management costs) concerning which the Tribunal made no findings of relevant error, that is to say, the type of error which falls within s 71C(1). In other words, the Tribunal’s conclusion about the AER exercising a discretion incorrectly cannot be sustained because it was not based on errors that fall within s 71C(1). The AER also submitted that in reaching its conclusion about the AER’s approach to the Australian data and normalising the data before modelling, the Tribunal failed to take into account certain matters and that the Tribunal failed to consider that the AER had lowered the target or benchmark comparison point to allow for any imperfections or limitations in the data or modelling process. The AER submitted that the matters not considered by the Tribunal were material aspects of the AER’s decision in relation to opex. The AER submitted that the Tribunal approached its task as if it was undertaking “some form of de novo merits review”. The AER submitted that the Tribunal erred in concluding that the AER had placed too much weight on the EI Model and that this was an incorrect exercise of discretion.

280    First, the AER submitted that the Tribunal started on the wrong basis because it proceeded on the basis that the AER had used the EI Model as the sole determinant of opex. The AER submitted that that was not correct and pointed to the fact that the first stage of the AER’s decision-making, that is to say, in deciding whether to accept the Revised Regulatory Proposals submitted by the electricity network respondents, the results of the EI Model were corroborated by the results of the other models and techniques the AER employed and the fact that the AER’s final estimates were reached after the results produced by the EI Model had been adjusted by the OEF adjustment process which we have described.

281    We reject this submission. It seems that the Tribunal found that the AER used the EI Model as the sole determinant of opex. It said, at [443]-[444]:

It is Networks NSW’s submission that the AER has used an experimental model as the sole determinant of opex, contrary to sensible regulatory practice including significant experience of modelling in other jurisdictions.

Networks NSW’s submission is cogent.

282    It should be noted that the context in which these observations were made was one where Economic Insights itself said that the approach it was advancing was the first time economic benchmarking was being used as the primary basis for an Australian regulatory decision (see [200] above).

283    The Tribunal approached the matter as one of substance and, in our opinion, it was entitled to do so. It referred to the other models and techniques at some length. The other two econometric models developed by Economic Insights used the same data from Australia, New Zealand and Ontario and the same explanatory variables as the EI Model. These models suffered from the same defects as the EI Model. The fourth model was the MPFP model which used only the Australian data and five outputs. It did not use overseas data and it used a different output specification. In other words, the MPFP model used only a subset of the data used for the three econometric models. In our opinion, it was not erroneous for the Tribunal to find (as we think it did at [461]) that the other models were “very closely related models” to the EI Model which “all derived from the same data and missing the same wider review of factors and sense checks”. The Tribunal was aware of the adjustment for OEFs carried out by the AER and, as we have said, discussed the process at length. It found that the process was flawed for reasons it gave and we will address the AER’s challenge to that conclusion in due course. Accepting the correctness of the conclusion for present purposes, we do not think that the Tribunal erred in concluding that the AER effectively used the EI Model as the sole determinant of opex. Finally, the Tribunal’s conclusion must be considered in the context in which it was expressed. That context was the Tribunal’s consideration of the proper approach (which finds expression in its order and directions) being the use of a number of models, including a bottom-up assessment, rather than a single model with few explanatory variables. In a context where the Tribunal was making it clear that the AER did not use the former approach, the Tribunal’s conclusion that the AER had effectively used the EI Model as the sole determinant of opex was not an error.

284    Secondly, the AER submitted the Tribunal’s conclusion that the AER had erred in placing too much weight on the EI Model was dependent on its conclusion that it had a “series of concerns” and that if none of those concerns involved a reviewable error within s 71C(1), then its conclusion at the general level that there had been an incorrect exercise of discretion must fall away. We will address the issue of whether the “concerns” expressed by the Tribunal involved reviewable errors within s 71C(1) when we address the particular matters later in these reasons. For reasons we will give, we think at least some of them did involve errors within s 71C(1). In any event, we think the submission that the Tribunal’s conclusion at the general level was dependent on the underlying concerns being reviewable errors within s 71C(1) must be rejected.

285    We think that on the proper construction of the Tribunal’s reasons, it was identifying an independent and freestanding error at the general level. It is true that the Tribunal said that underlying its view at the general level were a series of concerns about the EI Model which concerns it identified, but it also said that there were underlying elements to the EI Model which meant that the available Australian data was not sufficiently extensive for appropriate modelling and the AER should not have placed the weight it did on the output of the EI Model. The Tribunal’s observations at [496(a)] are important in this context:

The AER’s undue reliance on the EI model as a determinative factor in the AER’s estimation of each DNSP’s required opex pursuant to r 6.12.1(4)(ii). That reliance being placed on the model notwithstanding that it recognised it had limitations with respect to the specification of outputs and inputs, data imperfections, and other uncertainties in a context where economic benchmarking is being used for the first time to set opex allowances – see eg: Attachment 7 to the Endeavour Final Decision, at pp 7-268 and 7-269, Attachment 7 to the Ausgrid Final Decision, at p 7-64 and Attachment 7 to the ActewAGL Final decision at p 7-250.

These observations are important because the Tribunal refers to error in light of recognised limitations (not error because of a series of underlying errors) and to data imperfections and other uncertainties (emphasis added).

286    In our opinion, the Tribunal was identifying two independent errors (treating the underlying concerns or errors as a single category) by the AER. In other words, the reliance on the EI Model with all its recognised features was a freestanding error. That is why the Tribunal did not need to decide whether the concerns it expressed involved errors of fact or incorrect exercises of discretion.

287    Thirdly, the AER submitted that even if the error which the Tribunal identified at the general level was treated by the Tribunal as a separate and freestanding error, it was not an error within the grounds for review within s 71C(1). The AER pointed to the two aspects of the Tribunal’s decision in support of this submission, being the reference to too much weight and an incorrect exercise of discretion. As to the former, the AER referred to two passages in the Tribunal’s reasons. First, the Tribunal said, at [471]:

At a general level, that is because the AER placed too much weight on the outcome of the EI model.

Secondly, the AER pointed to the first sentence in [496(a)] of the Tribunal’s reasons which we have already set out (at [285]).

288    As to the latter (i.e., an incorrect exercise of discretion), the AER emphasised that this was the ground upon which the Tribunal relied and not s 71C(1)(d) which provides a ground of review where the AER’s decision was unreasonable, having regard to all of the circumstances. The AER submitted that it was not open to the electricity network respondents to seek to uphold the Tribunal’s decision by reference to the unreasonableness ground because that was not a ground relied on by the Tribunal.

289    Before assessing the AER’s third submission, it is necessary to note some other aspects of the Tribunal’s decision. As we have said, the Tribunal’s reasons must be read as a whole.

290    In a number of sections in its reasons, the Tribunal described the principal issue in relation to opex as being whether the AER’s application of the EI Model discharged its obligation under rr 6.5.6 and 6.12.1(4) of the NER. The following are examples:

123.    The principal issue that may be drawn from the following overview of the parties’ challenges is whether the AER’s application of what the parties referred to as the EI model discharged its obligations under rr 6.5.6 and 6.12.1(4).

227.    Networks NSW and ActewAGL’s main submissions addressing the principal issue whether the AER’s application of the EI model discharged its obligations under rr 6.5.6 and 6.12.1(4) are addressed …

463.    Conceptually, the parties submissions address the principal issue (whether the AER’s application of the EI model discharged its obligations under rr 6.5.6 and 6.12.1(4)) at two levels.

495.    Having regard to the DNSPs and PIAC’s submissions as a whole the Tribunal concludes that the AER’S reliance on the EI model failed to discharge its obligations under rr 6.5.6 and 6.12.1(4).

291    It is appropriate in this context to refer to how senior counsel for the AER described the principal issue before the Tribunal. It was quoted by the Tribunal and we have set it out above (at [193]). Senior counsel for the AER said that non-compliance with the regulatory rules would involve an incorrect exercise of discretion. Furthermore, reliance on a flawed model would involve an incorrect exercise of discretion or the making of an unreasonable decision. We refer to these comments, not because they are to be treated as some form of admission, but because they provide an important part of the context in which the Tribunal made its decision.

292    Returning then to the AER’s third submission, there is no doubt that the Tribunal does not have a purely substitutive function in terms of its review of the AER’s decision. The fact that the Tribunal would have preferred a different expert to that preferred by the AER, or put more weight on a particular matter than the AER, are not matters which of themselves give rise to reviewable error within s 71C(1). However, in our opinion, that is not what the Tribunal did in this case. It found that the flaws in the AER’s approach were more fundamental than that and amounted to a failure to comply with rr 6.5.6 and 6.12.1(4), and that led to an incorrect exercise of discretion. The AER had a number of options as to how it proceeded to make its estimate of forecast opex. The fact is that up until the present decision, the AER had adopted what its counsel described as the traditional form of benchmarking often described as a bottom-up or engineering approach (see the passage referred to below at [379]). That in whole or in part was an option as was a broader range of modelling or a combination of these methods. Instead, it used a model with recognised limitations as the sole basis or, at least even on the AER’s case, the primary basis for arriving at its estimate in circumstances where that was being done for the first time for an Australian regulatory decision. The Tribunal was entitled to conclude (as it did) that that was an incorrect exercise of discretion. Furthermore, there is reason to think, as the electricity network respondents submitted, that the Tribunal did (implicitly) or would have found that, in addition, the AER’s decision was an unreasonable decision, having regard to all of the circumstances. The way in which senior counsel for AER described the principal issue and the conclusion that the AER failed to comply with rr 6.5.6 and 6.12.1(4) would support such a conclusion see Chan Yee Kin v Minister for Immigration and Ethnic Affairs [1989] HCA 62; (1989) 169 CLR 379. The AER accepted that a lack of reason or logic would make out the unreasonableness ground, but said that the Tribunal made no such finding. In view of our conclusion concerning an incorrect exercise of discretion, we do not need to address this issue any further.

293    There are two further points to be made at this stage. First, the AER submitted that in considering the AER’s reliance on the EI Model the Tribunal referred to the opinions of experts engaged by the electricity network respondents, but did not refer to the answering report of EI which contained (it was submitted) opinions which could not be described as unreasonable. We reject this submission. The Tribunal was clearly aware of the Second EI Report and made extensive reference to it (see, for example, at [145], [162], [170]-[174]). Secondly, the parties made detailed submissions to the Tribunal about the proper construction of r 6.5.6. The Tribunal did not find it necessary to rule on these submissions in view of its conclusion, at [479], that “the AER was, (because of the inherent weaknesses in the EI model and the ex post adjustments to its outcomes) wrong to rely on the EI model to estimate the DNSPs’ required opex”. However, consistent with its other conclusions, the Tribunal did say the following, at [480]:

Suffice to say at this point that in a context where it is applying benchmarking for the first time, the AER’s application of the EI model gave a discordant weight to r 6.5.6(e)(4) (benchmark opex that would be incurred by an efficient DNSP) vis á vis the other r 6.5.6(e) opex factors.

294    We turn now to consider each of the Tribunal’s underlying “concerns”.

295    We start with the use of the RIN data. The AER submitted that the Tribunal’s conclusion with respect to the RIN data which was that, at this point in its evaluation, the RIN data was not data upon which the AER might rely on to the extent that it did, could only be reached if the Tribunal found that the RIN data was unreliable for the use to which it was put. The AER submitted that the Tribunal made no such finding. The AER made the same submissions concerning the criticism that a significant proportion of the RIN data was estimated and “backcast”, as opposed to actual data.

296    The AER also submitted that the Tribunal did not address a number of the AER’s submissions and that included the following:

(1)    The fact of estimation does not establish unreliability;

(2)    Where the electricity network respondents provided estimates, they were generally not required to create a new series of data;

(3)    In most cases, the electricity network respondents produced estimates by drawing together actual information from a number of different sources;

(4)    The Tribunal ignored the fact that the AER made an allowance for any estimation errors in the RIN data by lowering the benchmark comparison point.

297    The AER further submitted that the result of the Tribunal’s decision is uncertain because it is uncertain whether the AER is entitled to use RIN data in undertaking modelling on the remittal.

298    In summary, the AER submitted that the Tribunal failed to address aspects of its case, failed to identify a relevant and reviewable error, and that its decision is uncertain.

299    It should be noted at this point that the AER accepts that, although Economic Insights attempted to estimate a simple opex cost function using the data for eight years for the 13 Australian National Electricity Market DNSPs, the parameter estimates it produced were relatively unstable and unreliable due to a lack of time series variation in the data.

300    In response, the electricity network respondents submitted that the Tribunal found that the different policies of electricity network respondents in classifying their opex and capex affects the AER’s calculation of the opex efficiency score and that the RIN data did not meet the necessary conditions for benchmarking, being that the data is long term reliable information, of high quality, consistent time series data and based on consistent definitions. The electricity network respondents submitted that the Tribunal found that a substantial portion of the data consisted merely of estimates by the DNSPs. The electricity network respondents submitted that the Tribunal found that the AER’s conclusion that the DNSPs’ chief executive officers had attested to the fact that the data was robust was incorrect. A key point made by the electricity network respondents was that the Tribunal found that the RIN data was unreliable for the use to which it was put by the AER and that, in that context, it was necessary to consider the Tribunal’s (later) conclusions about how the AER used the EI Model. As to the first three matters identified in [296] above, the electricity network respondents submitted that the Tribunal in fact gave careful consideration to the AER’s arguments before rejecting them. They further submitted that it should not be inferred that the Tribunal did not give due consideration to all of the AER’s submissions and that the Tribunal is not required to refer in its reasons to every piece of evidence and every contention made before it, let alone provide a rebuttal of it. As to the fourth matter, the electricity network respondents submitted that the Tribunal dealt later in its reasons with the lowering of the benchmark comparison point and the reasons the lowering of the benchmark comparison point did not make proper allowance for estimation errors. Finally, the electricity network respondents submitted that the Tribunal’s decision is not uncertain. The Tribunal is able to make a determination under s 71P(2)(c) of the NEL without telling the AER, in prescriptive terms, “exactly what to do and how to do it”.

301    In reply to the electricity network respondents’ submissions, the AER sought to characterise the issue as follows. The AER understood the characteristics of the RIN data and there was no error of fact at that point. The issue was whether it was suitable for use in the EI Model. In that respect, the AER relied on the advice of its expert, while the experts for the electricity network respondents held a contrary view. The AER submitted that that approach did not involve a question of fact or an incorrect exercise of discretion. Nor did it involve an unreasonable decision, having regard to the AER’s careful reliance on expert advice. In any event, the Tribunal did not make such a finding.

302    The submissions of the parties raise three issues. First, whether the Tribunal identified an error or even if it did, whether it was a reviewable error within s 71C(1) of the NEL. Secondly, whether the Tribunal erred in failing to consider contentions or submissions advanced by the AER. Thirdly, whether the Tribunal’s decision is uncertain.

303    We are of the view that the Tribunal reached the conclusion that the RIN data was not suitable to be used for the purpose to which it was put. Another way of expressing that conclusion, is that the RIN data was not sufficiently reliable to be used for the particular purpose for which it was used. That conclusion necessitates a consideration of the purpose for which the data was used. The Tribunal did deal with that topic and found that the RIN data was used in the EI Model and that the EI Model had been used as the sole determinant of opex. We will address the correctness of the latter finding in due course. As we have said, the context is that the AER accepted that the RIN data was not suitable for use by itself in the EI Model. Suitability for use for a particular purpose depends on the purpose and the soundness of other matters also used for the same purpose. In this case, those other matters include, for example, the overseas data. It is in this sense that the various “concerns” raised by the Tribunal relate to each other and are interconnected. It seems to us that the Tribunal stopped short of finding that the RIN data should not be used for any purpose in connection with forecast opex and again, we think that this emphasises the interconnectedness of the Tribunal’s conclusions and the importance of the purpose of the use and other matters to a finding of unsuitability or unreliability in relation to a particular matter, such as the use of the RIN data. We think a conclusion of unsuitability for a particular purpose is a finding of fact. In a sense, in view of the AER’s own view about the limitations of the RIN data, the dispute between the parties before the Tribunal with respect to that data was as to the extent of the limitations, rather than whether there were any limitations. We are of the opinion that the Tribunal was entitled to consider the evidence before it and assess the extent of the limitations of the RIN data. We are of the opinion that, having regard to the purpose for which the RIN data was put and the other matters that bear on that purpose (which we address below), the Tribunal was entitled to conclude that the AER made a material error of fact in deciding that the RIN data was suitable or sufficiently reliable to be used for the purpose for which it was put.

304    We do not think that the Tribunal erred by failing to consider relevant matters or submissions. We refer to our discussion above (at [278]). The Tribunal did not find that the fact of estimation established unreliability so it did not err in failing to address the contention that the fact of estimation does not establish unreliability. The Tribunal specifically mentioned, at [256], the AER’s contentions that where the electricity network respondents provided estimates, they were generally not required to create a new series of data and that in most cases, the electricity network respondents produced estimates by drawing together actual information from a number of different sources. The Tribunal was not bound to accept the contentions. The AER’s submission that the Tribunal failed to take into account that the AER had made an allowance for estimation errors in the RIN data by lowering the benchmark comparison point must be rejected because the Tribunal, at [311(d)], did refer to the approach taken by Economic Insights in its Second Report.

305    We reject the AER’s submission that the Tribunal’s decision is uncertain because it does not indicate whether the AER is entitled to use the RIN data in undertaking modelling on the remittal. It will become apparent from our reasons that we are of the opinion that the Tribunal identified adequately the reviewable errors made by the AER. We do not think it was incumbent on the Tribunal to indicate whether the RIN data should or should not be used. That will depend on a whole range of matters which it will be for the AER to consider on the remittal. The matter may be tested in this way. Had the Tribunal said that the RIN data was not to be used in any circumstances, then there might be a legitimate complaint that there was no basis for such a final and absolute conclusion. Had the Tribunal said that the RIN data might be used in certain circumstances, then the complaint might be that the circumstances are not identified. It cannot be incumbent on the Tribunal to identify the circumstances, even assuming that it could do that. Such an obligation would be tantamount to an obligation on the Tribunal to direct the AER as to the decision it is to make. There is no such obligation on the Tribunal.

306    We turn now to consider the use of the overseas data. The AER submitted that the Tribunal accepted three of the electricity network respondents’ criticisms of the use by the AER of the overseas data: the AER should have consulted more about the use of the overseas data; the country dummy variables did not correct for systematic reporting differences in the data; and the AER should have conducted more or better sensitivity testing of the overseas data.

307    With respect to the lack of consultation, while the Tribunal acknowledged that it was not a ground for review, it formed (so the AER submitted) one of the most significant bases for the Tribunal’s decision on opex. The AER submitted that it was illogical for the Tribunal to acknowledge that it was not a ground of review and, at the same time, to rely on it as a major reason for setting aside the AER’s decision with respect to forecast opex.

308    With respect to use of country dummy variables, the AER submitted that the Tribunal’s conclusion that the use of country dummy variables reveals a significant weakness in the model was not a conclusion that involved an error of fact. The issue was a methodological one and the AER relied on expert advice. Further, the AER submitted that it was not a matter expressly referred to by the Tribunal in reaching its conclusion.

309    With respect to sensitivity testing of the overseas data, the AER submitted the Tribunal found that the lack of such testing did not increase confidence in the AER’s reliance on the overseas data and that this finding did not involve an error of fact. Again, the issue was one of methodology. Further, the AER submitted that the Tribunal did not rely on this matter in reaching its final conclusions.

310    The AER submitted that the Tribunal’s decision is uncertain because it is uncertain whether the AER is permitted on remittal to place any weight on the EI Model.

311    In summary, the AER submitted that the Tribunal did not identify an error of fact, relied on a matter which was not a ground of review, and made a decision which is uncertain.

312    In response, the electricity network respondents made the following submissions. First, with respect to the lack of consultation, the Tribunal did not find that this was a reviewable error. The Tribunal found that the AER did not have the benefit of a proper and robust exchange of views with the electricity network respondents and their experts about the propriety of its chosen modelling technique prior to the application of that technique in a distribution determination making process of the kind contemplated by the NER. The electricity network respondents submitted that, in considering whether the AER erred in placing the reliance it did on the EI Model, the Tribunal was entitled to view the lack of consultation as a relevant matter, without it being an independent ground of review.

313    Secondly, with respect to the Tribunal’s approach to the use of the country dummy variables, the electricity network respondents submitted that the Tribunal had identified an error of fact. The AER found that the country dummy variables accommodate different operating environments across different countries in the sense of picking up differences in opex coverage, whereas the Tribunal found that a country dummy variable does not necessarily control for these within and across country differences, and a dummy variable only controls for level cost differences between datasets, not cost relationship differences.

314    Thirdly, with respect to the AER’s failure to conduct sensitivity testing, the electricity network respondents submitted that, like the failure to consult, the Tribunal did not identify this as a ground of review in itself, but rather a limitation in the EI Model which was relevant to the reliance which could be placed on it.

315    In reply to the electricity network respondents’ submissions, the AER addressed the lack of consultation and the use of the country dummy variables. As to the former, the AER submitted that the electricity network respondents were seeking to minimise the importance of the lack of consultation to the Tribunal’s decision. The Tribunal emphasised it in its conclusions at [308] on the AER’s use of overseas data and its overall conclusion on opex at [496]. As to the use of the country dummy variables, the AER submitted that the electricity network respondents’ submissions mischaracterised what the Tribunal had in fact done. The Tribunal had stated the view of the expert advising the AER correctly, that is, that a dummy variable controls for level differences between datasets, not differences in cost relationships. The real issue was whether the differences in cost relationships were likely to be material in connection with the use of the EI Model. EI advised the AER that the difference in cost relationships were unlikely to be material. The AER submitted that this was a matter of opinion not fact and furthermore, the Tribunal did not find that the AER’s decision was unreasonable.

316    The submissions of the parties raise three issues. First, whether the Tribunal erroneously treated a lack of consultation as a ground of review. Secondly, whether the Tribunal erroneously found a ground of review made out in relation to the AER’s application or use of country dummy variables. Thirdly, whether the Tribunal’s decision is uncertain because of its conclusion with respect to the overseas data.

317    The Tribunal was clearly aware that a lack of consultation or a denial of procedural fairness was not a ground of review. It made that point on a number of occasions throughout its reasons: at [288], [308] (overseas data), [325] (OEFs), and [376] (vegetation management costs). It is true that in expressing its conclusions with respect to opex, the Tribunal referred to the restricted opportunity afforded to the parties to test the veracity of the EI Model as one of the two relevant matters.

318    We do not think that the Tribunal viewed the lack of consultation as a denial of procedural fairness and, therefore, a ground of review. We think that what the Tribunal was saying was that the limited opportunity or lack of opportunity to have input into the methodology and resulting estimates of the electricity network respondents’ required opex was relevant to the reliability or soundness of the EI Model. We are of the opinion that the Tribunal’s conclusion must be read in that light. We think that the Tribunal was entitled to take that approach. Common sense suggests that the more data is scrutinised and examined, the more likely flaws and limitations in the data will be exposed and the Tribunal had expert evidence to that effect. Furthermore, there was evidence before the Tribunal of overseas regulatory authorities, such as Ofgem engaging with Distribution Network operators about proposed models with a view to identifying matters that might call for adjustments. The extent to which consultation had or had not taken place in relation to the EI Model was not in dispute. It was not an error for the Tribunal to characterise that and then take it into account as relevant to its evaluation. The Tribunal took a similar approach in the case of Endeavour’s increased vegetation management costs and the 2015 Deloitte Report.

319    With respect to the use of country dummy variables, the argument before the Tribunal seemed to proceed in the following way. First, there was evidence that the overseas entities adopted different definitions for the purpose of collecting data. Secondly, the difficulty created by that was overcome by the use of country dummy variables which are designed to correct for systematic reporting differences. Thirdly, the use of country dummy variables account for level differences, but not for cost relationship differences. Finally, accepting that, the AER contended by reference to the EI Report that the differences would not have a material impact on the model results.

320    We are of the opinion that the Tribunal found that the AER made a material error of fact in finding that the country dummy variables accounted for systematic reporting differences or cost relationship differences or (and this is more likely) that the country dummy variables accounted for level differences and that systematic reporting differences would not have a material impact on the model results. We think that that conclusion is a finding of fact and that its character is not altered by the circumstance that it is informed by expert opinion.

321    With respect to the AER’s contention that the Tribunal’s decision is uncertain in relation to the overseas data, we reject this contention for the same reasons we have already given with respect to a similar contention in relation to the RIN data.

322    We turn to consider the AER’s decision to lower the benchmark comparison point. The AER submitted that the Tribunal erred in finding (insofar as it did find) that its decision to lower the benchmark comparison point involved reviewable error. The first submission seemed to be that it was not open to the Tribunal to criticise the AER in relation to its use of the RIN data and at the same time criticise it for lowering the benchmark comparison point to allow for any imperfections in the data. This approach was said by the AER to be inconsistent. Furthermore, the Tribunal failed to understand that the AER was engaged in a task involving estimation and that necessarily involves making approximations and allowances. Such an exercise could not give rise to errors of fact. The second submission was that, in any event, the Tribunal did not make findings of reviewable error. It referred to submissions as having merit or being cogent, but it did not make firm findings. The AER also submitted that the Tribunal’s decision is uncertain for similar reasons as those advanced in connection with the RIN data and the overseas data.

323    In response, the electricity network respondents submitted that the Tribunal had accepted ActewAGL’s submission that the choice between the benchmark comparison point made by the AER in its Draft Decisions and the benchmark comparison point in its Final Decisions was arbitrary and that the fact that such an arbitrary choice was necessary because of deficiencies in the EI Model suggest that no reliance could be placed on the results of the model whatsoever. The electricity network respondents submitted that the Tribunal’s conclusions about the benchmark comparison point were part of its reasoning that led to its ultimate conclusion that the AER had placed too much weight on the results of the EI Model and the AER’s ad hoc post modelling adjustments. The electricity network respondents submitted that the Tribunal found that the AER made an error of fact in concluding that its approach in lowering the benchmark was consistent with the approach taken by Ofgem. The electricity network respondents submitted that there was no inconsistency in the Tribunal finding that data used in the EI Model was unreliable and the lowering of the benchmark comparison point was arbitrary. Finally, the electricity network respondents submitted that the Tribunal’s decision with respect to the benchmark comparison point was not uncertain for the same reasons it gave in relation to the same submission with respect to the Tribunal’s decision concerning the RIN data.

324    In reply to the electricity network respondents’ submissions, the AER submitted that it did not make a finding of fact that its approach in lowering the benchmark comparison point was consistent with Ofgem’s approach or, if it did, that finding was not a material finding.

325    We see no inconsistency in the Tribunal’s approach. If the model and the data used in the model had imperfections and deficiencies, then the Tribunal was entitled to so conclude. In fact, this circumstance appears not to have been in dispute. In its Final Decision in relation to Ausgrid, the AER said that it had decided to reduce the benchmark comparison point to increase the margin for error “for modelling and data issues” and to incorporate an appropriately wide margin for “potential modelling and data errors and other uncertainties”. It is not inconsistent with that conclusion to find fault with the process the AER adopted to try and correct the results for potential errors.

326    We think the Tribunal did make an affirmative finding of material error in connection with the lowering of the benchmark comparison point. PIAC advanced five matters in support of its contention that the AER’s reasons for reducing the benchmark comparison point had no proper basis. We do not need to outline all of them. Two will suffice. PIAC submitted that the Economic Insights’ decision to argue for a lowering of the benchmark comparison point was informed by no more than the adoption of an even more conservative benchmark than the adjustment it described as “conservative” and “generous” at the Draft Decision stage. PIAC submitted that the AER erred in concluding that its approach was consistent with that taken by Ofgem. At one point, at [311], the Tribunal described PIAC’s submissions as having merit. In its conclusion, at [316], the Tribunal said that PIAC’s submissions were cogent. We think that that is a conclusion by the Tribunal that the reasons advanced by the AER for lowering the benchmark comparison point cannot be sustained. The Tribunal also described the submissions of ActewAGL as cogent and those submissions include reference to the deficiencies in the model and a submission that the reduction in the benchmark comparison point was arbitrary. Again, we think that they are conclusions of the Tribunal and we reject the submission that the Tribunal did not make findings. The AER’s choice of the benchmark comparison point was a finding of fact and the Tribunal’s conclusion that it was arbitrary and without a proper basis demonstrates that the Tribunal considered it was an error of fact. We do not think it could be suggested that it was not material. In those circumstances, we do not need to consider the electricity network respondents’ submission that the AER’s conclusion that its approach was consistent with that taken by Ofgem was also an error of fact, other than to note that there is force in the submission. We reject the submission that the Tribunal failed to understand that the process the AER was engaged in involved estimation. The Tribunal did understand that. Furthermore, we reject the further submission made by the AER that there cannot be an error of fact in an estimation. The fact is that the AER’s estimate was the result of a complicated process involving a number of steps and there is no difficulty in concluding that one or more of these steps involved an error of fact. Finally, we reject the submission that the Tribunal’s decision is uncertain for the same reasons we rejected a similar submission in relation to the RIN data.

