AUSTRALIAN COMPETITION TRIBUNAL

Application by ActewAGL Distribution [2017] ACompT 2

IN THE AUSTRALIAN COMPETITION TRIBUNAL

THE TRIBUNAL DETERMINES THAT:

1.        The reviewable regulatory decision, being the Final Decision: United Energy distribution determination 2016 to 2020, is affirmed.

THE TRIBUNAL NOTES: the AER’s error in calculation identified at [376] of its reasons herein and, in accordance with [377] of those reasons, leaves it to the AER to determine the appropriate response to its error.

IN THE AUSTRALIAN COMPETITION TRIBUNAL

THE TRIBUNAL DETERMINES THAT:

1.    The reviewable regulatory decision, being the Final Decision: ActewAGL Distribution Access Arrangement 2016 to 2021, is affirmed.

THE TRIBUNAL NOTES: the AER’s error in calculation identified at [376] of its reasons herein and, in accordance with [377] of those reasons, leaves it to the AER to determine the appropriate response to its error.

IN THE AUSTRALIAN COMPETITION TRIBUNAL

THE TRIBUNAL DETERMINES THAT:

1.    The reviewable regulatory decision, being the Final Decision: Jemena distribution determination 2016 to 2020, is affirmed.

THE TRIBUNAL NOTES: the AER’s error in calculation identified at [376] of its reasons herein and, in accordance with [377] of those reasons, leaves it to the AER to determine the appropriate response to its error.

REASONS FOR DETERMINATION

THE TRIBUNAL:

INTRODUCTION

1    On 16 June 2016, ActewAGL Distribution (ActewAGL) lodged with the Australian Competition Tribunal (the Tribunal) an application for leave to apply for review and an application for review of the final decision by the Australian Energy Regulator (AER) entitled Final Decision - ActewAGL Distribution Access Arrangement 2016 to 2021 for the access arrangement period comprising 1 July 2016 to 30 June 2021. The Final Decision was published on 26 May 2016, together with the access arrangement for the ACT, Queanbeyan and Palerang gas distribution network 1 July 2016 - 30 June 2021.

2    On 16 June 2016, United Energy Distribution Pty Limited (United Energy) lodged with the Tribunal an application for leave to apply to the Tribunal for review of a reviewable regulatory decision and an application for review of a reviewable regulatory decision of the AER being the Distribution Determination entitled Final Decision - United Energy distribution determination 2016 to 2020.

3    On 16 June 2016 each of Jemena Electricity Networks (Vic) Ltd (Jemena), CitiPower Pty Ltd (CitiPower), Powercor Australia Ltd (Powercor) and AusNet Electricity Services Pty Ltd (AusNet) filed their applications.

4    On 26 August 2016, the Tribunal granted these applicants leave to apply for a review of these decisions of the AER published on 26 May 2016 in respect of each of the grounds for review set out in their applications.

5    By notice dated the same day, the Minister for Energy, Environment and Climate Change for the State of Victoria (the Minister) intervened in the Victorian matters, ACT 3, 4, 5, 7 and 8 of 2016 (see in particular [6(iv)] and [6(vi)] below).

6    The matters raised were in summary as follows:

(i)    return on debt (transition): the AER’s decision to reject the relevant applicants’ (ActewAGL and Jemena) proposed trailing average approach (with no period of transition) to estimating the return on debt for use in the determination of the allowed rate of return for the 2015/16 year and the 2016-21 access arrangement period (together, 2015-21 period), and to instead determine the relevant applicants’ return on debt using an approach that involved a form of transition from the previous 'on-the-day' approach to the trailing average approach;

(ii)    value of imputation credits (gamma): the AER’s decision to reject the relevant applicants’ (ultimately only ActewAGL and Jemena) proposed value of imputation credits of 0.25 for use in the estimation of the cost of corporate income tax, and to instead determine that the value of imputation credits ought to be 0.4;

(iii)    forecast inflation: the AER’s decision to reject ActewAGL’s proposal to adopt a breakeven method estimate of inflation for the 2015-21 period of 1.96 per cent, and to instead determine that forecast inflation ought to be 2.18 per cent (United Energy withdrew this ground of its application);

(iv)    advanced metering infrastructure (AMI) rollout: the AER’s decision the subject of the Minister’s intervention in the Tribunal and affecting all applicants except ActewAGL;

(v)    labour price growth rates: the AER’s decision the subject of an application to the Tribunal by only CitiPower and Powercor;

(vi)    corporate overhead expenses: the AER’s decision the subject of the Minister’s intervention in the Tribunal and affecting only CitiPower and Powercor;

(vii)    return on debt (BVAL curve): the AER’s decision the subject of an application to the Tribunal only by AusNet; and

(viii)    forecasting opex/self-insurance: the AER’s decision the subject of an application to the Tribunal only by AusNet.

7    The grounds raised by ActewAGL were under the National Gas Law (NGL) and National Gas Rules (NGR). The grounds raised by the other applicants were under of the National Electricity Law (NEL) and under the National Electricity Rules (NER). Where the laws and rules governing electricity and gas networks are effectively the same we have dealt with them together, with reference where relevant to variations or additions under the corresponding Law or Rules.

8    For convenience, the Tribunal is giving today three sets of reasons structured so that issues common to more than one applicant are dealt with together. For example, the discussion of the return on debt grounds in this decision provides the basis for the conclusions on that issue in relation to Jemena. There were also common submissions in relation to the value of imputation credits (gamma).