327    We turn to consider the adjustments made for the OEFs. The AER submitted that the Tribunal erred in its approach to the AER’s adjustments to the results of the EI Model to account for differences between electricity network respondents’ opex efficiency that were attributable to OEFs that were not included as variables in the EI Model. It submitted that the Tribunal expressed no overall conclusion with respect to the OEFs, but that it appeared to make three findings.

328    First, the Tribunal found that the AER’s approach to OEF adjustments was, in certain respects, arbitrary. However, the AER submitted that the Tribunal made no express finding in those terms. It did not explain why a 0.5% materiality threshold was unreasonable or what it would regard as a reasonable threshold. Furthermore, the AER’s approach could not be characterised as an error of fact.

329    Secondly, the Tribunal found that the adjustments to the data should be made before modelling by normalising the data, rather than after modelling. However, the Tribunal did not address the AER’s reasons in support of its methodology, namely, that the output specification for the model inherently accounts for some of the most important OEFs; it is not possible to adjust the data for an OEF prior to modelling unless data in respect of the particular OEF was available for every DNSP and it was not available; other regulators, such as Norway’s NVE, adopt a similar approach to the AER; and the only alternative was to leave out most of the OEFs which was what the electricity network respondents’ consultants had done and which would have disadvantaged the electricity network respondents. In summary, the Tribunal did not support its conclusion with any technical analysis and it did not explain why the opinion of AER’s consultant, Economic Insights, was incorrect. Furthermore, the AER’s approach cannot be characterised as an error of fact.

330    Thirdly, the Tribunal found that there was merit in the criticisms made of immaterial and directionally ambiguous OEFs. However, the Tribunal did not identify an error in the AER’s approach. Having criticised the AER for making adjustments for immaterial OEFs equal to the outer bounds of 0.5%, the Tribunal did not make a finding about how the AER should have quantified adjustments for immaterial OEFs. In relation to the AER’s approach to directionally ambiguous OEFs, the Tribunal went no further than saying that the AER’s approach acknowledged the immaturity of the data and that this should have put the AER on notice of the consequences of relying on the data to the extent that it did.

331    In summary, the AER submitted that the Tribunal erred in finding a reviewable error. The Tribunal failed to consider a relevant matter, namely, that it was not possible to normalise the data for OEFs before modelling. The Tribunal’s decision is uncertain because it is uncertain how the AER should use the EI Model on remittal when it is not possible to normalise the data for OEFs before modelling.

332    In response, the electricity network respondents pointed to the significance of the OEF adjustments. As will be seen from the Tables set out above (at [202]), in the case of ActewAGL, the AER made adjustments for four “material” OEFs and for the collective effect of 20 immaterial OEFs resulting in a reduction in the benchmark efficiency score applied to ActewAGL by 23%. In the case of the NSW service providers, the benchmark efficiency scores applied to Ausgrid, Endeavour and Essential were reduced by 11.7%, 12.9% and 10.7% respectively. The electricity network respondents submitted that the Tribunal found that the AER had erred in three respects.

333    First, the Tribunal in substance found that the OEF adjustments made by the AER were arbitrary. The conclusion that the materiality threshold of 0.5% was arbitrary did not require a prior finding that it was too high or too low by a particular amount. The electricity network respondents submitted that the assertion that the AER’s approach did not involve an error of fact, although correct, was not to the point. The conclusion that the OEF adjustments were arbitrary reinforced its finding that the AER was wrong to consider that the RIN data was sufficiently reliable for the purpose to which it was put.

334    Secondly, the Tribunal found that the costs relationships within the model had been skewed by the inclusion of non-comparable data and that this could not be corrected by post modelling adjustments to only one of the electricity network respondents in the data set. The AER’s attack on the Tribunal’s conclusion in this respect is no more than raising a matter which went to the merits. The Tribunal was aware that it was not possible to normalise the data and did not find to the contrary. The point the Tribunal made was that that was not a reason to ignore the deficiency. The deficiency was a practical limitation in terms of the EI Model and it diminished the probative value of the model.

335    Thirdly, electricity network respondents submitted that the Tribunal did not err in its treatment of directionally ambiguous OEFs. The AER’s complaint that the Tribunal did not identify an error of fact misses the point which is that the Tribunal’s conclusions about directionally ambiguous OEFs was part of a number of concerns held by the Tribunal about the AER’s modelling methodology that led the Tribunal to the overall conclusion that the AER had placed too much weight on the outcome of the EI Model and, therefore, one or more of the grounds of review advanced by the electricity network respondents was made out.

336    In reply, the AER submitted that the Tribunal did not find that the OEF adjustments were arbitrary. It did not find that the materiality threshold adopted by the AER of 0.5% was unreasonable. The AER submitted that the Tribunal did not find an error within s 71C(1) and there was no such error in the decision of the AER to accept the advice of its own expert.

337    In our opinion, the Tribunal was entitled to proceed on the basis that the EI Model contained potential modelling and data errors and uncertainties before the benchmark comparison point was lowered and the adjustments for OEFs were made. As we have said, that was indeed the view of the AER and its adviser, Economic Insights. The Tribunal found, without legal error as we have said, that the RIN data was not suitable for the purpose for which it was put and that there were limitations associated with the overseas data. The estimate made by the AER was not made in one step, but by a series of steps which started with the use of the RIN data and concluded with the OEF adjustments. Each step potentially related to other steps whether they were earlier or later steps. For example, if there were errors associated with the use of the RIN data, later steps may reveal whether those errors were material or were in some way cured by later steps. Furthermore, a later step, and the need for it, may reveal errors in earlier steps. It seems to us that it is not simply a matter of considering the final result and finding no reviewable error because the final step was supported by expert evidence. Nor is it a matter of ignoring the interim steps because they do not involve reviewable error if the interim steps throw light on whether there was reviewable error at an earlier or later stage or on the materiality of those errors. The Tribunal took that approach, which we think is without legal error, and its examination of the OEFs was part of that approach.

338    As we said earlier, the Tribunal did not always state in express terms that it was making a particular finding as one might expect from a court. On occasions, it commented on the correctness or otherwise of a submission or submissions made by a party. It did that in the case of its treatment of the AER’s adjustment for OEFs. The Tribunal, at [331]-[332], found that the allowance for the collective influence of immaterial variables was a subjective estimate and that there was no detailed analysis of materiality with respect to these variables. The Tribunal, at [335], found that the failure to normalise the data set to deal with the fact that the costs were affected by non-comparable data and, therefore, were skewed by heterogeneous differences between the electricity network respondents were not corrected by post-modelling OEF adjustments made to some only of the electricity network respondents. It is to be noted that the AER’s response to that conclusion was that it was not possible to normalise the data in the manner suggested. The Tribunal, at [336], acknowledged that response, but we think reading its reasons as a whole, the Tribunal decided that if that was the case it had the consequence of diminishing the probative value of the EI Model. Finally, the Tribunal concluded from the AER’s approach to the directionally ambiguous OEFs that the AER’s approach was evidence of the immaturity of the data. Although the AER did not characterise its approach in that way, that conclusion standing alone was one that was difficult to dispute.

339    The Tribunal’s decision is not uncertain. We reject the AER’s submission for the same reasons we have given in relation to the same submission in relation to the RIN data, the overseas data and the lowering of the benchmark comparison point.

340    We should say that with respect to the two topics which we have just addressed (i.e., the AER’s decision to lower the benchmark comparison point and the adjustments made for the OEFs) PIAC made submissions to the effect that the AER had not characterised correctly its own reasoning and that of the Tribunal. In addition, it sought to refute the AER’s submissions.

341    As we have noted at [9] above, we gave PIAC leave to intervene shortly prior to the hearing of these applications. We doubt whether some of the submissions justify the order for intervention, but because of the conclusions we have reached, we do not need to address those submissions.

342    We turn now to consider the Tribunal’s approach to the electricity network respondents’ vegetation management costs. The Tribunal stated that each of Endeavour, Essential and ActewAGL challenged the AER’s assessment that the vegetation management costs in their respective 2012-13 base year opex were not efficient. The Tribunal addressed the challenges. The AER submitted that the Tribunal erred in the course of doing so because it criticised the AER’s approach to vegetation management costs and its criticisms formed one of the grounds upon which the Tribunal ultimately concluded that the AER had committed reviewable errors in relation to its estimate of forecast opex. The AER submitted that when the Tribunal’s reasons in relation to vegetation management costs are properly analysed, it is clear that the Tribunal did not find any errors or, if it did, those errors were not reviewable errors within s 71C(1) of the NEL.

343    The response of the relevant electricity network respondents to these submissions differed from their response to similar submissions made by the AER in relation to other items of forecast opex. The electricity network respondents submitted that whatever conclusions the Tribunal drew with respect to vegetation management costs, those conclusions did not form one of the grounds for the Tribunal’s ultimate conclusion that the AER’s estimate of forecast opex involved reviewable error.

344    We think that the submission made by the electricity network respondents is correct when the Ausgrid reasons are read together with the separate reasons in the case of each of Endeavour, Essential and ActewAGL.

345    We start with Endeavour’s vegetation management costs and the Ausgrid reasons. We have already referred to those reasons, but we will recapitulate briefly. The Tribunal found, at [373], that the AER’s reasons for rejecting Endeavour’s vegetation management opex were “tenuous”, having regard to the fact that Endeavour had provided its internal analysis to the AER, and the fact that Endeavour was committed to paying contractors retained through a competitive tender process so that it may comply with an applicable regulatory obligation. The other matter identified by the Tribunal, at [378], was that the NSW service providers had not been given the opportunity to be heard in relation to the 2015 Deloitte Labour Report and, although a failure to accord a hearing was not a ground of review, it was a matter of common sense (according to the Tribunal) that a report such as the 2015 Deloitte Labour Report might carry greater weight if it had been the subject of a response from Endeavour, depending on the terms of that response. In its conclusions in the Ausgrid reasons dealing with forecast opex, the Tribunal identified as one of the issues with the EI Model and the AER’s application of the EI Model, the fact that the AER relied on a qualitative analysis rather than a bottom-up quantitative assessment to test issues such as the issues raised by the electricity network respondents in relation to their vegetation management opex (see above at [261]). Later in the conclusions, the Tribunal referred to the AER’s treatment of the vegetation management costs as one of a series of concerns it had. Those conclusions in the Ausgrid reasons perhaps leave the matter unclear.

346    However, the matter is made clear by the Tribunal in the Endeavour separate reasons. In the Endeavour separate reasons, the Tribunal noted, at [24], that in the AER’s Final Decision with respect to opex, the AER found that Endeavour was not materially inefficient. However, the AER did not make an allowance for the labour and vegetation management costs that Endeavour would incur in the 2014-2019 regulatory control period. The Tribunal noted that the common contentions of the NSW service providers, which of course included Endeavour, regarding the opex allowance in each of the NSW service providers’ applications relating to the selection and use of data, the model construction, and the use of the EI Model and the reflection of the OEFs, had been made out. The Tribunal noted that despite the flaws in the EI Model as articulated in the common contentions, the AER found that Endeavour’s opex was not materially inefficient on the basis of the EI Model. However, the AER’s decision concerning Endeavour’s efficiency was that, although not materially inefficient, that was only marginally so, and its conclusion would not have been the same had the additional expenditure been included in Endeavour’s base year opex. The flaws in the EI Model as articulated in the common contentions, affected Endeavour because they affected the starting point from which the additional expenditure is considered. The Tribunal said, at [35], that it was not appropriate to order that the AER simply reconsider the disputed costs because compliance with the Tribunal’s orders (as finally made) might affect the starting point and the AER might “revisit the VM expenditure and redundancy expenditure claimed in the more general context of the opex expenditure claimed”. The Tribunal made it clear in the Endeavour separate reasons, at [47], that it was not making a decision about Endeavour’s challenge to the AER’s decision with respect to its claim for vegetation management costs and redundancy costs.

347    We are of the opinion that a proper analysis of the Ausgrid reasons and the Endeavour separate reasons leads to the following conclusions. First, although expressing some criticisms of the AER’s approach to Endeavour’s claim for increased vegetation management costs and redundancy costs, the Tribunal did not make a final decision about the AER’s decision with respect to that claim. Indeed, the Tribunal was of the view that it would be undesirable for it to do so. Secondly, other than noting that a bottom-up analysis would enable the AER to test issues such as those raised by the electricity network respondents in relation to their vegetation management opex (a point the Tribunal also made when it referred to the dictates of sensible administration in the context of the vegetation management costs of Essential) (at [389]) and ActewAGL (at [408]), the Tribunal did not base its conclusions in terms of reviewable errors on any findings or criticisms it made with respect to Endeavour’s vegetation management costs and redundancy costs.

348    It follows that the AER’s submissions that the Tribunal did not find error, let alone reviewable error, do not advance its applications under the ADJR Act.

349    The AER also submitted that the Tribunal’s decision is uncertain because it is uncertain what approach the AER should take to Endeavour’s proposed step change increase in vegetation management costs on the remittal. This submission stands or falls with the more general submission that the Tribunal’s orders and directions, insofar as they refer to a broader range of modelling and benchmarking against Australian businesses and including a “bottom-up” review of Endeavour’s forecast operating expenditure, are uncertain. We will consider it in that context because Endeavour’s vegetation management costs and redundancy costs are encompassed by those matters: see the Endeavour separate reasons at [76].

350    We are of the opinion that precisely the same conclusions as we have reached in the case of Endeavour’s vegetation management costs and redundancy costs apply in the case of Essential’s vegetation management costs. We acknowledge that the position does not emerge as clearly from the Essential separate reasons as it does from the Endeavour separate reasons. Nevertheless, we think the same conclusions apply for the following reasons. First, even accepting that the Tribunal did not, when making findings, always express itself as clearly as a court might, the Tribunal, in the context of Essential’s vegetation management costs, did not state that it accepted Essential’s challenge to the AER’s approach, but rather, at [390], said no more than that it was unconvinced by the AER’s response to some of the criticisms. Secondly, if the Tribunal had reached firm conclusions that there were errors, then one would have expected it to say so in the Essential separate reasons. It did not do that. Thirdly, what the Tribunal did say, at [28], in the Essential separate reasons is that the common contentions of the NSW service providers were made out. Finally, it is clear from the ActewAGL separate reasons that the Tribunal did not reach a final conclusion with respect to ActewAGL’s vegetation management costs. There is no reason to think that the Tribunal took a different approach in the case of the vegetation management costs of Essential than it did in the case of the vegetation management costs of Endeavour and ActewAGL respectively.

351    The same conclusions apply in the case of the Tribunal’s approach to ActewAGL’s vegetation management costs. As we have said, that is made clear in the ActewAGL separate reasons. In those reasons, the Tribunal said, at [30], that the NSW service providers had made out their criticisms of the AER’s benchmarking methodology and, therefore, their grounds of review, and ActewAGL’s contentions in large measure followed the NSW service providers contentions and, at [35], that it was not necessary for it to determine “specifically the correctness of the AER’s assessment that ActewAGL’s vegetation management costs are inefficient”.

352    In Ground 7 of its applications for judicial review, the AER claims that the Tribunal erred because it did not consider or address significant aspects of the AER’s decision. We have already considered and rejected the AER’s submissions (at [304], [325] and [338] above).

353    In Ground 9 of its applications for judicial review, the AER claims that the Tribunal erred because it took into account an impermissible consideration, being the AER’s alleged failure to give the relevant DNSP an adequate opportunity to make submissions with respect to overseas data. Again, we have already considered and rejected this submission in relation to the overseas data (at [318] above).

354    In Ground 21 of its applications for judicial review, the AER claims that the Tribunal erred in its approach to the grounds for review in s 71C(1) of the NER and to the grounds in respect of which it granted leave. The AER pointed to various sections in the NER which emphasise the centrality of the grounds for review to the Tribunal’s review function. We do not propose to repeat those submissions. It is enough to say that at a general level those submissions may be accepted. The AER criticises the Tribunal for not setting out the grounds for review and dealing with them seriatim. It criticises the Tribunal for proceeding as if it was conducting a de novo hearing on the merits, for not clearly identifying when it was making findings as distinct from observations, and for not identifying or clearly identifying the reviewable errors it found. We have concluded that the Tribunal did not err and we do not propose to debate how the Tribunal might have better structured its reasons or expressed itself. The question for this Court is whether one or more of the grounds of review in s 5 of the ADJR Act have been established. In relation to forecast opex, we have concluded that none of the grounds advanced by the AER have been established.

355    It is convenient to add at this point that we reject this complaint on the same grounds insofar as it is made in relation to the other matters addressed by the Tribunal and challenged by the AER in this Court.

The Tribunal’s approach to the Labour Costs of the Electricity Network Respondents (Ground 8)

356    The AER submitted that the Tribunal erred in concluding that it is the policy of the legislative arm of government that, to the extent that the EBAs are (if they are) an inefficient imposition on the electricity network respondents, nevertheless they are a cost to be borne by the consumers of electricity. The AER submitted that, having said that, the Tribunal then immediately contradicted itself by saying that it was open to the AER to assess the extent of inefficiency reflected by the number of employees, and that it may review and consider the terms upon which the number of employees may be reduced under the EBAs and the timing for the exploration of the EBAs.

357    The AER submitted that the Tribunal “side-stepped” the argument advanced by the electricity network respondents to the effect that the EBAs were a “regulatory obligation or requirement” within s 2D of the NEL. The AER submitted that the Tribunal’s reference to rr 6.5.6(a)(4) and 6.5.6(c)(3) (a realistic expectation) of the NER was “question begging” and that if, by its reference to those rules, the Tribunal was suggesting that the operating expenditure criteria in r 6.5.6(c)(1) (efficient costs) (2) (a prudent operator) were irrelevant then it was in error.

358    The AER submitted that the Tribunal, at [344], erred in concluding that the EBAs were not an endogenous managerial choice, at least in circumstances where the AER “has quite radically shifted from an itemised bottom-up approach to assessing opex to benchmarking total opex per se – particularly where that benchmarking has not been exposed to the rigours of the consultation the NEL and NER envisaged for such a radical change”. The AER submitted that the linking of these two propositions lacked a logical basis because whether the EBAs were an endogenous managerial choice could not be effected by the number of times the AER had adopted a benchmarking approach.

359    The AER submitted that whilst the Fair Work Act is an exogenous matter, the EBAs are not.

360    The AER submitted that the Tribunal erred “if and so far” as it concluded that the electricity network respondents’ EBAs are “applicable regulatory obligations or requirements associated with the provision of standard control services” within r 6.5.6(a)(2) and, therefore, an operating expenditure objective. In support of this contention, the AER submitted that neither the Fair Work Act nor the EBAs are an Act or instrument that materially affects the provision of electricity network services within s 2D(1)(b)(iv) of the NEL because the electricity network respondents are not required to engage labour pursuant to the Fair Work Act or the EBAs. Further, or in the alternative, neither the Fair Work Act nor the EBAs is a regulatory obligation or requirement “associated with the provision of standard control services” within r 6.5.6(a)(2) as both are independent of the provision of distribution services. Both the Fair Work Act and such agreements operate in a much wider field.

361    The AER summarised its submissions as follows. First, the Tribunal misconstrued the phrase “applicable regulatory obligations or requirements associated with the provision of standard control services”. Secondly, the Tribunal misconstrued the meaning of “error of fact” and failed to identify an error of fact as that term is properly construed. Finally, the Tribunal’s decision is uncertain because it is uncertain what approach the AER should take to the electricity network respondents’ labour costs on the remittal.

362    In response, the electricity network respondents submitted that the Tribunal did not decide whether the EBAs were regulatory obligations or requirements associated with the provision of standard control services within r 6.5.6(a)(2). The Tribunal found that the EBAs may reasonably be regarded as otherwise required to achieve the opex objective of maintaining the safety of the distribution system through the supply of standard control services, and as reasonably reflecting the opex criteria, including a realistic expectation of the demand forecast and cost inputs required to achieve the opex objectives. The electricity network respondents pointed out that before the Tribunal, the AER accepted that the opex criteria included a realistic expectation of cost inputs and that labour costs are one such input. However, the Tribunal identified the AER’s error as treating the EBAs as an endogenous factor to be ignored in the AER’s estimate under r 6.12.1(4)(ii). That error came about because of the AER’s focus on benchmarking and, in particular, the EI Model and its total opex outcomes.

363    The electricity network respondents submitted that the Tribunal found that the AER’s conclusion that the EBAs were an endogenous factor was wrong. They submitted that the Tribunal’s conclusion in this respect was related to the particular circumstances of the case, and that those circumstances were not in dispute. The NSW service providers high number of employees was a result of changes to Ministerial licence conditions in 2005 and 2007 which placed considerable pressure on the NSW service providers during the 2009-14 regulatory period. The 2015 Deloitte Report expressed the view that in response, the service providers acted in a manner consistent with a “prudent and efficient DNSP” by aiming to be largely compliant with the Ministerial conditions by 2014. In the same report, the restrictions which are associated with the EBAs were noted.

364    The electricity network respondents submitted that the Tribunal did not suggest that the AER could not consider inefficiencies. The Tribunal said that the AER could assess the extent of any efficiencies reflected by the number of employees and it could review and consider the terms upon which the number of employees may be reduced under the EBAs, and the timing for the expiration of the EBAs.

365    The electricity network respondents addressed the three conclusions advanced by the AER. First, they submitted that the Tribunal did not, in fact, decide that the EBAs were a regulatory obligation or requirement associated with the provision of standard control services. Secondly, they submitted that the Tribunal found that the AER erred in considering that the EBAs were an endogenous factor to be ignored when making an assessment of the total required opex under r 6.12.1(4)(ii) and that that error was properly characterised as an error of fact. In the alternative, the erroneous characterisation of the EBAs led to an incorrect exercise of discretion. The electricity network respondents submitted that the Tribunal’s focus on benchmarking and, in particular, the EI Model, led it to treat the EBAs as endogenous rather than exogenous. Finally, the electricity network respondents submitted that the Tribunal’s decision is not uncertain. As submitted in the context of a similar submission first made by the AER in the context of the RIN data, the electricity network respondents submitted that the Tribunal was not required to tell the AER precisely how it was to perform its task on the remittal.

366    In reply to the electricity network respondents’ submissions, the AER accepted that it was not entirely clear whether the Tribunal found that the EBAs were a regulatory obligation or requirement. Nevertheless, the Tribunal erred in concluding that the EBAs, even if they are an inefficient imposition on the electricity network respondents, are a cost to be borne by the consumers of electricity. The AER submitted that that is clearly an error because there is nothing in the NER which suggests that the costs associated with the EBAs are not to be assessed in the same way as other costs. Furthermore, the AER reiterated its earlier point that it was illogical to link the finding that the EBAs are not an endogenous managerial choice with the fact that the AER had adopted a benchmarking method of estimating the required opex.

367    A number of the AER’s submissions were directed to whether the EBAs were a regulatory obligation or requirement within s 2D of the NEL and an applicable regulatory obligation or requirement associated with the provision of standard control services within r 6.5.6(a)(2) of the NER and the first question is whether the Tribunal held that they were properly so characterised. We do not think the Tribunal drew such a conclusion. It decided that the matter could be approached in another way and that was by reference to the operating expenditure objective in r 6.5.6(a)(4) and the operating expenditure criteria in r 6.5.6(c). The approach taken by the Tribunal was open to it and to proceed in that way was not an error.

368    Two interrelated matters loomed large in the Tribunal’s consideration of the matter and the AER’s approach. The first was the AER’s reliance on the EI Model, and the second was the AER’s conclusion that the EBAs were an endogenous managerial choice. These two matters are interrelated because the differences between the NSW service providers and the frontier DNSPs identified in the AER’s economic benchmarking analysis (in particular, the EI Model and its total opex outcomes), were attributed to the NSW service providers’ labour management practices and that led the AER to treat the EBAs as endogenous and irrelevant to the AER’s estimate of the total required opex made pursuant to r 6.12.1(4)(ii). That this was the Tribunal’s view of the AER’s approach appears from the following passage in the Tribunal’s reasons (at [423]):

Notwithstanding the AER’s statement that: “Recruitment and removal of staff are both ‘legitimate costs’ that the service providers would need to incur.” and its submission that labour costs are relevant to its assessment of required opex, its focus on benchmarking (in particular the EI model and its total opex outcomes) have lead it to treat the EBAs as endogenous (rather than exogenous) – an endogenous factor to be ignored in the AER’s estimate of the total required opex made pursuant to r 6.12.1(4)(ii).

369    The Tribunal took the view that this approach was erroneous. The benchmarking and the EI Model, in particular, was not sufficiently reliable to enable a conclusion to be drawn that NSW service providers’ labour management practices were inefficient and that that was attributable, or partly attributable, to the EBAs which were an endogenous managerial choice. That conclusion was open to the Tribunal, having regard to its earlier conclusions about the EI Model which we have concluded were not in error. Furthermore, it explains the Tribunal’s linking of the novelty of benchmarking and flaws in the EI Model on the one hand, with conclusions about the efficiencies or otherwise of the NSW service providers’ labour management practices on the other. We reject the submission (summarised, for example, at [358] above) that the Tribunal’s approach was illogical.

370    The Tribunal went on to consider the AER’s conclusion that the EBAs were endogenous and, therefore, to be ignored. The Tribunal rejected an approach that simply characterised an obligation as endogenous and to be ignored or as exogenous and to be considered, and said that a closer analysis was required. We think this was what the Tribunal was saying when it referred to the pressure placed on the NSW service providers by the Ministerial licence conditions and their response to it and, that that having happened, to the Fair Work Act being an exogenous factor. Leaving aside precisely what is meant by an endogenous matter and exogenous matter, we do not think that the Tribunal erred in finding error in this approach of the AER. The distinction might be a useful one, but it should not be used in a way which precludes an examination of all the facts and circumstances.

371    We have reached the conclusion that as to the two interrelated matters which we have identified (i.e., the AER’s reliance on the EI Model and the AER’s conclusion that the EBAs were an endogenous managerial choice), it was open to the Tribunal to conclude that the AER committed material errors of fact or (our preferred characterisation) incorrectly exercised its discretion. We would add that it may be considered that standing alone, the Tribunal’s statement that it is the policy of the legislative arm of government that, to the extent that the EBAs are (if they are) an inefficient imposition on the electricity network respondents, nevertheless they are a cost to be borne by consumers of electricity (at [436]) is incorrect However, the statement is not to be construed as standing alone and the Tribunal went on to explain, not contradict as the AER submitted, what it meant by this statement. Even if it is an overstatement, it does not affect the Tribunal’s earlier conclusions with respect to the labour costs.

372    Finally, we reject the submission that the Tribunal’s decision is uncertain in relation to the labour costs of the electricity network respondents. For the reasons we have given in relation to particular matters, and will give in relation to the Tribunal’s orders generally, we do not think the Tribunal’s decision is uncertain.

Whether the Tribunal’s reasons and orders are uncertain and unreasonable (Grounds 10 and 11)

373    As we have said, the Tribunal made the following direction with respect to opex in Ausgrid:

(a)    the AER is to make the constituent decision on opex under r 6.12.1(4) of the National Electricity Rules in accordance with these reasons for decision including assessing whether the forecast opex proposed by the applicant reasonably reflects each of the operating expenditure criteria in r 6.5.6(c) of the National Electricity Rules including using a broader range of modelling, and benchmarking against Australian businesses, and including a “bottom up” review of Ausgrid’s forecast operating expenditure;

374    The Tribunal made the same direction in the matters of Endeavour, Essential, and ActewAGL.

375    The AER submitted that this direction is legally uncertain and unreasonable. In support of this submission, it made the following submissions.

376    First, the AER submitted that the requirement to conduct a “bottom up” review is uncertain in its terms and it lacks a legal and evidentiary justification.

377    We start with uncertainty. The AER submitted that the phrase, “a bottom up review” has no one meaning. It is generally used in contra-distinction to the phrase, “top down review”. The AER submitted that in the context of assessing a forecast opex, a top down review examines the forecast opex for efficiency without disaggregating that total into its various components, by comparing one business against comparable businesses, adjusted for any differences, whereas a bottom-up review examines the particular features of a business to examine whether it is operating efficiently. The scope of a bottom-up review is critical and an itemised bottom-up review could involve an examination of the efficiency of every work practice and the necessity and quantum of every expense.

378    The Court notes that s 71P(2) of the NEL provides relevantly that the Tribunal’s determination may set aside the reviewable regulatory decision and “remit the matter back to the AER to make the decision again in accordance with any direction or recommendation of the Tribunal”. We see nothing in this paragraph which requires the Tribunal to direct the AER in precise terms as to how it is to go about making the decision again. In our view, that is not surprising. The Tribunal may detect error in the AER’s approach, but there may be a number of other approaches which do not involve error. It seems to us that it is unlikely that the Tribunal (in many cases not fully informed) would be required to direct the AER to take a particular approach. We agree with the submission of the electricity network respondents that it is for the AER to determine the nature and scope of the bottom-up review provided that it otherwise complies with the NEL and the NER and makes it decision consistently with the criticisms in the Tribunal’s reasons about its original methodology.

379    We do not understand the AER to submit that it does not know what a bottom-up review is or that it could not conduct a bottom-up review. It could not say that in view of the following matters. First, in the course of the hearing before the Tribunal, senior counsel for the AER said:

Mr Moore made a submission to the tribunal that benchmarking has many meanings. Undoubtedly that can be right stated generally, and he gave an example of looking at whether a DNSP uses four people to replace a pole or uses three people to replace a pole, and you can look at that and you can compare different DNSPs and that’s an example of benchmarking.

Leaving aside whether that’s really opex or capex, the illustration still holds for present purposes and that’s certainly true and that can be done; but essentially what he is describing through the use of that example, is the traditional form of benchmarking that has been undertaken by the AER up until the present, which is often described as a bottom-up or engineering approach to benchmarking, and that approach is to go to very specific line items or very specific practices at quite a micro level and do that sort of examination that Mr Moore gave the illustration of, and look at it in that way.

380    We also note that the AER’s own Better Regulation Explanatory Expenditure Forecast Assessment Guideline, November 2013 uses the term, “bottom up” approach.

381    The Tribunal’s direction to use a bottom-up review was prefaced by the injunction to do so in accordance with the Tribunal’s reasons and we think that it was perfectly clear that the extent of the bottom-up review is up to the AER. The Tribunal, at [408], referred to sensible administration suggesting that the regulator would apply “some form of quantitative “reasonableness check bottom-up analysis in at least some, if not all, of the opex components”. It is also informative to note that in considering the AER’s use of the EI Model as the sole determinative of opex, the Tribunal, at [449], noted evidence that Ofgem used a “toolkit” of approaches (top-down econometric models, bottom-up unit cost analysis, bottom-up engineering assessments, assessments of historic costs and assessments of forecast costs) and referred, at [452], to the triangulation of “top down” benchmarking with other sources of information, for example, review by expert engineering consultants of unit costs, volumes of work, policies and practices.

382    As to the submission that the reference to a bottom-up review lacked a legal justification, the AER submitted (correctly in our view) that r 6.5.6(c) requires it to form a view about the total of the forecast opex and that it does not require an assessment of individual components of opex. It is the next step in the AER’s submission that is difficult. It submitted that the Tribunal’s direction purports to direct the AER in the performance of its functions “in a manner that is contrary to the requirements of the NER”. We do not accept this. There is no provision in the NER which states or even suggests that it is inappropriate for the AER to conduct a bottom-up review as part of the performance of its function under rr 6.5.6 and 6.12.1.

383    As to the submission that the reference to a bottom-up review lacked an evidentiary justification, the AER’s main submission was that the direction was based on an erroneous conclusion by the Tribunal that the AER had not conducted any form of bottom-up review. That this was the view taken by the Tribunal was said to be revealed in various paragraphs in its reasons (at [389], [408], [434]) and that it was erroneous was said to be supported by the AER’s examination of labour costs and vegetation management costs and the work done when adjusting the figures to take account of the OEFs. We do not think this is a material point in light of the statement by AER’s counsel that the AER had conducted bottom-up review until the present (emphasis added) and its own statement that this was the first time it was using benchmarking as the primary basis for determining forecast opex.

384    Secondly, the AER submits that the direction by the Tribunal that it use a broader range of modelling and benchmarking is uncertain in its terms and has no evidentiary justification. The submission was that the Tribunal gave no indication of the type of modelling which it had in mind and that was in a context in which (so the AER had submitted) it had used a broad range of modelling. The AER submitted that the AER’s techniques included two partial performance indicators, two productivity index number models, three econometric models and category analysis metrics to examine key categories of opex (i.e., labour costs and vegetation management costs). The AER submitted that the Tribunal did not identify an alternative form of modelling or suggest what data might be available to use in the model.

385    The Tribunal did not accept that the AER carried out a broad range of modelling. As we have said, the Tribunal referred to the other models as closely related models, all derived from the same data and missing the same wider review of factors and sense checks. The value of the Partial Performance Indicators was limited and the reviews in the areas of labour costs and vegetation management costs were qualitative not quantitative. As far as the models to be used, we agree with the electricity network respondents’ submissions that the experts’ reports indicated that there were a wide range of models that could be used and that it was open to the Tribunal to proceed on that basis.

Conclusion with respect to Forecast Opex

386    For the reasons we have given, the AER has not established any of the grounds of judicial review in relation to forecast opex.

RETURN ON DEBT

Introduction

387    The AER’s case is that the Tribunal’s decision and orders with respect to the AER’s estimation of the return on debt for each of the electricity network respondents involved one or more of the grounds of reviewable error identified in s 5 of the ADJR Act. We identify these grounds with more precision below. It is sufficient for us to say at this stage that the AER’s case focused, firstly, on whether some contentions advanced by the electricity network respondents could properly have been raised in the limited merits review before the Tribunal in light of the provisions of s 71O(2) of the NEL and, secondly, on whether the Tribunal erred in its construction and application of the allowed rate of return objective referred to in r 6.5.2 of the NER, including in relation to r 6.5.2(k)(4) which pertains to the need for the AER to have regard to impacts arising from any change in the methodology used to estimate the return on debt from one regulatory period to another. Such a change occurred in the present case.

388    This section of our reasons is structured in the following way. First, we will set out those provisions of the NER which are relevant to the AER’s task in estimating the return on debt. Secondly, we will summarise the AER’s final decisions and then the Tribunal’s reasons, in some detail. It is necessary for us to do so in order to put the detailed submissions of the parties in an appropriate setting. Thirdly, we will return to the grounds of review relied on by the AER – although, as we will explain, the AER distilled these grounds into a number of propositions, which the parties then addressed. Finally, we will summarise, to the extent necessary, the competing submissions of the parties (and, where relevant, of the intervenors) on each proposition and, in relation to each proposition, set out our conclusions and reasons on whether the AER’s grounds of review have been made out.

The National Electricity Rules

389    Return on debt is an aspect of the return on capital building block. The return on capital building block is designed to provide a service provider with a return on capital that gives a return on equity to investors and enables the service provider to service the interest on its loan debt by means of an “allowed rate of return”.

390    Under r 6.5.2(a) of the NER, a service provider’s return on capital for each regulatory year is calculated by applying a rate of return for that service provider to the value of the service provider’s regulatory asset base. This rate of return (that is, the allowed rate of return) must be determined so that it achieves the “allowed rate of return objective”: r 6.5.2(b).

391    The allowed rate of return objective is important. It is defined in r 6.5.2(c):

The allowed rate of return objective is that the rate of return for a Distribution Network Service Provider is to be commensurate with the efficient financing costs of a benchmark efficient entity with a similar degree of risk as that which applies to the Distribution Network Service Provider in respect of the provision of standard control services (the allowed rate of return objective).

392    It is not in dispute that “standard control services” are services that are provided by a regulated entity.

393    The NER specify that the allowed rate of return is a weighted average of the return on equity for the regulatory control period and the return on debt for each regulatory year in the regulatory control period. The allowed rate of return must be determined on “a nominal vanilla basis” that is consistent with the estimate of the value of imputation credits referred to in r 6.5.3: r 6.5.2(d).

394    For present purposes, specific discussion of the return on equity aspect of the capital building block can be put to one side. There is no application for judicial review of the Tribunal’s determinations in that regard. It is important to note, however, that the return on equity for a regulatory control period, and the return on debt for a regulatory year in that period, must each be estimated so as to contribute to the allowed rate of return objective: rr 6.5.2(f) and (h). Thus, the efficient financing costs of a benchmark efficient entity, as described in r 6.5.2(c) of the NER, is a central consideration in the implementation of the capital building block.

395     The return on debt may be estimated using a methodology which results in either the return on debt for each regulatory year being the same or the return on debt being, or potentially being, different for different regulatory years in the regulatory period: r 6.5.2(i). The methodology adopted to estimate the return on debt may be designed to result in the return on debt reflecting the return that would be required by debt investors in a benchmark efficient entity if it raised debt at the time of, or shortly before, the making of a determination for the regulatory control period: r 6.5.2(j)(1). This is referred to colloquially as the “on-the-day” approach. The methodology adopted to estimate the return on debt may also reflect the average return that would have been required by debt investors in a benchmark efficient entity if it raised debt over an historical period prior to the commencement of a regulatory year in the regulatory control period: r 6.5.2(j)(2). This is referred to colloquially as the “trailing average portfolio” or, simply, “trailing average” approach (for convenience, we will use the latter description). Rule 6.5.2(j)(3) contemplates that the methodology for estimating the return on debt may be a combination of these two approaches. That said, r 6.5.2(j) makes clear that the AER is not limited to these methodologies. Nevertheless, in the present case, these two methodologies were the focus of the AER’s attention in making its final determinations.

396    In estimating the return on debt, r 6.5.2(k) requires the AER to have regard to certain identified factors. One of those factors is the interrelationship between the return on equity and the return on debt: r 6.5.2(k)(2). Importantly for present purposes, another factor is prescribed in r 6.5.2(k)(4):

In estimating the return on debt under paragraph (h), regard must be had to the following factors:

(4)    any impacts (including in relation to the costs of servicing debt across regulatory control periods) on a benchmark efficient entity referred to in the allowed rate of return objective that could arise as a result of changing the methodology that is used to estimate the return on debt from one regulatory control period to the next.

397    As we explain in the next section of these reasons, in its final determinations the AER considered that, for the regulatory period in question, the methodology for estimating the return on debt should reflect a trailing average approach. This approach was discussed in the Rate of Return Guideline entitled ‘Better Regulation: Explanatory Statement Rate of Return Guideline’ published by the AER on 13 December 2013, following a period of consultation (the ROR Guideline). This represented a departure from the previous on-the-day approach, which had been required by the NER at the time of previous determinations. The AER reasoned that, amongst other benefits, its resort to the trailing average approach would be more reflective of the actual debt management approaches of non-regulated businesses and, therefore, more likely to represent efficient financing practice. The AER also reasoned that, because of this change in approach, a transition arrangement should be implemented. As will become apparent, the AER’s determination of the need for a transition arrangement became the focal point for the disagreement between the parties on the topic of the return on debt.

398    For relevant purposes, the return on debt has two components. The first component is the risk free (or base) rate. The second component is a risk premium over the base rate, called the debt risk premium (DRP). It is common ground that, at relevant times, a service provider could hedge the base rate but not the DRP.

The AER’s final decisions

399    The AER considered four options for determining the return on debt. These options combined forms of the on-the-day and trailing average approaches. The first option was to continue the on-the-day approach. The second option was to start with an on-the-day rate for the first regulatory year and gradually transition into a trailing average approach over 10 years (Option 2). The third option was to start with an on-the-day rate for the base rate component and gradually transition this component into a trailing average approach over 10 years. The DRP component of the return on debt would be determined using a backwards looking trailing average DRP. In other words, there would be no transition for this component. As we have noted, the DRP component could not be hedged. The AER referred to the combined elements of this option as a “hybrid transition”. The fourth option was to adopt a backwards looking trailing average approach, with no transition on either the base rate component or the DRP component (Option 4).

400    The AER adopted Option 2, as foreshadowed in the ROR Guideline. In its calculation of the allowed return on debt for the 2014/15 regulatory year, the AER applied an on-the-day rate shortly before the start of the regulatory period to 100% of the debt portfolio. The rate of return reflected the return on BBB+ rated bonds. For the 2015/2016 regulatory year, it applied the on-the-day rate to 90% of the debt portfolio with the remaining 10% updated to reflect prevailing interest rates during the averaging period for that year. The averaging period was a period of 10 to 40 business days occurring as close as practicable to the start of the regulatory year. For the 2016/17 regulatory year, the AER applied the on-the-day rate to 80% of the debt portfolio, with 10% based on prevailing interest rates during the averaging period for the 2015/16 regulatory year and 10% updated to reflect prevailing interest rates during the averaging period for the 2016/17 regulatory year. In the same way, the AER applied this methodology for subsequent regulatory years by updating 10% of the return on debt each year to reflect prevailing interest rates during the averaging period for each year. On this approach, the return on debt would be, on completion of the transition period, a simple average of prevailing interest rates over the previous 10 years. This transition was based on the approach recommended by the Queensland Treasury Corporation (QTC). The AER referred to this as “the QTC approach”.

401    It should be noted that the AER’s final decision, in each case, only determined the return on debt methodology for the 2014/19 regulatory period, which was the first five years of the 10 year transition period. This is because the AER only had power to determine, at this point in time, the return on debt methodology for the 2014/19 period.

402    The AER described its approach as one that involved a “gradual forward looking transition to a trailing average”. It saw this approach as one based on well-established economic, financial and regulatory principles. The AER said that its approach would reflect its position regardless of whether prevailing interest rates were higher or lower than the 10 year historical average, thereby avoiding a potential bias in regulatory decision-making that could arise from choosing an approach that used historical data after the results of that historical data were already known. The AER saw Option 2 as having numerous other advantages over the other options. For present purposes, it is not necessary to descend to the detail of the AER’s findings and reasons in that regard.

403    Option 2 was opposed by the electricity network respondents, who argued for the adoption of Option 4.

404     It is necessary at this stage to say something about the AER’s conception of a benchmark efficient entity. The AER determined that the benchmark efficient entity was a single entity across all service providers that was “a pure play, regulated energy network business operating within Australia”.

405    By way of explanation, a “pure play” business is one that offers services focused in one industry or product area. The AER’s conception was that the benchmark efficient entity provides only regulated energy network services. Energy network services comprise gas distribution, gas transmission, electricity distribution or electricity transmission services. The AER saw the benchmark efficient entity as operating within Australia because this location determined the conditions under which the business operated, including the regulatory regime, tax laws, industry structure and broader economic environment.

406    Importantly, the AER conceived of the benchmark efficient entity as a regulated entity that was subject to economic regulation (that is, revenue price cap regulation) under the NER and/or the NGR. Its reasons for coming to this conclusion are discussed in the Explanatory Statement to the ROR Guideline.

407    The AER saw the NSW service providers opposition to Option 2, and advocacy of Option 4, as relating to their actual financing practices as opposed to those of a benchmark efficient entity. It is to be noted in this regard that the NSW service providers each held a portfolio of staggered debt. ActewAGL held no debt. However, as we have noted, ActewAGL also opposed Option 2 and advocated Option 4.

408    The AER did not consider the position of the NSW service providers to be efficient from the perspective of a benchmark efficient entity. For example, the AER found that Ausgrid had managed its refinancing risk, but had not taken steps to manage its interest rate risk. Here, “refinancing risk” refers to the risk that a benchmark efficient entity would not be able to refinance its debt when it matured and “interest rate risk” refers to the risk associated with a mismatch between the allowed return on debt and a benchmark efficient entity’s actual return on debt. The AER was satisfied on the evidence before it that the efficient practices of a benchmark efficient entity, with similar characteristics to Ausgrid, would have managed both its refinancing risk and its interest rate risk.

409    In this connection, the AER considered that an efficient financing practice of a benchmark efficient entity under the on-the-day approach would have been to borrow long term and stagger the borrowing so that only a small proportion of the debt matured each year. Importantly, it also considered that a benchmark efficient entity would have combined this practice with interest rate swap contracts to broadly match the base rate component of its actual return on debt to its return on debt allowance. The AER explained how this financing practice would have been undertaken. Once again, it is not necessary to descend to the detail of the AER’s discussion in this regard.

410    The AER reasoned that, if it were to apply Option 4, it would determine a return on debt that was founded on inefficient financing practices from the perspective of a benchmark efficient entity. It reasoned that this would be inconsistent with the allowed rate of return objective.

411    In its final decisions, the AER gave consideration to a number of arguments raised by the electricity network respondents, including an argument that there was more than one efficient financing practice under the on-the-day approach. The AER rejected all those arguments.

412    One of the arguments was that a transition, as in Option 2, delayed the imposition of the best approach to estimate the return on debt, while prolonging the use of an inferior approach. The NSW service providers considered that a transition was unwarranted for service providers that currently adopted a staggered debt portfolio with no interest rate swaps.

413    The AER rejected this particular argument on the basis that its task was to set the return on debt allowance by reference to a benchmark efficient entity in the context of the prevailing regulatory regime – not by reference to actual network service providers. The AER expressed its satisfaction that a gradual transition closely matched a benchmark efficient entity’s financing costs over the next regulatory control period as it transitioned its financing practices to match the trailing average approach.

414    It is important to understand the rationale for the AER’s approach to transitioning. Because it had defined the benchmark efficient entity as a pure play, regulated energy network business, the AER could not observe directly what the efficient debt management practices of such an entity would be under the trailing average approach. This is because, hitherto, the NER required the AER to apply the on-the-day approach to such entities when determining the return on debt. The AER’s decision to provide for a transition relied on what the AER considered a benchmark efficient entity, as described by it, would need to transition from the on-the-day approach to the trailing average approach. The electricity network respondents contended that the imposition of a 10 year transition to the trailing average approach did not reflect the cost of debt of the NSW service providers and did not reflect the cost of debt for a hypothetical benchmark entity in the position of the NSW service providers and ActewAGL. However, the AER eschewed the idea that any transitional arrangement should take into account the individual circumstances of businesses and, in particular, their debt financing practices. It considered that such an approach would be inconsistent with the allowed rate of return objective and represent a break from the fundamentals of incentive-based regulation.

The Tribunal’s decision

415    Before the Tribunal, the electricity network respondents raised the question of the proper construction and application of r 6.5.2 of the NER, in particular the proper construction and application of the allowed rate of return objective specified in r 6.5.2(c). The matter of concern was how the AER’s new approach of adopting a trailing average to calculate the return on debt was to be transitioned. There was no disagreement that the trailing average approach was an acceptable methodology that was available to the AER under the NER. The matter of concern – whether and how the change to a trailing average methodology should be transitioned – is illustrated by Ground 118 of Ausgrid’s application to the Tribunal:

The AER’s decision involved an error of fact, or alternatively an incorrect exercise of discretion or was unreasonable, in that the AER identified the wrong “benchmark efficient entity”. The AER’s decision to impose a transition on the movement to the new trailing average methodology was dependent upon the AER’s view of the debt management practices that a hypothetical entity would have adopted under the previous “on the day” approach. In adopting this stance, the AER erred in identifying the relevant benchmark efficient entity as a benchmark regulated efficient entity, whereas the correct benchmark entity was a benchmark unregulated efficient entity, i.e. an entity competing in a workably competitive unregulated market. Such a benchmark entity would have a conventional debt portfolio of the type held by privately-owned entities in unregulated markets, namely a staggered portfolio of fixed rate debt. The debt financing costs of such a portfolio matches the debt cost calculated under the AER’s new trailing average approach. Therefore, once the correct benchmark entity is identified, there is no basis for the imposition of a transition. The AER’s selection of an incorrect benchmark entity, and imposition of a transition, involved:

(a)    an error in the construction and application of clause 6.5.2, particularly subclauses 6.5.2(c), (j) and (k); and

(b)    a decision contrary to the NEO and the revenue and pricing principles, in that the selection of a regulated entity as a benchmark would not impose an appropriate pricing signal for investment, i.e. the pricing signal that would be sent as a result of competition in a workably competitive market, rather than a pricing signal from the idiosyncratic application of a prior regulatory methodology.

(Underlining in original.)

416    As we have noted, the electricity network respondents contended that there should be no transition, with the consequence that, for the regulatory period in question, there should be an immediate application of the trailing average approach.

417    PIAC was granted leave to intervene on this issue in relation to the decisions concerning the NSW service providers. Its contention was that there should be a transition which should commence from 2015/2016 rather than 2014/2015. This was because the on-the-day rates at the time of, and leading up to, the AER’s final decisions were declining. PIAC argued that its approach better complied with r 6.5.2 and would lead to allowances over the regulatory period that would be very much less than the effect of the AER’s final decisions.

418    The question of the proper construction and application of r 6.5.2, as argued before the Tribunal, involved a number of issues. Relevantly, these included the related issues of:

1.    whether the electricity network respondents were precluded by the operation of s 71O(2)(a) of the NEL from raising the issue, in the review before the Tribunal, of whether the benchmark efficient entity was an unregulated firm operating in a workably competitive market;

2.     whether the benchmark efficient entity was a regulated entity;

3.    whether the benchmark efficient entity was a common entity for all distribution network service providers; and

4.    in the circumstances, whether, for the purposes of r 6.5.2(k)(4) of the NER, the AER was in error in adopting the transitioning arrangement represented by Option 2.

419    We turn to consider the Tribunal’s conclusions and reasons on each of these issues.

The operation of s 71O(2)(a)

420    We have already discussed the operation of s 71O(2)(a) of the NEL, in general terms: see [151] above.

421    The Tribunal found (at [878]) that the issue of whether the benchmark efficient entity was an unregulated firm operating in a workably competitive market was one that had been raised by the electricity network respondents in submissions made to the AER.

422    At [880], the Tribunal quoted from submissions made by the NSW service providers to the AER. Although not identified in the Tribunal’s reasons, there does not seem to be any doubt that these submissions were made in a letter dated 13 February 2015 from Herbert Smith Freehills, who were acting for those service providers. The Tribunal quoted the following passage from the submissions:

The notion of the hypothetical “benchmark efficient entity” is a tool designed to ensure that the relevant service provider only recovers revenue in respect of the efficient conduct of the business in a hypothetical competitive environment, not the inefficient conduct of the business in a monopoly environment.

423    At [881], the Tribunal noted that these submissions made the claim that, in moving to the trailing average methodology, the efficient financing costs of a benchmark efficient entity with a similar degree of risk as that which applied to the NSW service providers was the cost of issuing debt on a fixed rate staggered portfolio basis; that the service providers had already issued debt on that basis; and that, accordingly, there was no requirement in the interests of efficiency, or for the avoidance of monopoly pricing, for imposing a delay in the movement to the preferred methodology.

424    At [882], the Tribunal concluded that these service providers had raised the matter that the efficient financing costs of a benchmark efficient entity were the financing costs that would be expected in a competitive environment. Before the Tribunal, the AER seems to have referred to particular paragraphs from the submissions which, according to the AER, argued for a benchmark efficient entity that was regulated. At [883], the Tribunal expressed the view that it did not consider that the submissions made this argument.

425    At [884], the Tribunal referred to the AER’s reliance on a report of Frontier Economics entitled Cost of Debt Transition for NSW Distribution Networks, January 2015 (the Frontier Economics Report) apparently as support for the proposition that the benchmark efficient entity was regulated. The Tribunal found that there was nothing in that report which said that the benchmark efficient entity was a regulated entity. On the contrary, the Tribunal noted that pages 8 to 9 of the report considered the efficient practices of an unregulated infrastructure service provider as a potential benchmark and noted that such a provider would have adopted a fixed rate staggered maturity approach.

426    At [885], the Tribunal referred to the AER’s submission that the revised regulatory proposals submitted by the NSW service providers did not contain any reference to the definition of the benchmark efficient entity. The Tribunal nevertheless found that it was clear that the NSW service providers were maintaining their position. In this connection, the Tribunal noted that each regulatory proposal made reference to the predominant debt management approach of non-regulated infrastructure firms (such as ports, airports, roads and railways) of issuing debt on a staggered portfolio/trailing average basis, and that the NSW service providers had relied on those examples as supporting the approach which they (the NSW service providers) had adopted.

427    The Tribunal found (at [886]) that ActewAGL’s submission (that the AER should have found that the benchmark efficient entity was an unregulated firm operating in a workably competitive market) was raised and maintained in its revised regulatory proposal where (at 473-474) it submitted that:

The financing practices of relevance to the term “efficient financing costs” do not encompass practices adopted in response to a pre-existing regulatory approach to the estimation of the return on debt notwithstanding whether one of the characteristics of the benchmark efficient entity that informs the degree of risk for which capital market investors require compensation is that that entity is regulated.

Such a construction of the term “efficient financing costs” is consistent with the objective of the regulatory regime established by the NEL and the Rules, as evinced by the NEO and the RPPs, which is itself concerned with creating incentives for efficiency and mimicking, so far as practicable, the outcomes of a workably competitive market, including in particular by creating incentives for providers to operate and invest in the manner of a firm in a competitive environment.

428    At [887], the Tribunal found that these submissions asserted that the pre-existing regulatory approach was of no relevance to the “efficient financing costs” of the allowed rate of return objective, regardless of the characteristics of the benchmark efficient entity. The Tribunal noted that ActewAGL’s revised regulatory proposal said that the financing practices of relevance to the term “efficient financing costs” did not encompass practices adopted in response to “the regulatory approach”.

429    At [888], the Tribunal noted a related contention by the AER that even if the “matter” was raised by any of the material referred to, it was not raised with sufficient precision to fulfil the requirements of s 71O(2) because the NSW service providers and ActewAGL did not identify and explain their approach as a departure from the ROR Guideline. At [889], the Tribunal rejected this contention, finding that the NSW service providers and ActewAGL had identified and provided reasons for a departure from the ROR Guideline. This was because they sought the immediate adoption of the trailing average approach, with no transition. The Tribunal found that the regulatory proposals submitted by these service providers did not apply the transitional arrangements set out in the ROR Guideline and explained why the guideline should not be followed in that respect. The Tribunal concluded that this was sufficient to satisfy the requirements of the NER: see, in that regard, S6.1.3(9) of the NER.

A regulated and common entity for all service providers?

430    The Tribunal considered the second and third issues together. It noted that, before the Tribunal, the AER had adhered to the views it had expressed in its final decisions.

431    At [904], the Tribunal identified that the critical step taken by the AER was its adoption of one regulated entity as the benchmark efficient entity. At [905], the Tribunal reasoned that, once that step was taken, some service providers would necessarily be materially disadvantaged. In this connection, it was common ground that each of the NSW service providers would recover a significantly greater sum (as the return on debt) by an immediate application of the trailing average approach. The Tribunal noted the NSW service providers’ argument that they would be deprived of this advantage by reason of (what they termed was) a “one size fits all” transition process. This was because, in their submission, their debt financing structures were the efficient structures that the AER was seeking to achieve by the introduction of the trailing average approach and that, by the transition process imposed by the AER, they were being given an artificial debt structure as a starting point. According to the NSW service providers, this artificial debt structure depressed their recoverable financing costs below their actual and (what they said were) efficient financing costs.

432    At [907], the Tribunal expressed its view that the benchmark efficient entity referred to in the allowed rate of return objective was not a regulated entity. The Tribunal also expressed the view that the benchmark efficient entity need not necessarily be the one entity for the purposes of all regulatory decision-making in a particular regulatory period for all regulated service providers – although, at [921], the Tribunal also said that it was not persuaded that the AER erred by adopting a single benchmark efficient entity.

433    The Tribunal’s reasons for coming to the view that the benchmark efficient entity was not a regulated entity were explained at [908]-[922]. We summarise those reasons in the remaining paragraphs of this section: [434]-[438] below.

434    The Tribunal said that the AER was required to make its regulatory determinations in relation to a regulated service provider as if the relevant service provider were operating in a competitive environment (even though it was not) and that the benchmark efficient entity was imposed as a proxy for the hypothetical unregulated competitor. The Tribunal explained that, if it were otherwise, the starting point for the AER would be a regulated competitor in a hypothetically regulated market. The Tribunal said that such an approach would not be consistent with the policy underlying the purpose of the NEL and NGL in relation to the fixing of terms under which monopoly providers may operate. The Tribunal also reasoned that the use of a regulated efficient entity as the base comparator would divert the AER from the role of fixing the terms for supply of services on a proxy basis (ie, compared to those likely to obtain in a competitive market) and focus attention on some different and (the Tribunal said) unidentified regulated market. In this connection, the Tribunal observed that the AER had imposed the trailing average methodology as that most likely to represent the proxy for the cost of debt for a supplier of the relevant services in a competitive market.

435    The Tribunal acknowledged that the AER’s decisions on return on debt (and other topics) were to be made by reference to the efficient financing costs of a benchmark efficient entity, rather than the actual financing costs of the particular regulated service provider. Once fixed, such costs provided the economic incentive to the provider to operate more efficiently. However, the Tribunal also noted that the benchmark efficient entity must have a similar degree of risk to that of the relevant service provider in respect of which the allowed rate of return is to be determined. As different service providers had, in fact, different degrees of risk, and so may have had different efficient financing cost structures, there would not be an identical benchmark efficient entity for all service providers.

436    The Tribunal then reasoned that the reference to “debt investors” in r 6.5.2(j) of the NER was likely to be a reference to investors in a competitive market rather than in a regulated service provider, otherwise the measure for comparison was unidentified and might not lean towards an efficient entity.

437    The Tribunal also reasoned that the interrelationship between the return on equity and the return on debt, as required to be taken into account by r 6.5.2(k)(2), pointed to the same conclusion. The Tribunal said that the return on equity and the return on debt had a complementarity that was significant and only meaningful if measured by similar, or similarly conceptual, yardsticks. As the return on equity was to be measured by the prevailing conditions in the market for equity funds (see r 6.5.2(g) of the NER), it would follow that market conditions for the benchmark efficient entity should also be used to measure the return on debt, rather than some undefined regulated conditions.

438    The Tribunal concluded that the comparison provided in the allowed rate of return objective was not one that involved an artificial or contrived comparator. It required the benchmark efficient entity to be one with a similar degree of risk as that which applied to the distribution network service provider (ie. a particular service provider). It was necessary, therefore, to consider how the particular service provider should have structured its financing costs under the former regulatory regime – relevantly, how it should efficiently have done so in response to the on-the-day methodology of estimating the rate of return. As different service providers may have had different degrees of risk, there was scope for a range of structures of efficient financing costs to exist at the end of one regulatory period. That range of structures then assumed significance for the purposes of r 6.5.2(k)(4) of the NER: see [396] above.

The correct transitioning arrangement?

439    At [925], the Tribunal noted that the AER’s starting point – that the efficient debt financing structure for the on-the-day methodology was one that involved hedging – meant that the debt financing structures of the NSW service providers (who did not hedge) and ActewAGL (who did not have debt financing) were, by the AER’s definition, inefficient and that the implementation of the trailing average methodology required transitioning, in their instance, in a manner which was obviously artificial.

440    At [926], the Tribunal remarked:

It is somewhat ironic that, by that process, the BEE at the end of the current regulatory period, under the trailing average approach, would (subject to particular considerations) have the characteristics of the financing cost structure of Networks NSW at the commencement of the current regulatory period. That is because, by its approach, the AER has treated that current financing cost structure as inefficient, even though that structure (subject to particular considerations) underlies the trailing average approach.

441    At [927], the Tribunal considered that if a different benchmark efficient entity financing cost structure were to be adopted as the starting point, it would be necessary to revisit the AER’s approach to, and consideration of, the factor to which regard must be had under r 6.5.2(k)(4).

442    At this point in its reasons, the Tribunal considered a contention made by the AER that the effect of debt transition on a particular service provider was, ultimately, a largely irrelevant consideration. The AER’s contention was summarised at [930] of the Tribunal’s reasons:

... The relevant matters that the AER must have regard to under r 6.5.2(k)(4) are any impacts on a BEE that could arise from a change in methodology, including in relation to the cost of debt across regulatory control periods. Accordingly, the effect of debt transition on a particular service provider can be relevant only to the extent that it provides some information about how a change in debt methodology would impact a BEE. Therefore, it says, it is not a mandatory relevant consideration whether an immediate transition to the trailing average approach would cause any cost and inconvenience for Networks NSW and ActewAGL because they either have no debt or staggered non-hedged debt. The relevant consideration is the efficient financing costs of a BEE, not the particular DNSPs the subject of a decision.

443    The Tribunal rejected this contention. In doing so, it said (at [932]):

The contention … is ironical. It takes the regulated BEE (which is chosen by the AER as a standard from the range of individual network providers financing costs structures, which are their idiosyncratic individual responses to the on-the-day methodology), and then selects a transition option to achieve a financing costs structure to that reflected by, and in, the trailing average approach. So it is converting the hypothesised regulated BEE from one financing costs structure which it therefore regards as “the efficient costs structure”, but which ultimately it regards as inefficient, to another financing costs structure. And in doing so, it does not have to have regard to the fact that Networks NSW already have that financing costs structure (not necessarily in the efficient form), but it deems Networks NSW and ActewAGL to have some other costs structure for the purposes of the transition process.

444    The Tribunal concluded (at [933]) that, in having regard to any impacts on a benchmark efficient entity that could arise as a result of changing the methodology used to estimate the return on debt, r 6.5.2(k)(4) required the AER to start with the efficient financing costs of a benchmark efficient entity (not a regulated entity but, as described at [914] of the reasons, a hypothetical, efficient competitor in a competitive market for the services); to determine whether the benchmark efficient entity would suffer any impacts as a result of the changed methodology; and, if so, to have regard to those impacts in deciding on the transition process to the new methodology.

445    At [934], the Tribunal explained that the starting point was not the service provider’s actual financing costs, but what would be the efficient financing costs having regard to the service provider’s degree of risk.

446    By way of elaboration, the Tribunal found that the financing costs of the NSW service providers were readily applied to the trailing average methodology. For those service providers, the Tribunal concluded that the relevant inquiry would start with whether their actual financing costs were efficient as at the commencement of the new regulatory control period. If not, those of the benchmark efficient entity would be applied prospectively.

447    In the case of the other service providers, the Tribunal concluded that the relevant inquiry would start with whether the actual financing costs (including hedging costs) were efficient at the start of the new regulatory period having regard to the service provider’s particular degree of risk. If so, the impacts of the changed methodology (to a trailing average approach) would require “the sort of transition process” which had been imposed in the AER’s final decisions. If not, the starting point for the transition would be “some refinement to the efficient financing costs within that structure”.

448    The Tribunal considered that r 6.5.2(c) – which speaks of a benchmark efficient entity with a similar degree of risk as that which applies to the service provider – supported these conclusions. It also considered that this approach had “a degree of common sense” in that it represented a means of looking realistically to the actual consequences of the changed methodology. At [935], the Tribunal explained:

… It means, contrary to the AER submission, that an actual assessment must be made of the efficient (not just the actual) financing costs of each DNSP as it has responded in its methodology for estimating the return on debt for the prior regulatory period and an actual assessment must be made of the impacts on those efficient financing costs of that DNSP by the changed methodology.

449    The Tribunal also considered that its conclusions on the requirements of r 6.5.2(k)(4) were consistent with the AEMC’s reasons for the changes introduced by the 2012 rule amendments: see at [936].

450    Accordingly, at [937], the Tribunal concluded that the NSW service providers and ActewAGL had established a ground or grounds of review. At [938], the Tribunal explained:

The selection or identification of the BEE as a regulated entity involved the wrong exercise of a discretion about the character of the BEE in all the circumstances, and as a consequence its decision on the topic was unreasonable in all the circumstances. It may have been possible to identify the specific features of the regulated BEE which then, as a matter of fact, might be said to involve errors of fact in its findings of fact, but it is not necessary to go into that detail. Similarly, its exercise of its discretion to apply the characteristics of its selected regulated BEE to the transition process in the case of Networks NSW and ActewAGL is also erroneous, and its decision on the transition process was unreasonable, in all the circumstances.

The grounds of review

451    In its applications to the Court for judicial review, the AER raised a number of grounds of review in relation to the Tribunal’s decision concerning the return on debt. At the hearing, these grounds were distilled into six contentions, with the AER nominating the following corresponding grounds of review in its written submissions:

1.    The Tribunal erred by receiving and acting upon a matter that was not raised and maintained by the service providers in their submissions to the AER: Ground 15 (electricity network respondents).

2.    The Tribunal erred in concluding that the AER was bound to address the allowed rate of return objective on the basis that the benchmark efficient entity must be an unregulated entity: Grounds 12(a)-(d), (f) and 13(a)-(f) (electricity network respondents); Grounds 6(a)-(d), (f) and 7(a)-(f) (JGN).

3.    The Tribunal erred in its construction of r 6.5.2(k)(4) of the NER: Grounds 12(e) and (g), and 13(g) (electricity network respondents).

4.    The Tribunal’s conclusions as to what was required to be done to comply with rr 6.5.2(c) and (k) revealed error: Grounds 13(f), (g), (h) and (j) (electricity network respondents).

5.    The Tribunal failed properly to undertake a review of the AER’s decision on return on debt: Grounds 13(i) and 14 (Ausgrid, but presumably meaning the electricity network respondents).

6.    The Tribunal erred in setting aside the AER’s decision with respect to JGN’s debt transition: Ground 8 (JGN).

452    Consistently with the approach of the parties, we have treated these contentions as raising the issues for determination in this proceeding. We summarise the parties’ respective written and oral submissions, and our conclusions, in respect of these issues at [453]-[616] below, other than the last contention concerning JGN’s debt transition. We have dealt with that contention in separate reasons: Australian Energy Regulator v Australian Competition Tribunal (No 3) [2017] FCAFC 80. It should be noted that JGN adopted the submissions of the electricity network respondents to the extent that those submissions covered the grounds of review discussed in these reasons that were common with JGN’s position with respect to the estimation of the return on debt under the NGR (specifically, Grounds 6, 7 and 9 of the application for judicial review relating to JGN corresponding to Grounds 12, 13 and 14 of the applications for judicial review relating to the electricity network respondents).

Whether the Tribunal erred by receiving and acting upon a matter that was not raised and maintained by the service providers in their submissions to the AER

The AER’s submissions

453    In its written submissions, the AER commenced by emphasising, firstly, the purpose of s 71O(2) which, it said, was to ensure that service providers “must engage” with the AER and other interested parties “throughout the entire decision-making process” and then, secondly, the terms of the allowed rate of return objective which, it said, was not directed to providing a rate of return on a service provider’s actual or expected financing costs but on the efficient financing costs of the benchmark efficient entity.

454    The AER also placed emphasis on the promulgation of the ROR Guideline including, in particular, the submissions that were made and supported by various parties at that time. The AER said that these submissions had advanced the conceptual definition of a benchmark efficient entity as a regulated entity. The AER referred, in particular, to a submission to this effect advanced by the Energy Network Association (the ENA) which, it said, was supported by the electricity network respondents. It said that conception of a benchmark efficient entity as a regulated entity was adopted in the final form of the ROR Guideline.

455    The AER argued that none of the electricity network respondents had identified in their proposals (or elsewhere) that they proposed a departure from the ROR Guideline, even though it was open to them to do so under S6.1.3(9) of the NER. The AER also argued that it did not understand that any electricity network respondent had taken issue with the notion that the allowed rate of return was to be calculated by reference to a benchmark efficient entity that was regulated. The AER said that, accordingly, it did not address this issue in its determination and that, consequently, the Tribunal had not reviewed a decision which addressed this issue.

456    The AER then returned to consider the four “documents” which the Tribunal referred to in reaching its conclusion that the electricity network respondents had raised the issue that the benchmark efficient entity was an unregulated firm operating in a workably competitive market. The first three documents concern the position of the NSW service providers. The fourth document concerns the position of ActewAGL.

457    The first document was constituted by the submissions made by the NSW service providers to the AER in the letter dated 13 February 2015 from Herbert Smith Freehills. This letter was sent after the initial proposals were submitted in May 2014, the draft decisions were published in November 2014, and revised proposals were submitted in January 2015. The AER argued that this letter contained no submission or contention that the benchmark efficient entity should be unregulated. The AER noted that the Tribunal itself only found (at [882]) that the NSW service providers had raised the matter that the efficient financing costs of the benchmark efficient entity were the financing costs that would be expected in a competitive environment. The AER focused on what it perceived to be the limited nature of the submission that the NSW service providers had made: it was a submission about how one determined what financing costs were efficient; it was not a submission about the nature or characteristics of the benchmark efficient entity itself.

458    The second document was the Frontier Economics Report. It is to be recalled that the Tribunal (at [884]) noted that nothing in that report said that the benchmark efficient entity was a regulated entity. Indeed, the Tribunal referred to the fact that the report considered the efficient practices of an unregulated infrastructure service provider as a potential benchmark. In this proceeding, the AER argued that the contention made in the Frontier Economic Report was whether efficient financing costs could be assessed against the financing practices of unregulated monopolies. The AER argued that the report did not take issue with the conception of the benchmark efficient entity as a regulated entity.

459    The third document was constituted by the revised regulatory proposals submitted by the NSW service providers. The AER argued that the most that could be said in relation to the revised proposals was that they adopted the same contention that had been advanced in the Frontier Economics Report concerning the debt management approaches of unregulated infrastructure service providers.

460    The fourth document was ActewAGL’s revised regulatory proposal in which it argued that AER had misconstrued the term “efficient financing costs” in r 6.5.2(c). We have set out the relevant passages at [427] above. The AER argued that these passages acknowledged that the benchmark efficient entity was a regulated entity, whilst at the same time advancing the contention that efficient financing practices should be informed by what would occur in a workably competitive market. The AER argued that these passages could not be understood as advancing a submission that the conception of the benchmark efficient entity should be changed to an unregulated entity.

461    The AER submitted that it was clear from these documents that the electricity network respondents were advancing submissions about what were, and what were not, efficient financing costs. They were not advancing a contention that the benchmark efficient entity must be unregulated or that there should be a departure from the ROR Guideline (to the extent that the Guideline suggested that the benchmark efficient entity was regulated).

462    The AER also pointed to two contextual matters. First, the AER argued that the conception of a benchmark efficient entity is relevant to the whole of the rate of return framework (in particular, to the allowed rate of return objective), not just in the context of debt. The AER said that the electricity network respondents’ submissions before the Tribunal were advanced only in the context of debt and not more broadly. The AER said that a change in the conception of the benchmark efficient entity from a regulated to an unregulated entity would have important flow-on effects for the determination of the return on capital. And yet, according to the AER, these flow-on effects were not addressed by the electricity network respondents in their proposals or revised proposals. Indeed the AER pointed to the service providers’ reliance, in the context of return on equity, on material and expert evidence concerning North American network service providers. The AER submitted that the reliance placed on this material was inconsistent with the contention that the benchmark efficient entity was an unregulated entity. The AER submitted that if the service providers wanted to take this issue before the AER, the entirety of their return on capital building block should have been premised on an unregulated benchmark efficient entity. The AER said it was not and that the Tribunal erred in concluding to the contrary. Secondly, the AER said that the electricity network respondents’ submissions were responsive to the view expressed in the AER’s draft decisions that the practice of the NSW service providers in hedging debt exposure under the previous on-the-day approach was not efficient. In other words, the submissions advanced before the AER were about efficient financing costs. They were not about whether the benchmark efficient entity was regulated or unregulated.

463    The AER also referred to the level of particularity with which an issue must be raised and maintained for the purposes of s 71O of the NEL by calling in aid general statements of principle which emphasise the need for clarity, precision and openness in the conduct of litigation. The AER argued that “an equal or even higher level of candour and particularity with respect to the identification of critical arguments and contentions” is to be expected in non-adversarial proceedings, such as the regulatory-decision making process before the AER, and that s 71O is “the statutory expression of that expectation”.

464    The AER contended that the first occasion on which the electricity network respondents proposed a departure from the conception of the benchmark efficient entity referred to in the ROR Guideline was when they made their applications to the Tribunal for leave to review the AER’s determinations.

The electricity network respondents’ submissions

465    The electricity network respondents emphasised the need to bear in mind the matter actually addressed in their submission to both the AER and the Tribunal on this aspect of the review, namely the correct approach to satisfying the allowed rate of return objective in r 6.5.2(c) of the NER when estimating the return on debt.

466    In order to elucidate the matter, the electricity network respondents described what they saw as the basic difference between the AER’s position and their position. The starting point for this analysis was the uncontroversial fact that, prior to the current regulatory control period, it had not been possible for the AER to adopt a trailing average approach to estimating the return on debt; only the on-the-day approach was available. It was equally uncontroversial that the AER had reasoned that, under the on-the-day approach, managing both refinancing risk (the risk that a benchmark efficient entity would not be able to refinance its debt when it matured) and interest rate risk (the risk associated with a mismatch between the allowed return on debt and a benchmark efficient entity’s actual return on debt) could be achieved by employing a staggered debt portfolio with interest rate swaps. The AER found that this was the financing strategy generally adopted by most privately-owned service providers under the on-the-day approach and concluded that this would be an efficient financing practice for a benchmark efficient entity. It is important to understand this point because it explains the perceived need for a transition. As the Tribunal explained at [863], an immediate change to a trailing average approach would require such an entity to “unwind” its hedging contracts which might be costly, if at all possible. The AER saw the need for a gradual transition so that the benchmark efficient entity could adjust to these changes.

467    The electricity network respondents distinguished their position, which was that the NSW service providers had staggered portfolios of fixed rate debt and thus, according to them, a debt portfolio of the type that would be held in a competitive market. They said that this strategy reflected the trailing average approach and that no period of transition was required. It is to be remembered, in this connection, that ActewAGL had no debt (it was funded by equity investment) and thus, for this reason, required no transition.

468    The electricity network respondents argued that, rather than considering whether debt costs were efficient, and rather than considering what would be the efficient financing costs incurred by a business operating in a competitive market, the AER focused on the hypothetical impact of hypothetical hedge contracts for a hypothetical business. The electricity network respondents argued that the AER’s approach was one which reflected its view of an appropriate response to a different regulatory regime (the change from the on-the-day approach to the trailing average approach) rather than focusing on the regulatory allowance for debt reflecting or simulating debt costs incurred by a business operating in a competitive market.

469    Having set this scene, the electricity network respondents submitted that the relevant issue was not whether the benchmark efficient entity was regulated or unregulated (the focus of the AER’s submissions) but, rather, whether efficient debt financing costs should be assessed by reference to a hypothetical response to a previous regulatory regime (as they said the AER had done) or by considering the efficient financing costs that would be incurred by a business operating in a competitive (i.e. unregulated) environment. In this connection, the electricity network respondents argued that references by the Tribunal to the benchmark efficient entity being “regulated” or “unregulated” were no more than shorthand expressions for this issue.

470    Having identified this issue, the electricity network respondents turned to the question of whether they had raised and maintained that matter before the AER.

471    The electricity network respondents referred to submissions that the NSW service providers had made in response to the original consultation paper on the ROR Guideline which had been repeated in those service providers’ original regulatory proposals, as well as to passages in their revised regulatory proposals. Importantly, the electricity network respondents also referred to passages from the NSW service providers’ submissions contained in the letter dated 13 February 2015 from Herbert Smith Freehills which had been quoted, in part, by the Tribunal at [880] of its reasons: see [422] above. It is desirable that we quote more extensively from those submissions, which appear under the heading “The proper role of the hypothetical benchmark efficient entity”:

6. The notion of the hypothetical “benchmark efficient entity” is a tool designed to ensure that the relevant service provider only recovers revenue in respect of the efficient conduct of the business in a hypothetical competitive environment, not the inefficient conduct of a business in a monopoly environment. For example, it is a tool to ensure that only efficient expenditure, rather than inefficient expenditure, is recovered from consumers. It is thus a tool for rewarding efficiency and ensuring that consumers are not exposed to monopoly pricing. It is a tool that has to be applied sensibly and rationally and with discretion, rather than dogmatically, in order to achieve the overall national electricity objective. It certainly does not require the entire revenue calculation exercise to be conducted on some hypothetical basis divorced from reality.

9. Given that the AER has correctly recognised, in moving to the trailing average methodology, that the efficient financing costs of a benchmark efficient entity with a similar degree of risk as that which applies to Networks NSW is the cost of issuing debt on a fixed rate staggered portfolio basis, and Networks NSW already issues its debt on that basis, there is no requirement in the interests of efficiency or the avoidance of monopoly pricing for imposing a delay in the movement to the best methodology. There is no scope for the positing of a hypothetical scenario of hypothetical contractual obligations preventing the immediate transition to an efficient financing basis where that scenario does not reflect reality. The AER’s approach assumes that there is some contractual obligation preventing immediate movement to a return on debt commensurate with efficient financing costs. That is nonsensical where there is no such obligation affecting Networks NSW. The imposition of a delay will not reflect the financing costs of Networks NSW which are already equivalent to the efficient financing costs of the benchmark efficient entity.

10. In this regard, it is relevant to observe that in amending clause 6.5.2, the AEMC stated:

… the Commission considered that the long-term interests of consumers would be best served by ensuring that the methodology used to estimate the return on debt reflects, to the extent possible, the efficient financing and risk management practices that might be expected in the absence of regulation.

11. As observed by Frontier Economics, unregulated infrastructure service providers tend to issue long-term fixed-rate debt on a staggered maturity cycle, and obviously do not enter into hedge contracts to fix their debt for five years at the rate prevailing during the averaging period. The AEMC’s observations, relevant to the proper construction of the NER, have been turned on their head by the approach of the AER.

12 The AER’s approach is not in accordance with the NER because it imposes a departure from efficient financing costs, which are those of a fixed rate staggered portfolio. The same efficient financing costs are applicable to a benchmark efficient entity. The AER’s approach is thus not in accordance with the allowed rate of return objective in clause 6.5.2(c). That position does not change because of some theoretical contractual obligation that could have been undertaken in a previous period, but which the DNSP did not undertake. That is to posit the wrong benchmark efficient entity for the purposes of clause 6.5.2. Put shortly, the “benchmark efficient entity” is (at most) a more efficient version of the actual entity. The concept of a “benchmark efficient entity” does not require the positing of an entity saddled with statutory or contractual obligations that do not otherwise exist.

472    The electricity network respondents also referred to ActewAGL’s response to the original consultation paper on the ROR Guideline and to the draft ROR Guideline, which was that the best approach to determining the cost of debt allowance was the trailing average approach. The electricity network respondents argued that ActewAGL maintained its position in its regulatory proposal. The Tribunal did not specifically refer to ActewAGL’s statements in this regard. Nevertheless, it is appropriate that we record that, after stating that it supported the adoption of a ten year trailing average approach, ActewAGL said:

ActewAGL Distribution considers, however, that a transition to this agreed long-term benchmark efficient debt management strategy is neither desirable nor permissible under the NER. This is because, in circumstances where ActewAGL Distribution has zero debt financing, leaving the issuance of debt funding to its equity investors, and those equity investors have adopted differing debt management strategies, there is no principled basis for the AER to depart from the estimation of ActewAGL Distribution’s return on debt consistent with the efficient debt management strategies of the benchmark efficient entity by imposing such a transition, nor is the imposition of such a transition permissible under the NER.

473    The electricity network respondents also referred to statements made by ActewAGL in its revised regulatory proposal. At [886] of its reasons, the Tribunal quoted some of these passages and summarised other passages: see [427] above. Once again, it is desirable that we quote more extensively. Prior to the passage quoted by the Tribunal at [886], ActewAGL said:

ActewAGL Distribution contends that the establishment of the transitional arrangements proposed by the AER is impermissible because those arrangements result in a return on debt that does not contribute to the achievement to the rate of return objective as required by clause 6.5.2(b) and (h) of the Rules. Put another way, ActewAGL Distribution considers that the transitional arrangements result in a return on debt that is not commensurate with the efficient debt financing costs of the benchmark efficient entity as is required by clause 6.5.2(b) and (h).

ActewAGL Distribution submits that, regardless of the characteristics of the ‘benchmark efficient entity’, the pre-existing regulatory approach to the estimation of the return on debt is of no relevance to the ‘efficient financing costs’ referred to in the rate of return objective or, thus, to the content of that rate of return objective.

The term ‘efficient financing costs’ is properly construed as referring to the costs of capital commensurate with the riskiness of the investment where efficient financing practices are adopted. Those costs are a product of the return required by capital market investors (in the case of the return on debt, debt holders) having regard to the degree of risk consequent upon the characteristics of the benchmark efficient entity.

474    Following the passage quoted by the Tribunal at [886], ActewAGL said:

Construing the term ‘efficient financing costs’ as encompassing the costs incurred as a consequence of a pre-existing regulatory approach to the estimation of the return on debt would be to effectively define the rate of return objective, being the criterion for selection of the regulatory approach to estimation of the return on debt, by reference to the pre-existing regulatory approach for estimation of the return on debt. Such a construction would be perverse.

475    Later in its revised regulatory proposal, ActewAGL criticised AER’s approach, saying that it would appear to:

misconstrue the rate of return objective, in that it construes ‘efficient financing costs’ as encompassing the actual financing costs incurred, and practices that, acting rationally, would have been adopted, by the benchmark efficient entity in response to a pre-existing regulatory approach that (on its own admission) did not, and was not designed to, estimate a return on debt that achieves the rate of return objective …

476    In light of these matters, the electricity network respondents submitted that they had raised and maintained submissions that the “efficient financing costs” were the efficient costs of an unregulated business (i.e. a business operating in a workable competitive market) and that the AER’s approach of setting the efficient costs of debt as a response to a previous regulatory regime was erroneous. The electricity network respondents submitted that the AER’s contention that they did not raise and maintain before the AER the matter which was the subject of the Tribunal’s consideration was, therefore, without foundation.

477    By way of further elaboration, the electricity network respondents argued that, in its submissions to this Court, the AER had posited a false dichotomy between, on the one hand, the benchmark efficient entity as an unregulated entity, and, on the other hand, the contention that the efficient financing costs of a benchmark efficient entity were those that were to be expected in a competitive (unregulated) environment. The electricity network respondents argued that this was a false dichotomy because, in the context of the return on debt, the relevant issue was the efficient financing costs of a benchmark efficient entity. If the efficient costs were those of an unregulated business, then that was the benchmark that is being propounded.

The AER’s reply submissions

478    The AER argued that the electricity network respondents’ submissions were an attempt to redefine the AER’s grounds of judicial review, which attempt, it said, revealed a misunderstanding by the electricity network respondents of the scope of the AER’s judicial review application. The AER quoted part of [118] of Ausgrid’s leave application before the Tribunal (see [415] above) which, it said, had been expressed identically in the other electricity network respondents’ leave applications. In this paragraph an express reference was made to, and distinction drawn between, a regulated efficient entity and an unregulated efficient entity, with an unregulated efficient entity being identified as the correct benchmark to be applied. The AER also called in aid selected parts of the electricity network respondents’ joint submissions to the Tribunal. The AER said that its challenge in the present judicial review proceedings arises because it contends that the Tribunal erred in construing the NEL and the NER to require the benchmark efficient entity to be providing unregulated services. The AER then argued that, on that basis, the relevant question is whether the electricity network respondents submitted to the AER that the benchmark efficient entity must be conceived of as unregulated and, if they did, whether the Tribunal erred in concluding that the NER compel the AER to conceive of the benchmark efficient entity as unregulated.

479    In their written submissions, the electricity network respondents had contended that there was no dispute that the AER should move to a trailing average approach when estimating the return on debt. We understand the electricity network respondents’ submissions to refer to the fact that they did not dispute that a trailing average approach should be adopted. However, in reply, the AER advanced and explained its own position, saying that the interrelationship between the debt estimation methodology used and any mode of transition was critical to its reasoning. It said that its approach was to consider both together in order to consider the appropriate overall debt estimation methodology. It presaged the possibility that, in the absence of a suitable transition which contributed (in its view) to the achievement of the allowed rate of return objective, it would have continued with the on-the-day approach (ie, it would have adopted Option 1).

480    The AER took issue with a submission by the electricity network respondents that, from the outset, they had raised with the AER their concern with the AER’s approach to the question of debt transition. The AER submitted that this was not the relevant question. It said that the relevant question was whether, in submissions before it, the electricity network respondents challenged the conceptual definition of a benchmark efficient entity which the AER had adopted in the ROR Guideline. The AER contended that nothing in the submissions advanced by the electricity network respondents before it made that challenge. In this connection, the AER also pointed to the fact that some of the documents relied upon by the electricity network respondents in this Court were not among the three documents which the electricity network respondents had relied upon before the Tribunal to argue that they had raised and maintained the relevant matter. The AER again canvassed what it said was disclosed in the documents to which the Tribunal had referred.

481    The AER accepted that the electricity network respondents had raised before it the fact that Option 4 was the debt estimation approach that should be adopted. It also accepted that the electricity network respondents had raised the appropriate approach to assessing what were the efficient financing practices under the on-the-day approach for the purpose of assessing whether a transition was necessary. However, the AER contended that a submission about what constitutes efficient financing practices, and a submission about the character of the benchmark efficient entity, were not “the same thing”. It also submitted that the conception of a benchmark efficient entity is critical because it provides the conceptual framework within which the appropriate debt estimation methodology is to be determined.

Analysis and conclusion

482    For the purposes of s 71O(2)(a) of the NEL, it is necessary to identify the “matter” that is sought to be raised in respect of the reviewable regulatory decision. As we have said at [151] above, this is a question of substance, not mere form.

483    At [877] of its reasons, the Tribunal identified the AER’s contention that s 71O(2)(a) precluded the raising of the issue whether the benchmark efficient entity was an unregulated firm operating in a workably competitive market. In this proceeding, the AER, on occasion, put the issue somewhat more narrowly, namely whether the benchmark efficient entity must be unregulated. In each case, the focus of the AER’s argument was about the benchmark efficient entity as such (i.e. its characteristics). However this is, at best, a fragment of the true matter raised by the electricity network respondents, which was that, in estimating the return on debt in accordance with the allowed rate of return objective, it was not appropriate to adopt Option 2 and that Option 4 should be adopted because the efficient financing costs of a benchmark efficient entity appropriate to their circumstances did not require any transition to the trailing average approach of estimating debt financing costs.

484    There is no real doubt that the Tribunal understood that this was the matter that the electricity network respondents were seeking to raise in the review before the Tribunal and in respect of which they said the AER had erred within s 71C(1) of the NEL. This is particularly evident from the Tribunal’s discussion at [880]-[883] and [885] (in respect of the NSW service providers), at [886]-[887] (in respect of ActewAGL), and at [889] (in respect of all the electricity network respondents). The gravamen of the electricity network respondents’ complaint was that, although it was appropriate to adopt a trailing average approach to estimate the allowed return on debt, it was not appropriate to transition to this approach on the basis that, theretofore, the NER required the AER to estimate the return on debt on the basis of the on-the-day approach. We would accept that there are some passages in the Tribunal’s reasons which, if isolated from the paragraphs we have noted immediately above, might be understood as considering the characteristics of a benchmark efficient entity as such: see [883]-[884] and perhaps [887]. However, those passages should not be read in that isolated way. We accept the electricity network respondents’ submission that references by the Tribunal to whether the benchmark efficient entity was regulated or unregulated should be viewed as a shorthand reference to the more broadly expressed issue that those parties were obviously seeking to agitate before the Tribunal.

485    Was this issue or “matter” raised and maintained by the electricity network respondents in submissions to the AER before the reviewable regulatory decisions were made? We are in no doubt that it was. We have already quoted extensively from the submissions that were made: see [471]-[475] above. We will not repeat those submissions. We are also in no doubt that the AER was seized of that issue and dealt with it.

486    In this latter connection, it is convenient to take, as an example, the AER’s final decision in respect of Ausgrid. It was not suggested in argument that the position in respect of the other electricity network respondents was any different. When, in that final decision, the AER decided to estimate the return on debt by applying an on-the-day rate for the first regulatory year in the 2014-2019 period and then, for subsequent years, applying a gradual transition to a trailing average using a “forward looking” approach, it specifically rejected what it described as Ausgrid’s “backwards looking” trailing average. The AER did not accept that Ausgrid’s debt financing practices (or those of the NSW service providers more generally) were efficient from the perspective of a benchmark efficient entity, which would manage both refinancing risk and interest rate risk. The AER considered that, if it were to apply Ausgrid’s proposed approach, it would determine a return on debt that was founded, in its view, on inefficient financing practices from the perspective of a benchmark efficiency entity which, in turn, would be inconsistent with the allowed rate of return objective. Thus, the AER engaged with and addressed the very same matter that, subsequently, the electricity network suppliers agitated in their applications for review before the Tribunal. It is true that, in its leave application to the Tribunal, Ausgrid, for example, drew a distinction between the benchmark efficient entity as a regulated entity and the benchmark efficient entity as an unregulated entity (meaning an entity competing in a workably competitive, and hence unregulated, market): see the ground expressed at [118] of the leave application. But this was in the context of the broader argument to which we have referred – in particular, whether there was any basis for imposing a transition to the new trailing average approach. As we later explain, a fixation on whether, for the purposes of r 6.5.2(c) of the NER, the benchmark efficient entity is a regulated or unregulated entity raises a false dichotomy that tends to mask the substance of the relevant matter in contention that the electricity network respondents had raised and maintained before the AER. Nonetheless, the NSW service providers’ reference to an “unregulated entity” in this context was, in large measure, allusive and is properly seen as a shorthand way of distinguishing their approach from the AER’s approach with a view to explaining what they saw as the error in the AER’s approach. Their basic proposition, both before the AER and the Tribunal, was that the AER had applied the wrong benchmark which, in the result, was antithetical to the allowed rate of return objective because, in a workably competitive market, efficient financing costs would be represented by a staggered portfolio of fixed rate debt, which they already held without the need to transition, and that their debt costs should be benchmarked accordingly.

487    For these reasons, there was no judicially reviewable error on the part of the Tribunal in concluding that the real issue or “matter” in question had been raised and maintained by the electricity network respondents before the AER. It follows that the grounds of review covered by this contention must be rejected.

488    We add the following observations.

489    First, whether or not, at the time that the ROR Guideline was promulgated, the electricity network respondents advanced a particular position as to the characteristics of the benchmark efficient entity that might be seen to be contrary to their position adopted before the Tribunal (see [454] above), is, with respect, beside the point. Whether, at that time, a contrary positon was advanced is a matter of contention between the parties and need not be addressed in these reasons. The real question is the identification of the matter advanced before the Tribunal and whether that matter was raised and maintained by the relevant service providers in submissions to the AER before the relevant reviewable regulatory decisions were made. For the reasons given, we are satisfied, in each case, that the identified matter was so raised and maintained.

490    Secondly, we are not in a positon to deal with the AER’s submission that the electricity network respondents advanced inconsistent positions before the Tribunal in relation to the return on equity (on the one hand) and the return on debt (on the other). Once again, this appears to be a matter of contention between the parties. There was no ground of judicial review before us concerning the Tribunal’s determination in respect of the return on equity and no detailed submissions were made which exposed any inconsistency as alleged. It is, perhaps, sufficient for us merely to note that we accept the electricity network respondents’ submission that the AER’s submission in this regard is not a proper matter for complaint given that the appropriate focus, for present purposes, is whether the respondents raised and maintained, in the context of the return on debt, the basis for their contention that it was not appropriate to adopt Option 2 and that Option 4 should be adopted.

Whether the Tribunal erred in concluding that the AER was bound to address the ROR Objective on the basis that the benchmark efficient entity must be an unregulated entity

Introduction

491    It is convenient to commence the consideration of this issue by recording the Tribunal’s stated understanding of the economic foundations of the regulatory regime imposed by the NEL.

492    After quoting the NEO set out in s 7 of the NEL, the Tribunal said (at [77]):

The ultimate objective reflected in the NEO and NGO is to direct the manner in which the national electricity market and the national natural gas market are regulated, that is, in the long term interests of consumers of electricity and natural gas respectively with respect to the matters specified. The provisions proceed on the legislative premise that their long term interests are served through the promotion of efficient investments in, and efficient operation and use of, electricity and natural gas services. This promotion is to be done “for” the long term interests of consumers. It does not involve a balance as between efficient investment, operation and use on the one hand and the long term interest of consumers on the other. Rather, the necessary legislative premise is that the long term interests of consumers will be served by regulation that advances economic efficiency.

493    The Tribunal also quoted (at [78]) the following passage from ElectraNet Pty Limited (No 3) at [15]:

The national electricity objective provides the overarching economic objective for regulation under the Law: the promotion of efficient investment in the long term interests of consumers. Consumers will benefit in the long run if resources are used efficiently, i.e. resources are allocated to the delivery of goods and services in accordance with consumer preferences at least cost. As reflected in the revenue and pricing principles, this in turn requires prices to reflect the long run cost of supply and to support efficient investment, providing investors with a return which covers the opportunity cost of capital required to deliver the services.

494    The Tribunal noted (at [80]) its acceptance, in Envestra Ltd (No 2), of the AER’s submission that service providers should be permitted a reasonable opportunity to recover “legitimate costs”, namely the costs that would be incurred in a “workably competitive market”. This phrase appears to have been sourced from the AEMC Rule Determination, National Electricity Amendment (the Economic Regulation of Transmission Services) Rule 2006 No. 18, where the AEMC described the fundamental objective of regulation as being:

… to reproduce, to the extent possible, the production and pricing outcomes that would occur in a workably competitive market in circumstances where the development of a competitive market is not economically feasible…

495    The Tribunal also referred to the High Court’s observations, in an analogous context, in EAPL at [18]:

The context and purpose of the Code is well understood, not least because the objectives of the legislation are articulated in the legislation itself in considerable detail. The Code as a whole provides for a regulatory regime of a kind which is “a surrogate for the rewards and disciplines normally provided by a competitive market”. Competitive pressures in a market stimulate efficiency of production and resource allocation, they stimulate efficient investment decisions and they minimise costs. No party disputed the fact that the regulatory process set out in the legislation was directed to eliminating monopoly pricing whilst nevertheless providing a rate of return to pipeline owners, commensurate with a competitive market…

496    The parties did not challenge these statements of general principle or criticise the Tribunal’s understanding of them. Indeed, all parties appeared to embrace them.

The AER’s submissions

497    The AER noted that the allowed rate of return objective specified in r 6.5.2(c) of the NER required it to determine a rate of return for a benchmark efficient entity with a similar degree of risk as that which applies to the service provider in respect of the provision of standard control services. As we noted at the outset, standard control services are regulated services. Thus, the AER submitted, the benchmark efficient entity was one that had a similar degree of risk (from the perspective of debt and equity investors) to the service provider’s regulated business.

498    The AER argued that this conclusion was “strongly reinforced” by the language of r 6.5.2(k)(4), which required the AER, when estimating the return on debt, to have regard to “any impacts … on a benchmark efficient entity … that could arise as a result of changing the methodology … used to estimate the return on debt” from one regulatory control period to another. The AER submitted that the only way that a change in methodology could have an impact was if the benchmark efficient entity was conceived as undertaking regulated activities; otherwise, a change in methodology for determining revenue could have no such impact and, according to the AER, r 6.5.2(k)(4) would be otiose.

499    The AER also argued that the application of revenue/price gap regulation to a service supplier’s business self-evidently affected its risk profile. The AER argued that this feature could not be ignored; to do so would fail to reflect an important level of risk faced by debt and equity investors, which could lead to either under-compensation or over-compensation in the circumstances.

500    Thus, according to the AER, the Tribunal erred by finding at [907] that the benchmark efficient entity referred to in the allowed rate of return objective was not a regulated entity.

501    The AER submitted that this was enough to found error in the Tribunal’s determination, because the Tribunal’s conclusions involved an erroneous construction of r 6.5.2(c) of the NER. Nonetheless, the AER submitted that further errors could be identified in the Tribunal’s reasoning.

502    First, the AER took issue with the Tribunal’s statement at [908] that the AER is to make its regulatory determination in relation to a regulated service provider “as if the relevant provider were operating in a competitive environment”. The AER put the matter this way:

The regulatory scheme seeks to obtain the benefits of economic efficiency through mimicking the processes that would occur in a competitive market. It seeks to eliminate monopoly pricing by imposing through regulation a revenue cap that would otherwise be imposed in a workably competitive market by the process of competitive rivalry. By the imposition of a revenue cap, the regulatory regime provides a surrogate for the rewards and disciplines normally provided by a competitive market, by stimulating efficient investment decisions and minimising costs…

(Emphasis in original.)

503    The AER submitted that there is nothing in the NEL which provided a basis for concluding that the rate of return on debt must be assessed by reference to an unregulated entity operating in a competitive environment or required an assumption as to what would be the outcome if the service provider were operating in a competitive environment. The AER argued that, to approach the determination of the return on equity and the return on debt in that way, would result in investors not being rewarded for the risks to which they were actually exposed. The AER argued that what the regulatory scheme sought to do was to ensure that sufficient revenue was permitted to cover efficient operating costs and then to allow a return on debt and equity for investors which was commensurate with the risk that they would actually face by investing in a benchmark efficient entity with a similar degree of risk to the service provider in question. The AER submitted that the error in the Tribunal’s approach was to focus on why the regulatory scheme existed (the absence of competition) rather than the end to be achieved (economic efficiency through the imperfect substitute of regulation). Thus, the AER submitted:

The approach of the Tribunal would perpetuate the inefficient outcomes the regulatory scheme is intended to eliminate by permitting returns to investors that are out of line with the risks their investments bear.

504    The AER contended that the same error was repeated at [914] of the Tribunal’s reasons where it said that the benchmark efficient entity was “likely to refer to the hypothetical efficient competitor in a competitive market for” the services in question. The AER submitted that this was not correct because regulation was a proxy for the rewards and disciplines of a competitive market when, in fact, no such market existed. Once again, if the rate of return were set by reference to what would be obtained in a competitive market, investors would be compensated for risks which they did not bear. The AER submitted:

The Tribunal should have found that the NER … requires the rate of return to be assessed by reference to an efficient entity that provides network services regulated by reference to a revenue control mechanism: that actually represents the risks faced by investors in [distribution network service providers].

505    The same point appears to have been made in the AER’s criticism of the Tribunal’s reasoning at [917] where the Tribunal said, in the context of r 6.5.2(j) of the NER (concerning the methodology that is to be adopted when estimating the return on debt), that debt investors in a benchmark efficient entity were likely to be investors in a competitive market. The AER submitted that r 6.5.2(j) contradicted, rather than supported, the Tribunal’s ultimate conclusion on the nature of a benchmark efficient entity. The AER did not elaborate on this submission other than to repeat the point that the Court should reject a construction of r 6.5.2(j) that would see debt investors remunerated for risks that they did not take. The AER emphasised, in this connection, the stipulation in the allowed rate of return objective that the benchmark efficient entity must be seen as having a similar degree of risk as that which applies to the service provider in respect of the provision of standard control (regulated) services. The AER submitted that this strongly indicated that the benchmark efficient entity, like the service provider, should be seen as an entity with a business governed by revenue or price cap regulation.

506    Secondly, the AER submitted that the Tribunal had, in various places in its reasons (including at [915]), conflated the separate issues of (a) whether the benchmark efficient entity should be conceptualised as carrying on either a regulated or unregulated business and (b) how an unregulated infrastructure firm might arrange its debt portfolio (which might be relevant to determining what the efficient financing costs of a regulated benchmark efficient entity might be).

507    Thirdly, the AER took issue with what it saw as the Tribunal’s attribution to the AER of an acceptance that different service providers have different degrees of risk and, consequently, may have different efficient financing cost structures. This led the Tribunal to conclude (at [916]) that “there would not be an identical BEE for all DNSPs”. The AER submitted, in effect, that it did not accept (and had not accepted) that different service providers had different degrees of risk. Indeed, it argued that both the ROR Guideline and its final decisions proceeded on the basis that all service providers had a similar degree of risk in relation to their regulated services.

508    At this point, the AER contrasted the Tribunal’s conclusion (that there will not be an identical benchmark efficient entity for all service providers) with its later conclusion (at [921]) that, in fact, it was not persuaded that the AER erred by adopting a single benchmark efficient entity for the regulated service providers. The AER submitted that these were inconsistent findings which introduced uncertainty as to its future task on remitter: was the AER required to have multiple conceptions of a benchmark efficient entity (as [916] of the reasons would suggest) or could it maintain its view that all service providers, in respect of their similarly regulated businesses, had a similar degree of risk?

509    Relatedly, the AER referred to the fact that, when making the final decisions, it had given detailed consideration to the question of risk as it informed the level of compensation required by equity and debt investors. The AER said that none of this material was mentioned by the Tribunal or found by the Tribunal to reveal reviewable error. The AER argued that the Tribunal’s failure to identify, specifically, which grounds of review had been upheld, showed that the Tribunal had failed to undertake its review in the manner required.

510    Fourthly, the AER criticised the Tribunal’s reasoning at [918] which called in aid r 6.5.2(k)(2) of the NER which requires the AER, when determining the return on debt, to have regard to the interrelationship between the return on debt and the return on equity. The Tribunal’s point was that, as the return on equity must be measured by having regard to prevailing conditions in the market for equity funds (see r 6.5.2(g)), so too should market conditions be used to measure the return on debt. As the Tribunal put it, there was a complementarity between the two measures which was only meaningful if measured by similar, or similarly conceptual, yardsticks.

511    The AER submitted that its approach was not inconsistent with the complementarity between the return on debt and the return on equity to which the Tribunal had referred. The AER accepted that the return on debt should be assessed by reference to competitive debt markets. It argued, however, that this said nothing about whether a benchmark efficient entity’s services were regulated (or not) or whether debt investors understood that they were investing in a business with regulated revenue.

512    In aid of its submission, the AER pointed to the fact that it had estimated the return on equity by reference to what investors in a competitive equity market would require for an investment in a benchmark efficient entity with regulated services. The AER argued that, as the Tribunal found no error in this approach, it must follow that the Tribunal accepted that the AER had correctly applied rr 6.5.2(f) and (g) (concerning the methodology that is to be adopted when estimating the return on equity) – in other words, by reference to a benchmark efficient entity subject to regulation. The AER submitted that the interrelationship required by r 6.5.2(k)(2) must be with respect to the same benchmark efficient entity and that the Tribunal’s contrary approach, when dealing with the estimation of return on debt (the benchmark efficient entity is not a regulated entity) was “internally inconsistent, irrational and unreasonable”.

513    The AER also submitted that the Tribunal’s contention (at [918]) that the AER’s approach involved “some undefined regulated conditions” was not correct, because “regulated” refers to the regulation applicable to standard control services, which means the imposition of price or revenue control on the entity concerned.

PIAC’s submissions

514    PIAC endorsed the AER’s submissions on this topic. In doing so, it emphasised the requirement of r 6.5.2(c) that the benchmark efficient entity is to be taken as having a similar degree of risk as that which applies to the service provider in respect of the provision of standard control services.

515    In this connection, PIAC argued that the Tribunal’s reasoning at [914] “miscarried”. It said that the “ultimate objective” of the revenue regulation of regulated monopolies was to ensure that the prices paid by consumers approximate the prices that would be charged in a workably competitive market. It said that this end would be “undermined” if the regulated prices charged to consumers reflected a higher degree of risk than that actually faced by investors in a regulated market.

516    PIAC said that it followed that, in applying r 6.5.2(k)(4), the AER was correct to have regard to the impacts of a proposed transition on a “regulated” notional benchmark efficient entity. PIAC also said that it would “make an oxymoron of r 6.5.2(k)(4)” to characterise the benchmark efficient entity as “unregulated” when the Tribunal had upheld the AER’s determination on the return on equity.

The electricity network respondents’ submissions

517    The electricity network respondents described the underlying principle of the present regulatory scheme as the regulation of the conduct of an asset owner (the service provider) to prevent it from charging monopoly prices by permitting it only to earn revenue and charge prices that would be charged and earned in a workably competitive market. They referred to the Tribunal’s statement of the economic foundations of the regulatory regime (summarised in [492]-[495] above) and noted that this regime was a surrogate for the rewards and disciplines normally provided by a competitive market. The electricity network respondents agreed with the AER’s summation (quoted at [502] above) concerning the aims and objectives of the present regulatory regime. They said that the AER’s observations were “entirely orthodox and well understood”. However, it was at this point that the electricity network respondents parted company with the AER, contending that the AER’s “ultimate submission” (that efficient financing costs should not be assessed by reference to an unregulated entity operating in a competitive environment) was a non sequitur.

518    In this connection, the electricity network respondents argued that the economic theory underpinning the regulatory regime meant that the recoverable costs of the benchmark efficient entity would only be the efficient costs of an entity operating in a competitive market. In other words, contrary to the AER’s submission, the benchmark was certainly not a regulated entity.

519    The electricity network respondents argued that a regulated monopoly could not provide the appropriate benchmark for the revenues or prices that should be earned or charged. In this connection, the electricity network respondents argued, firstly, that the regulation process was a “simulation exercise” (seeking to simulate a competitive market) so that positing the benchmark efficient entity as a regulated entity would be “circular”.

520    Secondly, the electricity network respondents argued that the costs of a regulated entity were the product of regulation itself, which created a different form of risk, including the risk that the regulator will make an incorrect decision.

521    The electricity network respondents advanced a similar submission based on the requirement of the allowed rate of return objective that the benchmark efficient entity must have “a similar degree of risk” as that of a service provider. The electricity network respondents submitted that this risk was the risk flowing from the nature of the business itself (which they described as an infrastructure business having certain characteristics), not from the fact that the business was regulated. The electricity network respondents argued that the assessment of risk flowing from a regulated business would include and require an assessment of the “track record” of the regulator in making correct decisions – an exercise which, they said, was alien to the purpose of the regulatory scheme.

522    Next, the electricity network respondents argued that none of the submissions raised by the AER properly engaged with its own approach to the question of debt financing costs. The issue considered by the Tribunal was not just whether regulation could be taken into account in assessing investment risk in the context of estimating efficient financing costs. The electricity network respondents submitted that even if it were appropriate to have regard to the impact of regulation, it would not provide a basis for the AER’s approach to debt transition, which looked to the optimal strategy of a benchmark efficient entity transitioning from the requirements of debt financing under a previous regulatory regime. The electricity network respondents said that, in adopting that approach, the AER took the notion of a benchmark efficient entity to “an absurd and illogical extreme”, which marked a “radical departure” from the accepted approach of mimicking the processes that would occur in a competitive market. The electricity network respondents submitted that the nature of the proper simulation exercise was correctly identified by the Tribunal, which also recognised (at [921]) that the AER’s approach involved a degree of circularity.

523    The electricity network respondents then turned to the particular position of ActewAGL which, they argued, highlighted the artificiality and inappropriateness of the AER’s approach of imposing a debt transition. As we have noted, ActewAGL had no debt. It was fully financed by equity. In these circumstances, the electricity network respondents posed the question: what was there to transition? They argued that, rather than being entitled to monopoly revenue, ActewAGL was entitled “to simulated efficient competitive market revenue which, in the case of debt, is calculated by the application of the trailing average approach”.

524    Finally, the electricity network respondents said that PIAC’s submissions added nothing of substance to AER’s submissions.

525    There is an additional matter to which we should refer. The electricity network respondents contended that the AER’s criticism that the Tribunal had acted inconsistently when dealing with the return on equity and the return on debt (see [512] above), was unwarranted. They argued that no party had raised any issue with the AER’s return on equity calculation based on any potential differences in the assessment of risk between regulated or unregulated entities. The electricity network respondents pointed to the fact that, when dealing with the return on equity, the Tribunal had rejected different complaints, and that the nature and derivation of an appropriate benchmark for the calculation of the return on equity was not a matter before the Tribunal. Thus, the electricity network respondents contended, no conclusion could be drawn about whether the AER had applied the correct benchmark when determining the return on equity.

The AER’s reply submissions

526    The AER criticised the electricity network respondents’ submissions as relying heavily on assertions about the overall outcome sought by the NEL and NER without examining the precise operation of the scheme in question. The AER elaborated on that proposition. It is not necessary for us to repeat that detail. Salient aspects of it have been noted in our earlier summary of the parties’ submissions. The AER contended that, while the parties agree that regulation is relevant to the determination of the degree of risk faced by a firm, their positions diverge from that point, with the electricity network respondents engaging in “faulty reasoning”. The AER again pointed to the language of the allowed rate of return objective, which refers to the benchmark efficient entity as having a similar degree of risk as the service provider in question. It said that the electricity network respondents’ submissions proceeded on the basis that, as the regulatory process is seeking to simulate a competitive market, this risk should not be taken into account when conceiving of the benchmark efficient entity, but rather the risk faced by an unregulated business. The AER argued that this approach does not properly consider or apply the words of the NER in describing the regulatory task.

527    The AER challenged the electricity network respondents’ submission that the risk addressed by the allowed rate of return objective is the risk flowing from the nature of the business itself, not the fact that the business is regulated. The AER argued that if, by this submission, the electricity network respondents were seeking to contend that the compensation required by hypothetical equity and debt investors was to be based on the service provider competing in a market for the supply of electricity services, then the electricity network respondents’ argument was flawed because it was based on a preconception of what the NEL and NER were seeking to achieve, and not an application of the relevant test for each element of the building block methodology. The AER said that the electricity network respondents’ approach would lead to hypothetical equity and debt investors being compensated for risks they do not face namely, absence of regulation and the risks of workable competition. The AER maintained its position that there was inconsistency between the Tribunal’s conclusion in relation to return on debt and its conclusion on return on equity. The AER said that it had made submissions to the Tribunal on this matter which the Tribunal never reconciled or referred to. The AER also pointed to the fact that the electricity network respondents had themselves relied on evidence concerning foreign regulated firms as additional comparators, as a further indication that the electricity network respondents had not submitted before the AER that the benchmark efficient entity should be unregulated.

528    The AER made other challenges to certain propositions attributed to the electricity network respondents. It is not necessary for us to descend to the detail of the debate on those points.

Analysis and conclusion

529    It is important, once again, to bear in mind the context in which this issue arose for determination by the Tribunal. As we have noted, the AER had adopted the conception that the benchmark efficient entity, referred to in the allowed rate of return objective, was a regulated entity. This was a critical step in the AER’s reasoning and conclusion that, when estimating the return on debt, transitioning from the on-the-day approach to the trailing average approach was required. The Tribunal’s reasons, on this issue, were directed towards the consideration of that step.

530    The Tribunal concluded (at [907]) that the benchmark efficient entity referred to in the allowed rate of return objective was not a regulated entity. This precise finding should be held steadily in mind. It should not be confused with a finding that, for the purposes of the allowed rate of return objective, the benchmark efficient entity must be characterised as an unregulated entity. The competing submissions made by the AER and the electricity network respondents raise what is, on final analysis, a false dichotomy. The allowed rate of return objective does not, in terms, posit the benchmark efficient entity as either a regulated entity or an unregulated entity. Moreover, in order to apply the allowed rate of return objective, there is no need to attribute either characterisation to the benchmark. Our reasons are as follows.

531    First, it is accepted by all parties that the regulatory scheme seeks to obtain the benefits of economic efficiency through mimicking the processes that would occur in a competitive market. Indeed, in promulgating the 2012 Rule Amendments which introduced the allowed rate of return objective, the AEMC noted that the long-term interests of consumers would be best served by ensuring that the methodology used to estimate the return on debt reflected, to the extent possible, the efficient financing and risk management practices that might be expected in the absence of regulation. The AEMC considered that the most appropriate benchmark to use in the regulatory framework for all service providers was the efficient private sector service provider.

532    Further, it may be accepted that, for the purpose of estimating the return on debt, the benefits of the competitive market are seen to be realised by and in the benchmark efficient entity itself, against whose performance the allowed revenue of the service provider is to be determined (i.e., benchmarked).

533    Thus, to this extent, there is, perhaps, something to be said for the AER’s criticism of the Tribunal’s statement at [908] that the AER is required to make its regulatory determination in relation to a regulated service provider “as if the relevant provider were operating in a competitive environment” for the services it provides. The “relevant service provider” is not, of course, operating in a competitive environment for the services it provides. Moreover, the required benchmarking is with respect to the efficiency of financing and risk management practices that are to be expected according to the disciplines of a workably competitive market. Nevertheless, we think it perfectly clear that the Tribunal did not deviate from the correct path of viewing the AER’s required decision on this topic as one to be made by reference to the efficient financing costs of a benchmark efficient entity, rather than the actual financing costs of the particular service provider. Indeed, the Tribunal made that very point at [909] of its reasons.

534    Plainly enough, a workably competitive market is not a regulated market. But, viewing the efficient financing costs of a benchmark efficient entity as those that would be incurred in a workably competitive market does not mean that, for the purposes of construing and applying the allowed rate of return objective, the benchmark efficient entity must then be fixed with the character of an unregulated entity.

535    Secondly, the degree of risk by which the efficient financing costs of the benchmark efficient entity are to be estimated must be similar to the risks that apply to the relevant service provider in providing the standard control services. We do not accept the electricity network respondents’ submission that this part of the allowed rate of return objective is directed to the risk of investing in a business of a generalised type, namely “an infrastructure business with long-lived tangible assets that produce long-term stable cash flows” such business also having a “BBB/BBB+ credit rating and a high book to market ratio”. That characterisation is a considerable, and, it seems to us, contrived, embellishment of what r 6.5.2(c) actually says. The risk is, simply, the particular service provider’s risk in respect of the provision of standard control services. The allowed rate of return objective is no more prescriptive than that. The assessment of that risk is a matter for the AER in the first instance.

536    Thirdly, while it is true that the standard control services provided by the service provider are regulated services, this does not mean that, by force of that fact, the benchmark efficient entity must, correspondingly, be fixed with the character of a regulated entity. The service provider is a regulated entity, not the benchmark efficient entity. Nonetheless, the legislative prescription is that the benchmark efficient entity is to be taken as having “a similar degree of risk” as that which applies to the particular service provider in providing its services. Thus, “a similar degree of risk” is to be attributed to the benchmark efficient entity. That degree of risk may be affected by the fact that the service provider’s provision of the relevant services is regulated. But, once again, this does not mean that the benchmark efficient entity must be characterised as a regulated entity.

537    Thus, in our view, it is not appropriate to characterise the benchmark efficient entity as either a regulated or an unregulated entity. The allowed rate of return objective does not do so, and there is no need to do so. The allowed rate of return objective confers on the benchmark its particular, necessary and defining characteristics: it must be efficient and it must face “a similar degree of risk” as that which applies to the particular service provider in question in relation to the provision of standard control services. But the attribution of the relevant “efficiency” (i.e., in respect of financing costs) is to be gauged by the disciplines of a workably competitive market (i.e., an unregulated market).

538    It follows that we do not accept, in their entirety, the submissions of either the AER or the electricity network respondents. We see no judicially reviewable error in the Tribunal’s primary conclusion (at [907]) that the benchmark efficient entity is not a regulated entity. To the extent that the Tribunal concluded, positively, that, for the purposes of the allowed rate of return objective, the benchmark efficient entity must be fixed with the character of an unregulated entity – and there is some suggestion that it might have done so (see, for example, [914]), although this is not entirely clear – we would respectfully consider that particular conclusion, if reached, to be erroneous. Nevertheless, we do not think that any such error then led the Tribunal to ignore the fact that, under the allowed rate of return objective, the benchmark efficient entity is to be taken as having a similar degree of risk as that which applies to the service provider in question. Indeed, the Tribunal’s reasons are replete with references to that requirement.

539    The Tribunal reasoned (at [938]) that the AER’s selection or identification of the benchmark efficient entity as a regulated entity involved the wrong exercise of a discretion about the character of the benchmark efficient entity in all the circumstances and that, as a consequence, the AER’s decision on return on debt was unreasonable in all the circumstances. In our view, the matter can be stated more directly. The AER erred by misconstruing the allowed rate of return objective in r 6.5.2(c) and the Tribunal correctly concluded that this raised a ground of review under s 71C(1)(c) of the NEL.

540    It follows that the grounds of review covered by this contention must be rejected.

541    Having expressed these views, it is not necessary for us to deal with each and every submission made by the parties in support of their respective positions. We do, however, make the following further observations.

542    First, contrary to the AER’s submission, we are not persuaded that there is a necessary inconsistency between the Tribunal’s conclusion (at [916]) that there will not be an identical benchmark efficient entity for all distribution network service providers and its later expressed conclusion (at [921]) that it was not persuaded that the AER had erred by adopting a single benchmark efficient entity for the regulated service providers in question.

543    The first conclusion was a statement that was predicated on the acceptance of particular facts (different service providers will have different degrees of risk in the provision of standard control services) and was referable to all DNSPs as a class. If one service provider’s risk was different to another service provider’s risk, it followed, as a matter of reasoning, that the benchmark efficient entity for the first service provider could not be the benchmark efficient entity for the second service provider because different degrees of risk would need to be attributed to the benchmarks appropriate to those service providers. We can see nothing impermissible in that reasoning or the general conclusion that was expressed (namely, that there will not be an identical benchmark efficient entity for all service providers). Indeed, the statement of the allowed rate of return objective in r 6.5.2(c) contemplates, at least, the possibility of different benchmark efficient entities, as does r 6.5.2(k)(4).

544    The second conclusion was a statement of a different character. It was directed to the AER’s treatment of the benchmark efficient entity in respect of the service providers whose applications for review were before the Tribunal. Further, the statement was made in the context of explaining the error which the Tribunal considered to be operative. As the Tribunal saw it, that error was not necessarily that the AER had adopted a single benchmark efficient entity for the purpose of applying the allowed rate of return objective but, in relation to those service providers, the wrong benchmark efficient entity, namely a regulated service provider.

545    Secondly, we do not think that, in relation to the question presented by this issue, anything can be drawn from the way in which the AER says that it had estimated the return on equity in its final decisions. The point sought to be made by the AER was that there was inconsistency between the Tribunal’s conclusion in relation to the return on debt and its conclusion in relation to the return on equity. Whether there was any inconsistency is, once again, a matter of contention between the parties. As we have noted, the electricity network respondents contended that, in the Tribunal, no party had raised any issue concerning the AER’s determination of the return on equity based on potential differences in the assessment of risk between regulated or unregulated entities. Further, there was no ground of judicial review before us concerning the Tribunal’s determination in respect of the return on equity. Once again, the parties’ competing contentions were not developed in any way that would enable us to determine whether the AER or, alternatively, the electricity network respondents are correct on this particular matter. The AER’s criticism that the Tribunal did not seek to “reconcile” the inconsistency which the AER perceived can be explained by the fact that none of the electricity network respondents attacked the AER’s decision on the return on equity by reference to whether the correct benchmark was applied. It is little wonder, therefore, that the Tribunal did not seek to “reconcile” allegedly inconsistent treatment when an aspect of the necessary comparison was not properly before it for determination.

Whether the Tribunal erred in its construction of r 6.5.2(k)(4) of the NER

The AER’s submissions

546    The starting point for the AER’s submissions was the Tribunal’s finding at [905] that, once one regulated entity had been taken by the AER as the benchmark efficient entity with the adoption of a ten-year progressive introduction of the trailing average approach, some service providers (such as the NSW service providers) would, necessarily, be materially disadvantaged. The AER argued that this finding revealed a misconstruction of r 6.5.2(k)(4) of the NER, which, it said, required it to take into account, when estimating the return on debt, any impacts on a benchmark efficient entity that could arise as a result of changing the methodology used to carry out that estimation. The AER submitted that r 6.5.2(k)(4) required it to compare (a) the position of the benchmark efficient entity if the existing approach (here, the on-the-day approach) were to continue, with (b) the position of the benchmark efficient entity under a “new” approach (here, the trailing average approach).

547    It should be recalled at this point that, at [935] of its reasons, the Tribunal concluded, contrary to AER’s submission, that an actual assessment must be made of (a) the efficient (not just the actual) financing costs of each service provider, as that service provider had responded in its methodology for estimating the return on debt for the prior regulatory period, and (b) an actual assessment of the impacts on those efficient financing costs of that service provider by the changed methodology.

548    In this connection, it should also be recalled that, prior to the regulatory period in question, the NSW service providers had in place a staggered portfolio of fixed rate debt which was amenable to the trailing average methodology. Thus, the Tribunal concluded (at [934]) that the relevant inquiry for the NSW service providers was whether their financing costs were efficient as at the commencement of the new regulatory period. If not, those of the benchmark efficient entity would be applied prospectively. In reaching this conclusion, the Tribunal acknowledged that, even though the NSW service providers held staggered portfolios of fixed rate debt, there remained an issue whether, in each case, that structure was in an efficient form: see at [906], [926], [932], [934] and [942]. The point of present significance is that the Tribunal concluded that no transition was required for the NSW service providers (because they already had in place a financing cost structure that was amenable to the new (trailing average) methodology) and that r 6.5.2(k)(4) did not require otherwise.

549    Before us, the AER argued that the Tribunal had erred because it did not compare the difference between the existing method and a new method (ie, the on-the-day approach v the trailing average approach). The AER submitted that, in fact, the Tribunal had compared, impermissibly, the differences between “two new methods”. The purport of this last-mentioned submission is not clear. We take the AER to be saying no more than that the Tribunal had regard to the fact that, at the commencement of the new regulatory period, the NSW service providers already had in place a staggered portfolio of fixed rate debt that was amenable to the trailing average approach and that that approach, when compared with the on-the-day approach, was a “new method” for estimating the return on debt.

550    The AER also contended that the Tribunal erred in concluding that the adoption of the trailing average approach necessarily disadvantaged some service providers. With respect to these service providers, the AER said that their adverse position was a result of their past financing practices, not the change in methodology that the AER had adopted. The AER said that there was no basis for the Tribunal’s conclusion in this regard which, the AER said, was unreasonable.

The electricity network respondents’ submissions

551    The electricity network respondents contended that r 6.5.2(k)(4) is to be construed as a transitional provision that dealt with the possibility that service providers may have taken steps or ordered their affairs on the basis of the “previous regime”. Based on statements made by the AEMC (some of which are referred to below at [557]-[561]), the electricity network respondents asserted that the previously required on-the-day approach for estimating return on debt caused some businesses to enter into contractual arrangements, such as hedge contracts or floating rate debt contracts, which could not be unwound immediately upon the implementation of the trailing average approach. The electricity network respondents argued that it was to circumstances of this kind that r 6.5.2(k)(4) was directed.

552    The electricity network respondents argued that r 6.5.2(k)(4) had no relevant application to them because they had no floating rate debt or any other hedge contracts that needed to be unwound in face of the adoption of the trailing average methodology to estimate the return on debt. They said that this did not mean that r 6.5.2(k)(4) was otiose, as the AER had contended, because it may have relevance to service providers who had such contracts in place.

553    The electricity network respondents said that the AER’s contention was based on a false premise because:

Nothing about the operation of cl 6.5.2(k)(4) requires that the “efficient financing costs” referred to in the allowed rate of return objective … in cl 6.5.2(c) be calculated by reference to what some hypothetical regulated business would have done in response to a previous regulatory regime, and having regard to some hypothetical hedge contract entered into by the hypothetical regulated business.

554    In this connection, the electricity network respondents said that the term “benchmark”, as used in r 6.5.2(c), was an “efficiency yardstick” for the particular costs under consideration at the relevant time.

555    With respect to the question of whether some service providers had been “materially disadvantaged” (see [546] above), the electricity network respondents argued that the AER’s submission had mischaracterised the Tribunal’s conclusion at [905]. The electricity network respondents said that the AER had misquoted the Tribunal, which did not say that the AER’s decision to change the method of determining return on debt would result in some service providers being materially disadvantaged. The electricity network respondents said that the Tribunal was merely observing that the imposition of the AER’s transition would materially disadvantage some service providers compared to the immediate imposition of the trailing average methodology. The electricity network respondents said that this observation by the Tribunal was “entirely correct” because the NSW service providers had been “penalised” by being given a debt cost “far below” their actual debt costs and the efficient debt costs in a competitive market.

The AER’s reply submissions

556    The AER submitted that the focus of the mandatory consideration in r 6.5.2(k)(4) is the impact of the change in methodology on the benchmark efficient entity, not on the service provider itself. For this reason, it said, the electricity network respondents’ submissions proceeded without consideration to the text of the relevant provision and thus mischaracterised the estimation task under r 6.5.2.

Analysis and conclusion

557    When promulgating the 2012 Rule Amendments (which included r 6.5.2(k)(4)), the AEMC noted that the amendments provided the AER with the flexibility to adopt an approach that would lead to the best estimate of the return on debt to reflect the efficient financing costs of the service provider in question at the time of the regulatory determination. Informed by a report prepared for it by SFG Consulting (SFG), the AEMC accepted that efficient benchmarking service providers may have different efficient debt management practices. When considering the characteristics of a benchmark service provider and how these characteristics might influence assumptions about its efficient debt management strategy, the AEMC said:

The current prevailing market conditions “one-size-fits-all” approach required under the NER, and applied under the NGR, may lead to various mis-matches between the regulatory estimate allowed by the regulator and the actual interest rate exposure of those service providers that employ debt management practices that are not closely aligned with the benchmark assumptions.

558    Concurrently, the AMEC considered whether transition arrangements might be required if there was to be a change in the methodology used by the AER to estimate the return on debt. In this connection, SFG had commented:

Many businesses would only consider a switch to a different method if appropriate transition arrangements could be put in place. For example, a business cannot today go back and borrow at a rate that applied ten years ago. If the regulatory allowance was set on this basis (by not allowing an appropriate transition arrangement), the result will be either a potentially material benefit or loss to the business – and conversely a potentially material loss or benefit for customers. Moreover, an appropriate transition arrangement effectively destroys any incentive or ability for a business to seek to “game” the regulatory allowance by proposing whichever method might result in the highest allowance …

559    When speaking of transition arrangements in its 2012 Rule Determination, the AEMC said:

The purpose of the [transition arrangement] is for the regulator to have regard to impacts of changes in the methodology for estimating the return on debt from one regulatory control period to another. Consideration should be given to the potential for consumers and service providers to face a significant and unexpected change in costs or prices that may have negative effects on confidence in the predictability of the regulatory arrangements.

It may be possible in many circumstances for the method to estimate the return on debt to take such concerns into account in the design of the method. Therefore, this criterion was intended to promote consideration of concerns raised by service providers with regard to transitions from one methodology to another. Its purpose is to allow consideration of transitional strategies so that any significant costs and practical difficulties in moving from one approach to another is taken into account.

560    There can be no doubt that, in these passages, the AEMC was speaking of actual impacts on individual service providers should the AER decide to change its methodology for estimating the return on debt.

561    In its 2012 Rule Determination, the AEMC specifically noted its intention that the AER could adopt more than one approach for estimating the return on debt having regard to the risk characteristics of benchmark efficient service providers. The AEMC also commented that:

… the return on debt estimate should reflect the efficient financing costs of a benchmark efficient service provider. It should try to create an incentive for service providers to adopt efficient financing practices and minimise the risk of creating distortions in the service provider’s investment decisions. If a service provider is run inefficiently then its shareholders, and not its customers, should bear the financial consequences of inefficient financing practices.

562    In the Explanatory Statement issued with respect to the ROR Guideline, the AER accepted that the holding of a portfolio of debt with staggered maturity dates was likely to be an efficient debt management financing practice for a benchmark efficient entity operating under the trailing average approach.

563    Even so, the AER determined that it would use a single definition of a benchmark efficient entity and would also adopt a single approach to estimating the return on debt for all service providers. As we have noted (see [404]-[406] above), one of the defining characteristics of the benchmark entity used by the AER was that it was a regulated entity. One of the AER’s cardinal considerations for arriving at that conclusion was the requirement of the allowed rate of return objective in r 6.5.2(c) that the benchmark efficient entity be taken as having a similar degree of risk as that which applies to the service provider. However, as we have endeavoured to explain, consideration of this requirement does not also require the benchmark efficient entity to be characterised as a regulated entity.

564    As the Tribunal correctly noted at [904]-[905], the AER’s determination that there would be a single definition of a benchmark efficient entity and a single approach to estimating the return on debt from all service providers was a critical step in the AER’s application of r 6.5.2(k) and, hence, its adoption of Option 2 and rejection of Option 4 in estimating the return on debt for all service providers. The AER adopted this approach notwithstanding the AEMC’s (but not, it would seem, the AER’s) acceptance that efficient benchmarking service providers may have different efficient debt management strategies depending on the risks involved.

565    The criticality of this step by the AER is manifest in its further reasoning that if the benchmark efficient entity were considered to be a regulated entity, it would follow that the efficiency of different debt financing practices would need to be considered in the context of the adopted regulatory regime and, specifically, the then adopted approach to return on debt estimation. This necessarily meant that the starting point for the AER in estimating the return on debt for all service providers was a regulated benchmark efficient entity whose return on debt was calculated according to the on-the-day approach. As we have previously discussed, the AER considered that such an entity would hold a staggered portfolio of debt with different terms to maturity to manage refinancing risk. But, importantly, the AER also considered that such an entity would use swap transactions to hedge interest rate exposure for the duration of the regulatory control period. This benchmark was then applied indiscriminately to all regulated service providers.

566    Further, in conformity with its view that there should be a single benchmark efficient entity and a single approach to estimating the return on debt, the AER also considered that, if a transition was needed, it should be implemented by a single transitional method. Accordingly, the AER then reasoned that the regulated benchmark efficient entity would need to unwind its hedging contracts in order to move from the current on-the-day approach to the trailing average approach.

567    The consequence of this reasoning, and the AER’s determination to have a single transition arrangement, was that the efficiency of the financing practices of those service providers who, immediately before the commencement of the new regulatory period, had adopted a staggered portfolio of fixed rate debt without hedging, was to be assessed as if they had entered into hedging contracts and needed to transition out of those arrangements over an extended period to reach a state of efficiency which, for them, already existed.

568    In adopting this single transition approach, the AER expressly eschewed the notion that transitional arrangements should be specific to individual service providers’ debt financing practices. The AER accepted that the debt financing practices of individual service providers might inform its view as to what were the efficient financing costs of (its conception of) the benchmark efficient entity but, having reached a view on that matter, the AER concluded that the debt financing practices of an individual service provider were irrelevant to the question of whether, in estimating the return on debt for that service provider, there was even a need for a transitional arrangement in order to estimate the efficient financing costs for that provider in accordance with the allowed rate of return objective in r 6.5.2(c). In other words, although those service providers had achieved efficient financing costs, at least to the extent of having a staggered debt portfolio amenable to the trailing average approach, the assessment of their financial efficiency was encumbered by a state of affairs (the unwinding of hedge contracts) which, for them, simply did not exist.

569    Having noted these matters, we now turn to consider the terms of r 6.5.2(k)(4).

570    Rule 6.5.2(k)(4) contemplates the possibility that there may not be a single benchmark efficient entity. The rule specifically refers to any impacts on “a” (not “the”) benchmark efficient entity referred to in the allowed rate of return objective in 6.5.2(c). Further, because r 6.5.2(c) is, in terms, conditioned on the degree of risk that applies to the service provider whose rate of return is to be estimated, it too contemplates that there may be different benchmark efficient entities, depending on the risk involved. Importantly, one and the same benchmark efficient entity is engaged by both rules, but, plainly enough, that benchmark must be one that is apposite for the service provider in question.

571    The requirement of r 6.5.2(h) is to estimate the return on debt for a particular service provider in a way that achieves the allowed rate of return objective. In this connection, the undoubted function of r 6.5.2(c) is to provide, as the electricity network respondents said, an “efficiency yardstick” for the particular costs under consideration. This was plainly the AEMC’s intention in the 2012 Rule Determination. While, for this purpose, the benchmark efficient entity is a construct, there is nothing in r 6.5.2(c) which requires one to be oblivious to the actual circumstances, including the debt management practices, of the particular service provider whose debt costs are being benchmarked. It is the efficiency of that service provider’s costs that is to be benchmarked according to what would be required by the disciplines of a workably competitive market. Considered against the yardstick of the efficient financing costs of the benchmark, the service provider’s debt management practices will either be efficient or inefficient. Where inefficiency exists, that inefficiency should not be reflected in the allowed rate of return for that service provider. Only efficient financing costs are to be allowed. But none of this means that, in applying r 6.5.2(c) to estimate the return on debt under r 6.5.2(h), and in considering whether impacts of the kind referred to in r 6.5.2(k)(4) exist that should be taken into account for that purpose, fictions should be imposed when the service provider has, for example, already implemented a debt structure that satisfies a required aspect of the intended benchmark efficiency.

572    At the commencement of the new regulatory period, the NSW service providers already had in place a debt structure that was both their particular response to the previous on-the-day approach to estimating the return on debt and, as it happened, a debt structure that was amenable to the “changed” methodology of the trailing average approach which the AER had decided to adopt. Significantly, in each case, that debt structure was not one complicated by the overlay of hedging contracts that needed to be unwound. The consequence of this was that, when the AER came to consider r 6.5.2(k)(4) in relation to those service providers, there were no impacts (in the form of hedging contracts that needed to be unwound) apposite to the benchmark efficient entity (or each benchmark efficient entity) for each such service provider that could be said to have arisen from the AER’s change in methodology from the on-the-day approach to the trailing average approach. Given that there were no such impacts, there was no need for the AER to take a step which, in the circumstances, r 6.5.2(k)(4) did not require for those service providers, namely a transition to the trailing average approach of estimating the return on debt by adopting a mechanism to unwind hedging contracts, such as that reflected in Option 2.

573    The same conclusion applies with respect to ActewAGL. Even though it held no debt, it was nonetheless necessary for the AER to arrive at an estimate for its return on debt in accordance with the allowed rate of return objective in r 6.5.2(c) so as to achieve the purposes of the return on capital building block. Even so, there was no meaningful, relevant impact on the benchmark efficient entity apposite for ActewAGL that could be said to have arisen from the change in methodology for estimating its allowed rate of return. The occasion or need for a transition simply did not exist.

574    For these reasons, we do not accept that the Tribunal erred in its construction of r 6.5.2(k)(4) of the NER. The Tribunal reasoned (at [938]) that the AER’s exercise of discretion to apply the characteristics of its selected regulated benchmark efficient entity to the transition process in the case of the electricity network respondents was erroneous and that the AER’s decision on the transition process was unreasonable, in all the circumstances. The Tribunal made no judicially reviewable error in so concluding, with the consequence that the grounds of review before the Tribunal under s 71C(1)(c) and (d) of the NEL were made out.

575    It follows that the grounds of review covered by this contention must be rejected.

Whether the Tribunal’s conclusions as to what is required to be done to comply with rr 6.5.2(c) and (k) reveal error

The AER’s submissions

576    This part of the AER’s case carried forward a number of the themes that the AER had advanced in support of its earlier contentions.

577    At [925] of its reasons, the Tribunal had referred to the AER’s transitioning approach, in the case of the NSW service providers (which did not hedge) and ActewAGL (which did not have debt financing) as “obviously artificial”. In light of that finding, the AER submitted that the Tribunal’s reasoning underlying its rejection of the AER’s transitioning approach to these service providers revealed “a misunderstanding of the incentive-based nature of the energy legislation” which “necessarily involves some artificiality”. The AER used the construct of “the benchmark efficient entity” as an illustration of such artificiality. It also cited, as an illustration of artificiality, the need for ActewAGL’s return on capital to be estimated as if it had debt financing costs, even though it did not, in fact, have such costs.

578    Thus, the AER submitted, to the extent that the Tribunal sought to approach the legislation with a view to avoiding “artificiality” by starting with each service provider’s actual debt structure or debt practices, the Tribunal had erred. The AER submitted that this error weakened the incentive basis of the legislation and the achievement of the NEO.

579    The AER then turned to the relevance of considering the benchmark efficient entity as a regulated entity. Here, the AER effectively repeated its submissions on the level of risk that would confront such an entity, including its warning about the possible mismatch of the appropriate return to investors in the benchmark efficient entity should it be taken as having a higher level of risk than the risk it actually faced. The AER argued that the receipt of regulated revenue would generally pose a lower level of investment risk to debt and equity investors and lead to a lower rate of return to a benchmark efficient entity in order to attract such investors. The AER also argued that the conception of the benchmark efficient entity as a regulated entity may affect the financing practices that are to be taken as efficient financing practices. The AER said that where a service provider has regulated revenue, and a benchmark efficient entity has the same form of revenue (so that it has a similar degree of risk), there may be financing practices that provide efficient ways of managing refinancing risks and interest rate risks. The AER said that these considerations led to its detailed analysis in both the ROR Guideline and final decisions about what were, and what were not, efficient financing practices. It said that if the benchmark efficient entity were to be taken as an unregulated entity in a competitive market, then the AER’s whole inquiry, in this regard, particularly in the final decisions, would have been misconceived.

580    Next, the AER referred to the NSW service providers’ reliance on the fact that they had adopted a staggered portfolio of fixed rate debt which, those providers said, was the debt structure that a benchmark efficient entity would have at the end of the AER’s ten year transition period. The AER argued that the error of this approach, reflected in the Tribunal’s reasons and conclusions, was that the NSW service providers’ debt structure was not efficient in light of the on-the-day approach. The AER argued that it was not correct to say that these service providers had adopted the efficient financing practices of a benchmark efficient entity because, on its finding (presumably on the basis of the on-the-day approach), an efficient entity would have managed interest rate risk, which the NSW service providers had not done.

581    All these submissions were directed to the proposition that the Tribunal erred in failing to characterise the benchmark efficient entity as a regulated entity, with the consequence that the Tribunal was also in error in its analysis of what was required to comply with r 6.5.2(c) and r 6.5.2(k)(4) of the NER.

582    Further, the AER argued that, by characterising the benchmark efficient entity as an unregulated entity, the Tribunal had rendered irrelevant (for the purpose of determining the rate of return on debt) any consideration of how a service provider should efficiently structure its financing costs in the face of the regulatory regime and any changes to it.

583    The AER also argued that the Tribunal’s reasons left the AER without any explanation of how it should “address the uncertainty” created by (what the AER said was) the Tribunal’s conflicting approaches to the conceptualisation of the benchmark efficient entity and the implementation of the transition from one methodology to another.

584    The AER also argued that the Tribunal had taken “the risk concept” out of its statutory context. This part of the AER’s submissions was not always easy to follow. Nonetheless, it seems that the AER was seeking to make the following points.

585    First, the allowed rate of return objective did not speak about the degree of risk associated with a service provider’s existing debt structure but the degree of risk that applied to the service provider in respect of the provision of standard control services. The service provider’s business in providing the standard control services was one that was subject to revenue or price regulation. The AER said that this risk related to the compensation required by debt or equity investors to invest in such an entity.

586    Secondly, the AER took issue with the Tribunal’s statement at [934] of its reasons that, in the case of service providers other than the NSW service providers, the relevant inquiry would start with whether, at the start of the new regulatory period, each of their financing costs (including hedging costs) were efficient having regard to their particular degrees of risk. The Tribunal reasoned that, if so, the impacts of the changed methodology would require “the sort of transition process” that was imposed in the final decisions. However, the Tribunal also reasoned that, if not, then the starting point for the transition process would be some refinement to the efficient financing costs within that structure. The AER’s particular concern appears to be that, by this reasoning, the Tribunal commenced with a consideration of the existing debt structure of a particular service provider and treated the AER’s determination of the service provider’s revenue allowance as constrained by the service provider’s existing debt structure rather than the efficient financing costs of a benchmark efficient entity.

The electricity network respondents’ submissions

587    In response to this aspect of the AER’s case, the electricity network respondents relied principally on the submissions they had made on whether the benchmark efficient entity should be seen as a regulated entity for the purposes of the allowed rate of return objective and whether the Tribunal had erred in applying r 6.5.2(k)(4).

588    The electricity network respondents seized on the AER’s comment that if the benchmark efficient entity was an unregulated entity in a competitive market then its (the AER’s) detailed analysis, particularly in the individual final decisions about what were and what were not efficient financing practices, was misconceived. The electricity network respondents treated this as a concession by the AER, and implied that the AER’s analysis of these matters was, in fact, misconceived.

589    The electricity network respondents also seized on the AER’s comment that the NSW service providers’ debt practices had not managed interest rate risk and had exposed them to considerable risk if interest rates dropped. The electricity network respondents said that the AER’s comment was “circular”. They explained this observation by saying that there would only be such a risk if the on-the-day methodology continued to be used because, with the on-the-day approach, the entire debt cost was calculated at prevailing rates immediately before the start of the new regulatory period whereas, with the trailing average approach, only those facilities that had to be renewed in the current year were exposed to prevailing rates. Hence, it was only those facilities that resulted in the return on debt changing. The electricity network respondents argued that the trailing average approach actually matched the position of an unregulated business in a competitive market, which was, of course, the position of the NSW service providers.

Analysis and conclusion

590    The particular matters raised by the AER have been substantially answered by the conclusions we have expressed at [529]-[545] and [557]-[575] above. At the risk of some repetition, we note the following matters.

591    First, in relation to the AER’s submissions recorded at [577]-[578] above, we accept that aspects of the regulatory regime may reflect, in implementation, a degree of what could be described as “artificiality”, although that is not a term which we would advance. We do not think, however, that the Tribunal’s particular comment – that the AER’s imposition on the electricity network respondents of a transition to the trailing average approach was “obviously artificial” – was directed to aspects of the regulatory regime under the NEL or the NER as such. Still less do we accept that this observation by the Tribunal reveals that it had a misunderstanding about the incentive-based nature of the “energy legislation”. The Tribunal’s observation was directed to the AER’s understanding of what was required by r 6.5.2(k)(4) and, more specifically, how the AER had exercised its discretion by imposing a transition on the electricity network respondents in respect of the adoption of the trailing average methodology when, in fact, the NSW service providers already had in place the structure that was amenable to the trailing average approach (subject nevertheless to their financing costs being shown to be efficient), and when ActewAGL did not have debt financing. The Tribunal’s comment was open to it on the facts before it and does not reveal any judicially reviewable error on its part.

592    Secondly, with respect to the AER’s submissions recorded at [579], we repeat our conclusion that the allowed rate of return objective in r 6.5.2(c) does not import the characterisation of the benchmark efficient entity as a regulated entity. It does, however, require the benchmark efficient entity to be taken as having a similar degree of risk as that which applies to the particular service provider in the provision of its standard control services. As we have remarked, this degree of risk may be affected by the fact that the provision of the services is regulated by price control. We do not think that, at the present time, further explication of the allowed rate of return objective is required or desirable. It will be a matter for the AER to determine how the relevant degree of risk is to be assessed and attributed to the apposite benchmark efficient entity in the circumstances of a given case. Moreover, given the relative degree of generality at which the submissions on this topic were advanced, we do not think that it is either necessary or desirable for us to venture any view as to whether the AER’s extensive analysis on risk, in each of the final decisions, was “misconceived”.

593    Thirdly, with respect to the AER’s submissions recorded at [580] above, we repeat our conclusion that the Tribunal did not err in its construction and application of r 6.5.2(k)(4). No transitioning was required in the case of the electricity network respondents by reason of the change in methodology from the on-the-day approach to the trailing average approach to estimate the return on debt. So far as the NSW service providers were concerned, the Tribunal left open the question of whether their portfolios of staggered fixed rate debt were efficient in form. No judicially reviewable error is made out in the Tribunal’s conclusion that, if not, the efficient financing costs of the benchmark efficient entity would apply prospectively.

594    Fourthly, with respect to the AER’s submissions recorded at [582]-[583], we do not accept that, by its reasoning, the Tribunal rendered irrelevant any consideration of how a service provider should efficiently structure its financing costs in the face of the regulatory regime and any changes to it. We accept that it is not clear as to whether the Tribunal made a positive finding that the benchmark efficient entity should be characterised as an unregulated entity. However, it is not necessary, in light of the above discussion, to consider this question further. We are not persuaded that the Tribunal’s analysis of how r 6.5.2(k)(4) should be applied leads to any uncertainty. Further, whether or not the Tribunal’s treatment of return on equity was inconsistent with its treatment of return on debt is not a matter on which we can opine: it is not a question that properly arises in this proceeding.

595    Fifthly, with respect to the AER’s submissions recorded at [584]-[586] above, we accept that the risk referred to in the allowed rate of return objective is the risk that applies to the service provider in respect of the provision of standard control services. However, we are not persuaded that the Tribunal either had, or applied, a different understanding in its analysis and reasoning or in its conclusions.

596    Further, we do not accept that the Tribunal treated the AER’s determination of the service provider’s revenue allowance as constrained by the service provider’s existing debt structure rather than the efficient financing costs of a benchmark efficient entity. It does not seem to us to be in any doubt that the Tribunal accepted that only the efficient financing costs of the appropriate benchmark efficient entity were to be allowed for the return on debt. But it does not follow that consideration of the service provider’s actual financing costs or its particular debt structure are irrelevant when determining what the efficient financing costs of the appropriate benchmark efficient entity might be when determining the allowance on debt for the particular service provider concerned.

Whether the Tribunal failed properly to undertake a review of the AER’s decision on return on debt

The AER’s submissions

597    This part of the AER’s case commenced with the proposition that, as the Tribunal’s review function was to review the decisions of the AER to determine what, if any, reviewable errors were made, it was incumbent on the Tribunal to understand and consider what the AER’s decision-making actually involved – in other words, what findings of fact the AER had actually made and what discretions it had actually exercised. The AER submitted that a “proper understanding” of its findings and reasons was, for the Tribunal, a “jurisdictional pre-requisite”.

598    Starting from these propositions, the AER raised a number of complaints concerning the way in which the Tribunal had recorded its understanding of how the AER had advanced the correctness of its (the AER’s) approach when estimating the return on debt for service providers.

599    The AER directed attention, firstly, to [925] of the Tribunal’s reasons where the Tribunal said that the AER had selected what it (the AER) had considered to be “the preferable or more representative” debt financing structure for a benchmark efficient entity under the on-the-day approach (namely, a portfolio of floating debt that was hedged) and then considered what would have been involved in moving to the trailing average approach.

600    In response, the AER argued that its attempt to identify efficient financing structures did not necessarily involve identifying one structure or even the “preferable” or “representative” structure. The AER argued that it made its assessment by reference to its understanding of efficient financing practices on the basis of a review of “a considerable amount of evidence”. The AER complained that the Tribunal “does not refer to any of this” or identify a ground of review “in respect of what the AER did”.

601    The AER argued that the Tribunal likewise erred when it had said (at [857]) that the AER viewed the benchmark efficient entity to be “of one size and shape only” and when it also had said (at [859]) that the benchmark efficient entity had the characteristics of a “small to medium” entity. The AER said that it had not applied a “one size fits all approach”, although we note that in other parts of its submissions, the AER appeared to embrace the proposition that it had adopted a single benchmark efficient entity for regulated service providers: see, for example, [507] above. Further, the AER said that it did not consider, by its definition, that the financing practices of the electricity network respondents were inefficient. The AER said that, nevertheless, it did consider “size” when assessing what financing practices were efficient and that it had acknowledged that the size of debt borrowings could affect the ability of some service providers to undertake optimum financing practices. However, the AER said that it had not reasoned in the way stated by the Tribunal and that, to the extent that the Tribunal had found error in the AER’s approach, the Tribunal had, itself, erred (even though the Tribunal did not find that any ground of review had been made out in this regard).

602    Secondly, the AER argued that the Tribunal had also mischaracterised the AER’s decision at [931] of the Tribunal’s reasons. In the preceding paragraph of its reasons ([930]), the Tribunal had noted the AER’s contention that the effect of debt transition on a particular service provider was “ultimately a largely irrelevant consideration” and the consideration of whether a transition to the trailing average approach would cause any cost or inconvenience to any of the electricity network respondents was not a mandatory relevant consideration. At [931] of its reasons, the Tribunal recorded, consequently, an argument by the AER that it had made no reviewable error in adopting a transition option on the basis that it would not have any effect on the electricity network respondents. However, the AER said that, in fact, it did not argue that the adoption of its transitioning approach would not have any effect on the electricity network respondents.

603    Thirdly, the AER also took issue with [932] of the Tribunal’s reasons which we have set out at [443] above.

604    With respect to this paragraph, the AER said that the Tribunal had wrongly characterised the AER as having “chosen” a standard from a range of structures. The AER said that the Tribunal’s characterisation ignored its “evidence-based approach” to identifying efficient practices; wrongly said that the methodology of estimating return on debt had been changed to achieve a particular debt structure, when it was not part of the AER’s role to decide what debt structures should be followed by service providers; wrongly stated that the AER had converted one hypothesised debt structure which it considered to be efficient into another debt structure which it considered to be inefficient; wrongly stated that the AER deemed service providers to have a different debt structure from the debt structure they actually had; and criticised the AER for not having regard to the service providers’ actual debt structures. The AER said that while it did not, in fact, have regard to the service providers’ actual debt structures for the purpose of considering the application of r 6.5.2(k)(4), the AER again said that its approach to that rule was not in error.

605    Next, the AER argued that the Tribunal erred by failing to consider the totality of the AER’s reasons for adopting its transitional approach (Option 2) and whether the AER’s decision to adopt that option was sufficiently supported by other reasons it had given.

606    Thus, the AER said, the Tribunal’s approach to return on debt demonstrated “its misunderstanding of its review functions”. The AER also said that, even if the Tribunal was justified in concluding that the AER had erred in its approach to “the benchmark efficient entity”, the materiality of that error, and thus whether that error established a ground of review under s 71C(1) of the NEL, could only be ascertained by having regard to the whole of the AER’s reasoning, which included matters that were independent of its characterisation of the benchmark efficient entity. The AER argued that the Tribunal had not properly undertaken its review function in assessing the materiality of the AER’s purported errors in its overall approach to the return on debt. Thus, the AER said, the Tribunal had failed to properly carry out its statutory function by not engaging with the whole of the AER’s reasons for adopting Option 2.

The electricity network respondents’ submissions

607    The electricity network respondents submitted that the AER’s complaint that the Tribunal had misunderstood the way in which the AER had advanced its case on return on debt was irrelevant having regard to the Tribunal’s ultimate conclusion. In this connection, the electricity network respondents argued that the AER’s “whole inquiry” as to what would have been the efficient response to the previous regime was misconceived.

608    Notwithstanding this general answer to this part of the AER’s case, the electricity network respondents also took issue with specific aspects of the AER’s submission in this regard.

609    First, with respect to [925] of the Tribunal’s reasons, the electricity network respondents said that the AER’s criticisms (see [599]-[600] above) were unfair. The electricity network respondents contended that the AER’s approach (which rejected any approach that was not the optimal response to the previous regulatory regime) did, indeed, lead to the AER identifying one approach only which, in the case of the electricity network respondents, the Tribunal had found to be the incorrect approach.

610    Secondly, with respect to [857] of the Tribunal’s reasons, the electricity network respondents argued that the AER’s criticism (see [601] above) was similarly unfair because the AER had indeed selected only one efficient response to the previous regulatory regime, which was in fact a response adopted only by certain small to medium regulated businesses, and not the larger regulated businesses, such as the electricity network respondents. According to the electricity network respondents, the result was that the NSW service providers’ debt financing practices were, by the AER’s “definition”, inefficient.

611    Thirdly, with respect to [931] of the Tribunal’s reasons, the electricity network respondents argued that the AER’s criticism (see [602] above) was mistaken. The electricity network respondents said that, at [930]-[933] of its reasons, the Tribunal was dealing with the AER’s contention that the effect of debt transition on a particular service provider was irrelevant, not the question of whether, in fact, the AER’s transitioning approach would have an effect on the electricity network respondents.

612    Fourthly, with respect to [932] of the Tribunal’s reasons, the electricity network respondents argued that the AER’s criticism (see [603]-[604] above) was “wrong in all its points”. The electricity network respondents submitted that each statement made by the Tribunal at [932] was correct and that the AER’s contrary contention was “a particularly egregious misrepresentation of the AER’s own rigid and illogical approach”.

613    Finally, with respect to the AER’s contention that the Tribunal failed to consider the totality of the reasons advanced by the AER for adopting its transitional approach (see [605]-[606] above), the electricity network respondents pointed to the fact that r 6.5.2(b) of the NER imposed a mandatory requirement: the allowed rate of return for a service provider was to be determined so that it achieved the allowed rate of return objective. Once the Tribunal reached the view that, in the case of the electricity network respondents, the AER had not determined the return on debt in accordance with that objective, it was beside the point that it might have preferred the AER’s incorrect approach because of other reasons that had been given by the AER. Thus, the electricity network respondents argued, it could not be said that the Tribunal erred in failing to give consideration to these “other reasons”.

The AER’s reply submissions

614    The AER contended that the electricity network respondents’ submissions did not address what it referred to as the “core” of its complaint, which was that the Tribunal was required to consider the whole of the AER’s reasons for its decision on return on debt before setting it aside. The AER argued that the Tribunal had fastened upon the definition of the benchmark efficient entity without addressing the AER’s other reasons for adopting Option 2, and failed to consider whether those other reasons provided a standalone basis to support the AER’s decision, even in the face of the error that the Tribunal had identified. Thus, according to the AER, the Tribunal had “failed to complete its statutory task”.

Analysis and conclusion

615    Based on the materials to which we were taken in the course of the hearing, we are not persuaded that the Tribunal misunderstood the way in which the AER had approached the task of estimating the return on debt in its final decisions. The AER’s criticisms are in substance directed to the way in which the Tribunal chose to express itself, which may not have accorded with the way in which the AER would have preferred its approach to have been summarised or discussed. However, we do not think that the way in which the Tribunal did describe the AER’s approach or reasoning exposes any judicially reviewable error on the part of the Tribunal. Having come to this view, there is little to be gained by dealing with the minutiae of each of the particular complaints or criticisms made by the AER.

616    Further, we are not persuaded that it was necessary for the Tribunal to deliberate, in its reasons, on all the matters addressed by the AER in its final decisions as to why the AER had adopted its particular approach to estimating the return on debt. We accept the electricity network respondents’ submission that the whole inquiry by the AER as to the efficient response to the previous regime was misconceived. Once the Tribunal had found error in the AER’s approach, based substantially on the AER’s misapplication of r 6.5.2(k)(4) of the NER, that was, effectively, the end of the matter. Once the relevant error had been identified, it was then up to the AER to make its constituent decisions on return on debt in accordance with the Tribunal’s direction. It was not necessary for the Tribunal to be any more explicit as to how the AER should then go about its task.

Conclusion

617    For the reasons which we have given, the AER has not established any of the grounds of judicial review in relation to return on debt.

THE VALUE OF IMPUTATION CREDITS (GAMMA)

The Tribunal’s reasons

618    The reasons of the Tribunal in relation to the value of imputation credits were as follows.

619    It first explained, at [1006]-[1007], that the value of imputation credits received by company shareholders for income tax paid at the company level was recognised by the NER and the NGR in estimating a regulated service provider’s allowed revenue. It cited with apparent approval an excerpt from an AER document as follows:

the estimation of the return on equity does not take imputation credits into account. Therefore, an adjustment for the value of imputation credits is required. This adjustment could take the form of a decrease in the estimated return on equity itself. An alternative but equivalent form of adjustment, which is employed by the NER/NGR, is via the revenue granted to a service provider to cover its expected tax liability. …This form of adjustment recognises that it is the payment of corporate tax which is the source of the imputation credit return to investors.

620    The Tribunal said that the return on equity was not reduced to take into account the value of imputation credits. Rather, the NER (r 6.5.3) and the NGR (r 87A) reduced the revenue that the service provider required to pay the estimated cost of its corporate tax by way of a formula in which the value of imputation credits is represented by the Greek letter γ.

621    At [1026], the Tribunal said the Rules continued to reflect the relationship between imputation credits and the cost of capital in the method of calculating the rate of return. Relevantly, r 6.5.2(d) of the NER provided that:

the allowed rate of return for a regulatory year must be:

(1)    a weighted average of the return on equity for the regulatory control period in which that regulatory year occurs (as estimated under paragraph (f) and the return on debt for that regulatory year (as estimated under paragraph (h)); and

(2)    determined on a nominal vanilla basis that is consistent with the estimate of the value of imputation credits referred to in clause 6.5.3.

(Tribunal’s emphasis in bold.)

622    At [1008], the Tribunal said the Rules required an estimate of “the value of imputation credits” (also referred to as “gamma” or γ) as an input into the calculation of the corporate income tax building block: r 6.5.3 of the NER:

6.5.3    Estimated cost of corporate income tax

The estimated cost of corporate income tax of a Distribution Network Service Provider for each regulatory year (ETCt) must be estimated in accordance with the following formula:

ETCt = (ETIt × rt) (1 – γ)

where:

ETIt is an estimate of the taxable income for that regulatory year that would be earned by a benchmark efficient entity as a result of the provision of standard control services if such an entity, rather than the Distribution Network Service Provider, operated the business of the Distribution Network Service Provider, such estimate being determined in accordance with the post-tax revenue model;

rt is the expected statutory income tax rate for that regulatory year as determined by the AER; and

γ is the value of imputation credits.

(Underlining added.)

623    The Tribunal set out the explanation of the application of the formula in Application by Energex Limited (No 2) [2010] ACompT 7 (Energex (No 2)). The Tribunal said that explanation was to be read subject to the 2012 Rule Amendments, which substituted “the value of imputation credits” for “the assumed utilisation of imputation credits” in the definition of gamma.

624    The Tribunal said the common approach between the parties to the assessment of gamma, expressed as a decimal ratio, was that it be calculated as the product of:

(a)    the distribution rate for imputation credits, expressed as a decimal ratio (also referred to as “F”); and

(b)    the value of distributed imputation credits (also referred to as “theta” or "θ").

625    The Tribunal identified the dispute between the service providers and the AER as being as to:

(a)    the appropriate interpretation of the distribution rate and theta parameters (including what is meant by “the value of imputation credits” in r 6.5.3 of the NER);

(b)    the appropriate method, sources of information and/or weight to be attributed to each data source when determining “the value of imputation credits”; and

(c)    the appropriate figures for each of the distribution rate, theta, and, ultimately, gamma.

626    In particular, the Tribunal said, at [1017], the service providers contended that the AER’s estimate of gamma:

(a)    did not reflect the best estimate of the value of imputation credits to investors, as reflected in market prices that investors were willing to pay for traded stocks; and

(b)    was significantly above even the upper bound for the value of imputation credits, as indicated by tax statistics.

627    The service providers submitted to the Tribunal that if their construction of the NER was correct, the AER erred in the weight it attributed to each source of data and incorrectly determined the appropriate figure for gamma. However, even if the AER’s interpretation was favoured, they said that the AER erred in the weight that it attributed to each data source and incorrectly determined the appropriate figure for gamma.

628    The Tribunal first considered the historical and legislative context.

629    The Tribunal then considered the AER’s approach to setting gamma. The AER rejected the service providers’ proposed value of 0.25 for gamma and adopted a gamma of 0.4 in its Final Decisions based on an analysis of various sources and estimates for both all equity and listed equity.

630    The Tribunal noted, at [1033], the dispute between the parties as to whether the distribution rate should be calculated from all equity, or from the subset of listed equity. The AER’s Final Decisions were based on consideration of the original 0.7 value and a new estimate of 0.8 (0.77 in the case of JGN) derived from the distribution rate based on data only for listed equity. The listed equity only estimate was produced in response to advice from the AER’s expert Associate Professor Handley: Report prepared for the Australian Energy Regulator: Advice on the value of imputation credits, 29 September 2014. He said that the distribution rate should be calculated on a consistent basis with the utilisation rate, theta. This, in turn was based on his understanding of a CAPM framework; consistent with the definition of a BEE, Associate Professor Handley had advised the distribution rate should be estimated only from the data for listed equity.

631    The Tribunal said, at [1035], it appeared that the AER effectively adopted the distribution rate of 0.8 when setting gamma. Whether this was correct or reasonable in the circumstances hinged on the validity of the rationale the AER provided for the emphasis placed on the listed equity estimate of the utilisation rate.

632    As to the utilisation rate (theta), the Tribunal said, at [1036]-[1037], that the AER interpreted theta as “the utilisation value to investors in the market per dollar of imputation credits distributed”, which reflected the extent to which investors could utilise the imputation credits they received to reduce their tax or obtain a refund. Three methods of estimating theta were considered by the AER in the Final Decisions: the equity ownership approach; tax statistics; and market studies. A fourth approach (the conceptual goalposts approach) mentioned in the ROR Guideline was not given any further consideration, but the reasons for excluding it were explained.

633    As to the equity ownership approach, the Tribunal said, at [1039], the AER rejected arguments suggesting that individual eligible investors could value imputation credits at less than their nominal dollar value. Consequently, the equity ownership approach assumed this dollar value of imputation credits to a relevant class of investors and then attempted to estimate the proportion of those investors in the total. The service providers criticised the equity ownership approach on the basis that it made no allowance for the percentage of Australian domestic investors who were unable to redeem an imputation credit because of restrictions on redemption, such as the 45 day holding rule. In its Final Decisions the AER specifically considered the extent to which the equity ownership data should be adjusted for the effect of the 45 day holding rule. The service providers also criticised the equity ownership approach on the basis that it essentially assumed the value of imputation credits rather than deriving it from market data and identified a number of reasons why, in aggregate, Australian resident shareholders will value a dollar of distributed imputation credits at less than a dollar. The service providers also contended that the AER erred by giving more weight to the equity ownership approach to estimating theta (which they said was, at best, an “upper bound” for theta) and that the AER erred by relying on equity ownership rates over a period commencing in July 2000. They submitted to the Tribunal that there was no apparent basis for taking figures up to 15 years old.

634    The Tribunal then considered the AER’s use of Australian Taxation Office (ATO) data and agreed with the Tribunal’s discussion in Energex (No 2), saying, at [1048], that tax statistics could only provide an upper bound on the estimate of theta.

635    The Tribunal next considered, at [1049] and following, market studies of the value of imputation credits, the third source of estimates of the utilisation rate. The Tribunal said, at [1054], the discussion in Energex (No 2) made it clear that the AER was correct not to place much weight in its Final Decisions on the results of early Australian dividend drop-off studies. Nevertheless, its summary of the range of potential theta estimates from these studies failed to focus on what the AER, in its ROR Guideline (Appendices at p 174), noted were “[t]he most relevant dividend drop-off studies, by SFG and Vo et al” which “present estimates in the range 0.35 to 0.55”.

636    The Tribunal turned to estimates of gamma and said, at [1058], without further clarification about which value was used, it was only possible to conclude that, by setting gamma at 0.4, from a range between 0.3 and 0.5, the AER relied upon a value for theta that was one of, or close to, 0.5, 0.52 or 0.57, depending on whether it used a distribution rate of 0.8, 0.77 (for JGN) or 0.7, respectively. All of these values exceeded the upper bound suggested by the tax statistics estimates of the redemption rate (0.43 (updated to 0.45 for JGN)).

637    The Tribunal then, at [1059], turned to the role of gamma in the NER in order to determine the appropriate methodology for calculating gamma. The Tribunal said, at [1061], the proper concern was not the extent to which imputation credits may be translated to real money. Instead, it involved a determination of the cost of taxation to a network service provider, and the extent to which that cost must be reduced to reflect the impact of the dividend imputation system on the network service provider. The reduction in the cost of income tax represented by gamma reflected the personal taxation benefits (as opposed to other benefits such as dividends) gained by shareholders from holding equity in the network service provider and the value of those benefits as ascribed by shareholders. Consequently, it was necessary to consider both the eligibility of investors to redeem imputation credits and the extent to which investors determine the worth of imputation credits to them.

638    Having outlined the respective approaches of the parties to calculating gamma, at [1064] and following, the Tribunal said, at [1076], that the recent studies relied on by the AER did not appear to present an empirically robust and internally consistent explanation for the link between the existence of imputation credits and the applicability of the vanilla (weighted average cost of capital) (WACC) that remained the basis for the allowed rate of return defined by r 6.5.2(d)(1) and (2) of the NER and r 87(4)(a) and (b) of the NGR.

639    At [1081], the Tribunal said it did not accept the AER’s approach that imputation credits were valued at their claimable amount or face value (as the AER said in the Final Decisions: the measure is what can be claimed). The Tribunal said the value was not what can be claimed or utilised, but what was claimed or utilised as demonstrated by the behaviour of the shareholder recipients of the imputation credits.

640    At [1084], the Tribunal said that the AER had not satisfied the Tribunal that its conception and estimation methods were consistent with the requirements of the NER and NEL, including the revenue and pricing principles in s 7A and s 24 respectively. That was understandable where the experts themselves, through their recent reports, presented no consistently coherent CAPM framework for the assessment of the components of the cost of capital. There were models with disputed applicability which may or may not be consistent with the application of a vanilla WACC, as required by r 6.5.2(d)(1) and (2) of the NER and r 87(4)(a) and (b) of the NGR.

641    At [1093], the Tribunal said that, leaving aside the issue of whether the AER was correct to assume that eligible shareholders value each dollar of imputation credits at a dollar, the Tribunal considered that the equity ownership approach overstated the redemption rate. The Tribunal said that it agreed with the service providers’ submission that “even on the AER’s own definition of theta (focussing on potential utilisation by eligible investors), equity ownership rates are above the true maximum possible figure for theta”. In the Tribunal’s view the estimates of the redemption rate produced by the equity ownership approach would be useful only, like the upper bound suggested by tax statistics, as a further check on other estimates.

642    At [1095], the Tribunal said the AER’s equity ownership and tax statistics approaches made no attempt to assess the value of imputation credits to shareholders and ignored the likely existence of factors, such as the 45 day rule, which, across all eligible shareholders, reduced the value of imputation credits to those shareholders below the “face” value assumed by the AER. The Tribunal considered these approaches to be inconsistent with a proper interpretation of the Officer Framework underlying r 6.5.3 of the NEL. It was the reason that the theta estimates produced by the equity ownership approach and tax statistics can be no better than upper bounds on the market value of imputation credits.

643    The Tribunal concluded, at [1100], that by placing most reliance on the equity ownership approach and effectively defining the utilisation rate as the proportion of distributed imputation credits available for redemption, the AER adopted a conceptual approach to gamma that redefined it as the value of imputation credits available for redemption. This was inconsistent, the Tribunal said, with the concept of gamma in the Officer Framework for the WACC which underlay the NER and NGR, and with the objective of ensuring a market rate of return on equity by making an adjustment to the revenue allowance for taxation to account for imputation credits.

644    Having considered the estimation of the distribution rate, the Tribunal concluded, at [1106], that it was appropriate to follow past practice rather than, as the AER had done, introduce the listed equity estimate to reflect the views of its expert Associate Professor Handley on the scope of the relevant markets for assessing theta.

645    The Tribunal’s conclusions on the range of possible gamma were stated at [1107]-[1109]. The Tribunal said that if the AER’s equity ownership approach was adopted, other than as a further check on the upper bound on estimates of theta, the AER’s gamma decision (0.4 from a possible range of 0.3 to 0.5) no longer aligned with the estimated ranges of the distribution and utilisation rates. The AER identified that the redemption rate from tax statistics was 0.43 (or 0.45 using updated data). The service providers’ preferred estimate of theta from the updated SFG study was 0.35. These values of theta produced estimates of gamma in the range between 0.25 and 0.30 with the all equity distribution rate in the ROR Guideline and Draft Decisions (0.7), or between 0.28 and 0.34 if using the higher listed equity distribution rate (0.8). Similarly, using the updated redemption rate that applied to JGN, gamma in that determination would either range between 0.25 and 0.32 with the original distribution rate, or between 0.27 and 0.35 when using the higher updated listed equity distribution rate (0.77).

646    The Tribunal said, at [1111], the AER’s decision set a value for gamma which was too high, where the relevant upper bounds for theta should be no more than the ATO statistical data of 0.43 (or 0.45 in the case of JGN).

647    For these reasons the Tribunal set aside the AER decision on gamma. The Tribunal concluded, at [1114]-[1117], that the AER’s error was material. It gave further reasons for the conclusion, having regard to s 71P(2a) and (2b) of the NEL, in the concluding section of its reasons.

The grounds of review relevant to gamma

648    The grounds in the AER’s application relevant to this topic were as follows, omitting particulars:

16    Insofar as the Tribunal has set aside the AER’s decision on the basis of the AER’s conclusions with respect to the valuation of imputation credits (gamma), the Tribunal’s determination involved reviewable error under ss 5(1)(c), (d), (f), and (j) of the ADJR Act, because the Tribunal misconstrued and misapplied s 71C of the NEL and therefore misunderstood the scope and limitations of its role.

17    Insofar as the Tribunal has set aside the AER’s decision on the basis of the AER’s conclusions with respect to gamma, the Tribunal’s determination involved reviewable error under ss 5(1)(c), (d), (e) (in conjunctions (sic) with (2)(g)), (f) and (j) of the ADJR Act, because the Tribunal misconstrued and misunderstood:

(a)    the meaning of the expression ‘value of imputation credits’ in NER 6.5.3; and

(b)    the requirements of the NER and the NEL, including the revenue and pricing principles set out in s 7A of the NEL.

18    Insofar as the Tribunal has set aside the AER’s decision on the basis of the AER’s conclusions with respect to gamma, the Tribunal’s determination involved reviewable error under ss 5(1)(c), (d), (f), and (j) of the ADJR Act, because it failed to consider or address significant aspects of the reasons for the AER’s decision in relation to the valuation of imputation credits.

19    Insofar as the Tribunal has set aside the AER’s decision on the basis of the AER’s conclusions with respect to gamma, the Tribunal's determination involved reviewable error under ss 5(1)(c), (d), (e) (in conjunction with (2)(g) and (h)), (f), (h) (in conjunction with (3)(a)), and (j) of the ADJR Act, because the Tribunal misconstrued the operation of the NER in determining the materiality of the change to gamma.

20    Insofar as the Tribunal has set aside the AER’s decision on the basis of the AER’s conclusions with respect to the valuation of imputation credits, the Tribunal’s determination involved reviewable error under ss 5(1)(c), (d), (e) (in conjunction with (2)(g) and (h)), (f),     (h) (in conjunction with (3)(a)), and (j) of the ADJR Act, because subparagraph (c) of the Tribunal’s determination is not supported or supportable by the reasons for the determination.

The AER’s submissions

649    The AER made the following submissions.

650    First it gave an overview of the grounds relating to the estimated cost of corporate income tax.

651    The AER submitted that the formula that appears in r 6.5.3 of the NER to estimate the cost of corporate income tax operated such that, all other things being equal, a lower value for γ (gamma) led to a higher revenue allowance. The AER, in its Final Decisions, adopted a gamma value of 0.4. The service providers contended before the Tribunal that the AER should have adopted their proposed gamma figure of 0.25.

652    The AER submitted that the NER did not contain any explicit methodology or formula for how gamma was to be estimated, save for requiring the AER to set out, in the ROR Guidelines, the estimation methods, financial models, market data and other evidence the AER proposed to take into account in estimating the value of imputation credits: r 6.5.2(n) of the NER. Imputation credits may be redeemed by eligible investors and used by them to reduce their tax liability or generate a tax refund. They were not, however, transferrable or tradable on any market. This led to inevitable difficulties in seeking to estimate their value. In Application by Energex Limited (Gamma) (No 5) [2011] ACompT 9 (Energex (No 5)), the Tribunal observed (at [45]) that “estimation of a parameter such as gamma is necessarily, and desirably, an ongoing intellectual and empirical endeavour”. In WA Gas Networks, the Tribunal said at [114] that “there is no one value of gamma that may be regarded as universally correct” and that “[o]ne has to consider the academic models and empirical research and decide on a case-by-case basis which outcome is most relevant for the matter at hand.

653    The AER submitted that before the Tribunal, the common approach between the parties to the assessment of gamma was that it was a decimal ratio, calculated as the product of:

(a)    the distribution rate (“F”), being the value of imputation credits distributed as a proportion of the value of the imputation credits generated; and

(b)    the utilisation rate (“theta” or “θ”), being the value of imputation credits distributed as a proportion of their face value: see reasons at [1010]-[1013].

However, there was disagreement as to how each of these sub-parameters, and thus gamma, should be measured.

654    The AER contended that there were a number of reviewable errors disclosed by the Tribunal’s reasoning.

Grounds 16 and 18

655    The Tribunal’s disagreement with the weight placed by the AER on the various methodologies for estimating gamma was not a matter which could establish a ground of review in s 71C of the NEL. In circumstances where there was no consensus amongst experts as to the best approach to estimating gamma, the Tribunal erred in assuming that its preference of a certain estimation technique for gamma – namely, market studies – had the result that the AER’s reliance on other techniques finding support in the expert evidence – namely, the equity ownership approach and tax statistics approach – led the AER to make a decision which was susceptible to review under the NEL. The Tribunal did not identify reviewable error in the AER’s approach to the estimation techniques for gamma.

656    This claimed error was developed as follows.

657    In estimating values for theta, and thus gamma, the AER relied, to differing extents, on figures obtained via the equity ownership approach, tax statistics approach and market value studies, including the SFG Study. The Tribunal, in its reasons, ultimately disagreed with the weighting assigned by the AER to the respective methodologies. Having accepted that the process of finding a value for gamma “is, or may be, a complex exercise depending upon the inference to be drawn from a range of data sources” (at [1082]) and involves “conflicting expert views” and “finely balanced decisions” (at [1118]-[1119]), the Tribunal’s preference of a different methodological/valuation approach to that preferred by the AER (and supported by the expert evidence) did not engage any ground of review under the NEL or NGL. As the Full Court made clear in ACCC v ACT (at [172]-[173]), the AER’s choice as to the most appropriate valuation methodology did not lead to an error of fact, miscarriage of discretion or unreasonable decision for the purposes of s 71C of the NEL. The Tribunal erred in holding otherwise.

658    An additional contention by the AER was that the Tribunal, in taking the view that the “equity ownership” approach and “tax statistics approaches” to estimating gamma could be used only as an “upper bound” or check for the purposes of ascertaining a figure for theta, and that the assessment of theta “must rely on market studies”, failed to consider or address significant aspects of the reasons for the AER’s decision in relation to gamma.

659    Turning to the tax statistics approach, the AER submitted that in its Final Decisions it had regard to statistics on tax returns submitted to the ATO and imputation credits refunded for the purpose of estimating the utilisation rate. It recognised that there were “residual concerns regarding the data” captured in tax statistics, but determined to give weight to the statistics in reaching a point estimate for gamma on the basis of expert evidence, including from Associate Professors Handley and Lally, supporting that approach. The Tribunal disagreed with the reliance placed by the AER on tax statistics. The AER submitted that the Tribunal’s reasons for that disagreement were, with respect, sparse. They were encapsulated at [1048].

660    The AER submitted that the Tribunal’s invocation of the approach to tax statistics taken in Energex (No 2) was insufficient to disclose error in the present case, of a kind identified in s 71C of the NEL, in relation to the AER’s reliance on tax statistics.

661    Notably, the AER submitted, the Tribunal did not find any factual error on the part of the AER in its interpretation of the ATO data that was available. To the contrary, the Tribunal observed (at [1047]) that the service providers had agreed that the AER correctly identified that the redemption rate from tax statistics was 0.43 (or 0.45 using updated data, applicable to JGN only). The Tribunal’s criticism was of the AER’s choice to rely on tax statistics (to obtain a point estimate of gamma) at all.

662    Critically, the AER submitted, the Tribunal’s reliance on Energex (No 2) as precluding the use of tax statistics in estimating gamma was misplaced in circumstances where that case proceeded on the basis of a concession made by the AER, and was necessarily confined to the material that was then before the Tribunal: see Energex (No 2) at [69(b)], [77], [91]. In the present case, the AER provided reasons in its Final Decisions, and before the Tribunal, as to why ATO tax statistics could be used to derive a point estimate for gamma; and as to why the use of such ATO data was not confined to simply providing a “check” or “upper bound” for the figure selected.

663    Associate Professor Handley stated, in a report relied upon by the AER:

To be clear, taxation statistics (like other methodologies) can be used to derive reasonable albeit imprecise point estimates of theta. The previously used “upper bound” terminology in no way invalidates its interpretation or use as a point estimate.

664    As to the “upper bound” nomenclature, the AER acknowledged that, in theory, tax statistics could be used to derive a measure of the total amount of imputation credits utilised by eligible investors to offset tax or to obtain a refund. However, the AER drew attention to the limitations of the ATO data that was available to it, as well as concerns expressed by the experts about its reliability, which led the AER to conclude that the ATO tax statistics did not reflect the total amount of credits in fact refunded to taxpayers. The AER thus contended that the 0.43 figure for theta, derived from an examination of the available tax statistics, was not, in fact, an accurate reflection of the total or maximum extent, across the market, to which imputation credits were redeemed or capable of holding value to investors.

665    The AER submitted that none of these matters were adverted to by the Tribunal, and the Court should find that it failed to consider them. In simply asserting, on the basis of the decision in Energex (No 2), that the estimates given by the tax statistics could be no better than an upper bound for theta (at [1048], [1090], [1095]-[1096]), without engaging in the AER’s contentions to the contrary and the expert evidence as to the use that may be made of the ATO data, the Tribunal failed to lawfully exercise its statutory review functions. It could not, absent consideration of the arguments that were advanced, properly draw any conclusions about the reasonableness or otherwise of using tax statistics to measure theta and thus gamma. The Tribunal’s disagreement with the AER’s reliance on tax statistics involved a merits-based assessment of a methodological choice that was open to the AER, and was not founded on any reviewable error.

666    The AER’s submissions then turned to the Tribunal’s treatment of the equity ownership approach. The AER submitted that the Tribunal’s disagreement with the AER’s reliance on that approach did not rest upon any identified ground of review in s 71C of the NEL or s 246 of the NGL.

667    First, the AER submitted the Tribunal’s disagreement with the AER’s reliance on the equity ownership approach to estimating theta was, in large part, a product of its broad brush and merits-based disagreement with the AER’s reliance on the tax statistics approach. It was uncontroversial before the Tribunal that the theta figure derived from the ATO tax statistics data was 0.43 (or 0.45 for JGN). It was also uncontroversial that the theta figure derived from the equity ownership approach – which the service providers contended to be inapposite to measure the utilisation rate – was higher; the AER found the equity ownership data to give a theta range of 0.38 to 0.55 for listed equity and 0.56 to 0.68 for all equity. The Tribunal, having found that the figure of 0.43 (or 0.45 for JGN) given by the ATO data reflected only an “upper bound” for theta, automatically rejected the equity ownership approach, given that it gave a higher theta value than that so-called “upper bound”: at [1090], [1092], [1093]. That reasoning process did not involve a proper exercise of the Tribunal’s review functions.

668    Secondly, the AER submitted, the Tribunal appeared to have disagreed with the AER’s treatment of the extent to which the 45 day holding rule impacted upon the reliability of the utilisation rate derived via the equity ownership approach. The 45 day holding rule, in general, required domestic investors to hold shares for at least 45 days before they were entitled to redeem associated imputation credits (Income Tax Assessment Act 1997, s 207-145(1)(a); former s 160APHO of the Income Tax Assessment Act 1936). The Tribunal found (at [1042]) that “[i]n its Final Decisions the AER specifically considered the extent to which the equity ownership data should be adjusted for the effect of the 45 day holding rule”. The Tribunal noted that the AER did not attribute any effect from the 45 day holding rule to its calculation of gamma based on the AER’s finding that there was no compelling evidence of a material class of investors who hold shares for less than 45 days (at [1091]). The Tribunal did not identify any factual error or unreasonableness in the AER’s approach in this regard. While the Tribunal identified what it considered to be an “issue” with the AER’s analysis – namely, that there remained a question as to the “extent to which the value of imputation credits is lower as a result of domestic shareholders being unable to use them” (at [1092]) – the mere identification of that concern could not, by itself, invalidate the AER’s reliance upon the equity ownership approach in such a way as to engage a ground of review in s 71C of the NEL or s 246 of the NGL.

669    The AER then turned to the Tribunal’s treatment of market value studies. It submitted that before the Tribunal, the service providers contended that, in deriving its estimate for gamma, the AER ought to have placed exclusive or at least predominant reliance on figures derived from market value studies, and in particular the SFG Study, commissioned by them, which gave a theta value of 0.35. As noted by the Tribunal at [1053], an earlier iteration of the SFG Study, prepared at the direction of the Tribunal in Energex (No 2), was relied upon in Energex (No 5). However, the Tribunal relied on that report in Energex (No 5) in a context where it made clear (at [45]) that its decision was “based on the material before it”, which did not include the voluminous analysis and conflicting expert evidence on gamma that was before the AER and Tribunal in this case.

670    In the present case, the AER submitted, the Tribunal found (at [1096]) that given its finding that the equity ownership and tax statistics approaches were no better than upper bounds, the assessment of theta “must rely on market value studies” and that such studies are “best placed to capture the considerations that investors make in determining the worth of imputation credits to them”. That conclusion was not based on any identification, on the part of the Tribunal, of a relevant ground of error in the AER’s gamma decisions, which found that market studies “can be subject to a number of limitations”, that they “employ complex and sometimes problematic estimation methodologies” and that there was “no consensus among experts on how to separate the value to the market of dividends from the value to the market of imputation credits”. Indeed, the Tribunal itself recognised the limitations of market value studies, at [1050].

671    The AER submitted that the Tribunal’s view that, notwithstanding the accepted weaknesses of market studies, it was necessary to rely upon them for the purposes of ascertaining a value for theta/gamma, amounted to an expression of opinion as to a matter upon which reasonable minds may differ; and in respect of which the AER had formed its own view as a result of its evaluative judgment and synthesis of the expert evidence before it. For example, the AER referred to concerns expressed by Associate Professor Lally that the examination of market prices in dividend drop off studies “will produce an estimate that reflects the tax position, transactions costs and motives of those investors who transact at the relevant times (such as tax arbitrageurs) and these investors may be quite unrepresentative of the entire market”; and to similar views expressed by Associate Professor Handley in relation to the SFG Study.

672    The AER submitted that the Tribunal, in deciding that market studies should be relied upon in preference to other estimation models for gamma, failed to address the AER’s extensive analysis of the utility and reliability of market value studies generally and the SFG Study in particular. In the premises, the Tribunal could not be satisfied that the AER’s treatment of market studies engaged a ground of review enlivening the Tribunal’s jurisdiction to interfere with the final decisions on gamma.

673    The AER next turned to the Tribunal’s treatment of the distribution rate. It submitted that it derived its gamma estimate using parameters for the distribution rate which were based on data for “listed equity” and for “all equity”. This was done on the basis of advice from Associate Professor Handley to the effect that the distribution rate should be calculated consistently with the utilisation rate. The AER submitted that the Tribunal appeared to have disagreed with this approach, although its reasoning on the issue was not entirely clear. The Tribunal said (at [1106]) that “[w]ithout questioning whether the option was open to the AER, the Tribunal on review is not of the view that this is a sufficient explanation for introducing the alternative measure” (ie, the distribution rate figure for listed equity) and that “[a]t present, the Tribunal is of the view that it is appropriate to follow past practice” (ie, using data from all equity to estimate the distribution rate). In so stating, without making any inquiry as to whether the AER’s approach was “open” to it, the Tribunal failed to demonstrate that the AER’s analysis of the distribution rate, and the choice of data set used to measure it, was affected by any ground of review.

Ground 17

674    The next broad submission made by the AER was that the Tribunal erred in its construction of the expression “the value of imputation credits” in the NER, which error was based upon its rejection of the estimation methodologies preferred by the AER. In this respect the AER submitted as follows.

675    The AER submitted the Tribunal set out its understanding of the function of gamma in the regulatory framework at [1061].

676    The AER’s challenge to the Tribunal’s construction of gamma concerned how the Tribunal proceeded to analyse the “extent to which investors determine the worth of imputation credits to them”. The AER submitted that the Tribunal rejected the approach advanced by the AER – namely, “that $1 of dividend, is equal to $1 of capital gains, is equal to $1 of imputation credits” – on the basis that “market-value studies of imputation credits suggest that investors may not value cash dividends and eligibility to reduce their income tax liabilities, equally” (at [1072]). The Tribunal concluded that the value of gamma “is not what can be claimed or utilised, but what is claimed or utilised as demonstrated by the behaviour of the shareholder recipients of the imputation credits” (at [1081]-[1082]). The AER submitted that that process of reasoning inverted the proper analysis.

677    On the Tribunal’s reasoning, the market value studies, which were designed to measure gamma, were instead deployed to define it conceptually. Conversely, the equity ownership approach and tax statistics approaches were sidelined as being inconsistent with the meaning of the expression “value of imputation credits” on the basis of the Tribunal’s views that they measure something different to market value studies. The AER submitted that that process of reasoning erroneously assumed that the “value” being measured by the market value studies was the same value that was referred to in the definition of gamma in the NER which, in effect, begged the question: cf. Tribunal at [1093], [1095], [1096].

678    The AER contended that, if the “face value” of a dollar of imputation credits was reduced by reference to, for example, the transaction costs (at the investor level) involved in their redemption, that would be inconsistent with the approach taken to return on equity, whereby dividends and capital gains received by shareholders are valued at their full face amount and are not so reduced: see Tribunal at [1067]. The AER contended that, for the Officer Framework underlying the NER to have coherent operation, a uniform approach needed to be taken to estimating the different components of the WACC calculation.

679    The AER submitted that the Tribunal took the view (at [1100]) that the AER had impermissibly “adopted a conceptual approach to gamma that redefines it as the value of imputation credits that are available for redemption”. That mischaracterised the AER’s approach, which sought to infer from the relevant data and expert evidence the value obtained from the redemption of available credits. As to measuring that “value”, the Tribunal’s preference (interpolating from what follows at [1101]-[1103]) was for an approach which accounted for the personal (investor level) taxes and costs of redeeming imputation credits. That involved a misconstruction of what the Officer Framework, and thus r 6.5.3 of the NER and r 87A of the NGR based upon it, required.

680    The AER submitted that the Tribunal also considered (at [1084]) that “the AER ha[d] not satisfied the Tribunal that its conception and estimation methods were consistent with the requirements of the NER and NEL, including the RPP”. The Tribunal’s reasons – based as they were on a mischaracterisation of the AER’s conceptual approach – did not however expose, the AER submitted, how the AER’s approach to gamma was said to be inconsistent with the RPP, or conceptually wrong. The Court should find that the Tribunal erred in so finding. The observation by the Tribunal that the experts presented “no consistently coherent CAPM framework for the assessment of the components of the cost of capital” and that there were “models with disputed applicability” did not absolve it from the need to identify and select an approach to gamma which was consistent with the approach taken to rate of return, as advocated by the AER and demanded by r 6.5.2(d)(2) of the NER: cf. at [1084].

Ground 19

681    The AER’s next broad submission was that the Tribunal, erred in concluding that a change to gamma would lead to a materially preferable NEO decision as it misconstrued the operation of the NER and how the value adopted for gamma interrelated with the return on equity.

682    That submission was developed as follows.

683    The AER submitted the Tribunal misconstrued the materiality of correction of the gamma decisions. The Tribunal concluded at [1110]-[1111] that the AER’s decision on gamma should be set aside, on the basis that it set a value of gamma which was too high. This conclusion was expressed by reference to the Tribunal’s view that the upper bounds for theta should have been no more than the number given by the ATO statistical data, being 0.43 (or 0.45 for JGN).

684    Even if the Court assumed – contrary to the AER’s arguments set out above – that the Tribunal correctly exercised its review functions in relation to its treatment of the AER’s reliance on the ATO statistical data – the Tribunal’s decision to set aside the Final Decisions on the basis of the AER’s conclusions with respect to gamma was nonetheless beyond jurisdiction. Pursuant to s 71P(2a)(b) of the NEL, the Tribunal was only empowered to set aside the Final Decisions if the Tribunal was satisfied that this would lead to a “materially preferable NEO decision”. However, the Tribunal’s reasons disclosed that it misconstrued the materiality of the change to gamma to the overall revenue allowances.

685    The Tribunal recognised (at [1026]) that the NER “continue[s] to reflect the relationship between imputation credits and the cost of capital in the method of calculating the rate of return”. That relationship was encapsulated in r 6.5.2(d)of the NER.

686    While correctly acknowledging the existence of an interconnection between gamma and the return on equity building block, the Tribunal misconstrued the effect of that interrelationship for the purpose of determining whether setting aside the AER’s gamma value would lead to a materially preferable NEO decision.

687    The Tribunal did not undertake any analysis of how the downward change in gamma might affect the overall revenue allowance of the service providers, which it needed to do in order to ascertain the materiality of the errors it identified with respect to gamma to the Final Decisions as a whole. Rather, the Tribunal referred at [1114] to the consequences, in dollar terms, and “without allowing for interrelationships with other issues”, of varying gamma from the AER’s 0.4 to the service providers’ preferred 0.25. The Tribunal considered that the amounts involved were, “of themselves, clearly significant”.

688    In circumstances where the overall return to equity investors is comprised of the value obtained through capital gains, dividends and imputation credits, and the NER and NGR – together with the AER’s revenue determinations – made provision for an interrelationship between those values, the AER submitted it was erroneous for the Tribunal to assume that correction of any error with regard to gamma would lead to a materially preferable NEO decision on the basis that the “impact of changing gamma ‘is what it is’”: cf. at [1117].

689    Nor was there any basis for the Tribunal to suggest that the materiality of the effects of a change in gamma should be increased as a result of the interaction between gamma and the rate of return on equity: cf. at [1116]. Rather, the Tribunal ought to have recognised that a downward change in gamma – and corresponding increase in the revenue allowance for the tax building block – would lead to a reduced market risk premium and rate of return on equity – and corresponding reduction in the revenue allowance for that building block. The interdependence of the market risk premium parameter in the return on equity building block and the gamma parameter in the corporate tax building block was described in the AER’s decisions on gamma and rate of return, the material referred to therein, and was also identified before the Tribunal as follows:

when the rate of return WACC parameters are determined, the market risk premium is adjusted to have regard to imputation credits, and in this sense a high theta leads to more money for the [network] applicants.

690    The AER’s approach to the interrelationship just referred to was not subject to any criticism by the Tribunal; indeed the service providers and their expert clearly appreciated that a lower gamma led to a lower market risk premium estimate. The existence of that relationship meant that any change to gamma would necessarily have “knock on” effects for other building blocks in the revenue allowance. Without quantifying the materiality of such effects, it was not open for the Tribunal to conclude that the materiality threshold in s 71P of the NEL for the setting aside of the Final Decisions in respect of gamma had been satisfied.

Ground 20

691    The AER submitted there was a significant disconnect between the Tribunal’s reasons in relation to gamma and paragraph (c) of its Determination: its final determination in respect of gamma was not supported by its reasons.

692    Section 71P(2c)(a) of the NEL and s 259(4c)(a) of the NGL imposed a duty on the Tribunal, where it made a determination to set aside the AER’s revenue decision and remit the matter back to the AER, to specify how it had taken into account the interrelationship between the constituent components of the AER’s decision and how these related to the grounds of review. This duty to give reasons necessarily required the provision of reasoning which supported the Tribunal’s determination on each constituent component of the revenue decision that was the subject of a ground of review (and any relevant interrelationships).

693    At [1103] the Tribunal stated that (emphasis added):

The Tribunal … provisionally considers that the best estimate of theta [as] derived by the updated SFG Study is 0.35. That provisional view is subject to the “Conclusions” section of this part of the Tribunal’s reasons.

694    The Tribunal later stated, in the “Conclusions” section of its reasons on gamma, that the Tribunal considered that the AER’s decision on the gamma topic should be set aside on the basis that it set a value for gamma which was too high, and was based on a value for theta which was above what the Tribunal considered to be an “upper bound”: at [1110]-[1111]. The Tribunal considered that the reduction of theta below the asserted “upper bound” involved a tension which “requires careful balancing”: at [1113]. In this regard, the Tribunal noted that the “SFG 2013 Study represents one point of view”, that there were “conflicting expert views” and that there were “finely balanced decisions to be made in that light” at [1118]-[1119]. The Tribunal indicated that the AER would be required to consider the competing expert views upon any remitter. In the final result, then, the “Conclusions” section of the Tribunal’s reasons on gamma do not express any concluded view on the correct estimate for theta, and thus gamma.

695    In circumstances where the Tribunal had left “finely balanced” decisions to the AER, including in relation to the selection between the experts”, its reasons did not support the adoption of a theta of 0.35 or a corresponding gamma of 0.25. The Tribunal’s Determinations were, consequently, inconsistent with s 71P(2c) and s 259(4c), and entailed an unreasonable and uncertain exercise of power which was beyond jurisdiction.

The electricity network respondents submissions

696    The respondents submitted that the Tribunal did not simply come to a different view as to the appropriate weighing of available estimation methodologies. Rather, the Tribunal started with an analysis of what was required to be measured under the NER, and then assessed the methodologies relied on by the AER in light of those requirements. The Tribunal concluded that the AER erred by giving primary weight to methodologies that were not capable of measuring what was required to be measured under the NER. The electricity network respondents submitted that the Tribunal was well aware that mere differences of view over legitimate choices to be made between properly available methodologies did not establish a ground of review. However, the Tribunal concluded that the AER’s choice of methodology for estimating gamma was neither permitted nor a legitimate choice because it was premised on a construction of the law that was incorrect.

697    The Tribunal gave careful consideration to the interpretation of “the value of imputation credits” in r 6.5.3 of the NER (r 87A of the NGR). It found that the AER’s conceptual approach to gamma, which redefined gamma as the face value of imputation credits available for redemption, was inconsistent with the WACC framework underpinning the NER and with the objective of ensuring a reasonable overall return to investors.

698    The Tribunal then considered the various methodologies relied on by the AER. It was only market value studies that accounted for all five factors that affected how investors valued imputation credits, those factors being:

1.    not all imputation credits created when companies pay tax are distributed, because some company profits are not paid out in dividends but are reinvested in the business;

2.    foreign investors are unable to redeem imputation credits that they receive;

3.    some domestic investors are unable to redeem imputation credits, for example due to the 45-day holding rule;

4.    some domestic investors eligible to redeem imputation credits do not redeem them; the cost or administrative burden for some shareholders may deter redemption;

5.    some investors who do redeem imputation credits may not value them at their full face value.

699    By contrast, the methodology given most weight by the AER, the equity ownership approach, only accounted for two of those factors. As a consequence, the AER was not simply weighing alternative methodologies designed to measure the same thing. The equity ownership approach was simply a ratio of domestic investors, rather than domestic investors who utilised imputation credits. It therefore did not measure even the AER’s conception of the value of imputation credits. Thus the Tribunal identified the problem as being that the AER’s misinterpretation of the NER requirements caused it to take an erroneous approach to the various methodologies. The Tribunal found that the AER erroneously gave little or no weight to the one methodology capable of accounting for all five factors when it should have recognised that all five factors were relevant to its task under the NER. The AER erred by giving most weight to the equity ownership approach, which was inconsistent with the NER. By proceeding on an incorrect interpretation of the NER, the AER engaged in an incorrect exercise of discretion.

700    The respondents then turned to the specific complaints made by the AER about the Tribunal’s approach to assessing the relevant methodologies.

701    As to tax statistics, it was submitted that although the Tribunal came to the same conclusion as in Energex (No 2), it did so independently and through careful analysis. The Tribunal first observed that the tax statistics estimate assumed a dollar value for each dollar of imputation credits issued and did not reflect any of the factors which may decrease the value of imputation credits to shareholders. The Tribunal then considered the proper role of the tax statistics approach, beginning with an analysis of what was required under the NER. It concluded that the tax statistics approach did not measure what needed to be measured.

702    The respondents took issue with the AER’s assertion that the Tribunal failed to address significant aspects of the AER’s reasoning for not treating the tax statistics as upper bounds. They submitted that a fair reading of the Tribunal’s reasons indicated that the Tribunal did not ignore the AER’s submissions but rather had regard to the answering submissions.

703    The Tribunal’s key conclusion was that tax statistics were measuring, whether or not reliably, the wrong thing and not taking into account matters that would lower the correct calculation of gamma.

704    The AER’s submission that the Tribunal failed to engage with the AER’s expert evidence as to the use that could be made of tax statistics failed to recognise that that expert evidence was based on the same misconception as to the meaning of gamma that, the Tribunal concluded, invalidated the AER’s overall approach.

705    As to the equity ownership approach, contrary to the AER’s submission that the Tribunal “automatically” rejected the approach because it returned higher figures than the tax statistics approach, the respondents submitted that the Tribunal rejected the approach as a point estimate for theta principally because it did not measure what needed to be measured under the NER. Further, even if the AER’s interpretation of “the value of imputation credits” had been correct, the Tribunal found that the equity ownership approach would return figures above the true maximum for theta, because it failed to measure the impact of the 45-day holding rule. The Tribunal’s conclusion that the equity ownership approach was inappropriate because it did not measure the right thing was not dependent on its reasoning regarding the impact of the 45-day holding rule. Rather, the Tribunal concluded that both the equity ownership approach and tax statistics were measuring the wrong thing, and in particular were not measuring the value to investors of imputation credits.

706    In any event, the respondents submitted, the Tribunal correctly rejected the AER’s approach to the 45-day holding rule, which assumed it had no effect on the ability of domestic investors to redeem credits. The Tribunal observed that there were three problems with the AER’s approach and set out its reasons therefor at [1092] and [1093].

707    As to market value studies, the respondents submitted that the Tribunal’s conclusion was based on its interpretation of r 6.5.3 of the NER and on its conclusion that market value studies were the only evidence before the AER that actually measured what that rule required to be measured, being the market value of imputation credits.

708    In any event, the submission that the Tribunal failed to address the AER’s extensive analysis of the weaknesses of market value studies was wrong because the Tribunal recognised, at [1052], that the AER “correctly identified a number of weaknesses in the market studies”. But that did not invalidate the Tribunal’s conclusion. Moreover, as noted by the Tribunal, the particular study relied on by the electricity network respondents was an updated version of the study reported and relied upon in Energex (No 5) which was the best available market value evidence.

709    As to the AER’s criticism of the Tribunal’s approach to the distribution rate, the respondents submitted it was unnecessary for the Tribunal to find error in the AER’s treatment of the distribution rate because it had already found material error in its assessment of theta and therefore in its ultimate conclusion on gamma. The Tribunal found that the AER’s estimate of gamma was erroneous regardless of which approach to the distribution rate was used, meaning that it was unnecessary to find error in the AER’s treatment of the distribution rate.

710    In ultimately directing the AER to use a gamma equal to 0.25, the Tribunal used a distribution rate based on data from “all equity” and explained why this was reasonable, at [1106]-[1107].

711    The submissions on behalf of the respondents then turned to the AER’s contention that the Tribunal misconstrued the meaning of the “value of imputation credits” in r 6.5.3 of the NER.

712    They submitted that, contrary to the AER’s contention, the Tribunal did not define gamma by using market value studies. It construed r 6.5.3 and that led the Tribunal to distinguish between the value that had to be measured under r 6.5.3 and, at [1082], how that value was to be assessed or measured. It was the Tribunal’s interpretation of the word “value” that meant that market value studies must be relied on when assessing the value of imputation credits, not the other way around.

713    As recognised by the Tribunal, the AER’s approach to gamma was flawed. The approach to assessing the required return on equity is, correctly, a market-based approach and it therefore incorporated all of the matters that cause investors to require a particular return.

714    As recognised by the Tribunal, the AER’s approach gave rise to an inconsistency between the assessment of the return on equity (a market-based measure) and the deduction for the “value of imputation credits”, which on the AER’s approach is not the value at all, but is the utilisation rate, or the face amount of utilised credits, not their value to the investor. The AER’s approach was internally inconsistent with the approach to the calculation of other parameters and building blocks and inconsistent with the NEO. It was also inconsistent with the conventional meaning of the words used in r 6.5.3, the value of imputation credits.

715    It followed that the AER’s contention was incorrect, that contention being that if the “face value” of a dollar of imputation credits was reduced by reference to the transaction costs involved in their redemption, that would be inconsistent with the approach taken to the return on equity, “whereby dividends and capital gains received by shareholders are valued at their full face amount”. Dividends and capital gains are not valued at their full face amount.

716    The electricity network respondents submitted that the AER’s reliance on expert opinion to justify its gamma interpretation was problematic, not only because the relevant person was not an expert in economics or economic regulation, but also because expert opinions cannot be used for determining statutory meaning: Pilbara Infrastructure [2011] FCAFC 58; 193 FCR 57 at [60].

717    The respondents submitted that when the correct principles of statutory construction were applied, it may be seen that the Tribunal’s interpretation of r 6.5.3 did not involve any reviewable error. The expression “value of imputation credits” referred to the value to investors of the imputation credits distributed by the business, where “value” took its ordinary meaning as in “worth”. The AER sought to read the expression to mean the “before personal tax and before personal cost value of imputation credits”.