9    In summary, these reasons deal with the issues in [6(i)]-[6(iv)] above, that is, all except the following: [6(v)] labour price growth rates and [6(vi)] corporate overheads, where the reasons are to be found in the separate CitiPower and Powercor decisions given today: Applications by CitiPower Pty Ltd and Powercor Australia Ltd [2017] ACompT 4; and [6(vii)] return on debt (BVAL curve) and [6(viii)] self-insurance, where the reasons are to be found in the separate AusNet decision also given today: Application by AusNet Electricity Services Pty Ltd [2017] ACompT 3.

THE STATUTORY STRUCTURE

10    The grounds for review are set out in s 246(1) of the NGL and the “materially preferable designated NGO decision” requirement is set out in s 259(4a). The grounds for review have recently been considered by a Full Court of the Federal Court in Australian Energy Regulator v Australian Competition Tribunal (No 2) [2017] FCAFC 79; 345 ALR 1.

11    Under s 246(1) of the NGL, an application under s 245(1) may be made only on one 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 one error of fact in its findings of facts, and 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.

12    Section 259(4a) provided as follows:

In a case where the decision is a designated reviewable regulatory decision, the Tribunal may only make a determination –

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

(b)    to set aside the designated 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 designated reviewable regulatory decision in making a contribution to the achievement of the national gas objective (a materially preferable designated NGO 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 designated 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 decision and remit the matter to the AER to make the decision again.

(Original emphasis.)

13    Section 244 defined “materially preferable designated NGO decision” by reference to s 259(4a)(c).

14    Sections 249(4b), (4c) and (5) should also be reproduced, as follows:

(4b)    In connection with the operation of subsection (4a) (and without limiting any other matter that may be relevant under this Law) –

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

(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 28); and

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

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

(i)    the establishment of a ground for review under section 246(1);

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

(iii)    that the amount that is specified in or derived from the designated reviewable regulatory decision exceeds the amount specified in section 249(2).

(4c)    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 designated reviewable regulatory decision and how they relate to the matters raised as a ground for review as contemplated by subsection (4b)(a); and

(b)    in the case of a determination to vary the designated reviewable regulatory decision – the reasons why it is proceeding to make the variation in view of the requirements of subsection (4a)(d).

(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 original decision maker.

15    The “national gas objective” is defined in s 2(1) to mean the objective set out in s 23, which provides as follows:

National gas objective

The objective of this Law is to promote efficient investment in, and efficient operation and use of, natural gas services for the long term interests of consumers of natural gas with respect to price, quality, safety, reliability and security of supply of natural gas.

16    Section 249(2) states that the amount that is specified in or derived from the decision exceeds the lessor of $5,000,000 or 2 per cent of the average annual regulated revenue of the covered pipeline service provider.

17    Time limits on applicants were imposed as follows: by s 247, an application under s 245 in respect of a reviewable regulatory decision of the present kind must be made no later than 15 business days after the reviewable regulatory decision is published in accordance with the NGL or the NGR.

18    The provisions of the NGR centrally relevant to the AER’s decision were as follows:

Division 3     Building block approach

76     Total revenue

Total revenue is to be determined for each regulatory year of the access arrangement period using the building block approach in which the building blocks are:

(a)    a return on the projected capital base for the year (See Divisions 4 and 5); and

(b)    depreciation on the projected capital base for the year (See Division 6); and

(c)    the estimated cost of corporate income tax for the year (See Division 5A); and

(d)    increments or decrements for the year resulting from the operation of an incentive mechanism to encourage gains in efficiency (See Division 9); and

(e)    a forecast of operating expenditure for the year (See Division 7).

…

Division 5     Rate of return

87     Rate of return

(1)    Subject to rule 82(3), the return on the projected capital base for each regulatory year of the access arrangement period is to be calculated by applying a rate of return that is determined in accordance with this rule 87 (the allowed rate of return).

(2)    The allowed rate of return is to be determined such that it achieves the allowed rate of return objective.

(3)    The allowed rate of return objective is that the rate of return for a 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 service provider in respect of the provision of reference services (the allowed rate of return objective).

(4)    Subject to subrule (2), the allowed rate of return for a regulatory year is to be:

(a)    a weighted average of the return on equity for the access arrangement period in which that regulatory year occurs (as estimated under subrule (6)) and the return on debt for that regulatory year (as estimated under subrule (8)); and

(b)    determined on a nominal vanilla basis that is consistent with the estimate of the value of imputation credits referred to in rule 87A.

(5)    In determining the allowed rate of return, regard must be had to:

(a)    relevant estimation methods, financial models, market data and other evidence;

(b)    the desirability of using an approach that leads to the consistent application of any estimates of financial parameters that are relevant to the estimates of, and that are common to, the return on equity and the return on debt; and

(c)    any interrelationships between estimates of financial parameters that are relevant to the estimates of the return on equity and the return on debt.

Return on equity

(6)     The return on equity for an access arrangement period is to be estimated such that it contributes to the achievement of the allowed rate of return objective.

(7)     In estimating the return on equity under subrule (6), regard must be had to the prevailing conditions in the market for equity funds.

Return on debt

(8)     The return on debt for a regulatory year is to be estimated such that it contributes to the achievement of the allowed rate of return objective.

(9)     The return on debt may be estimated using a methodology which results in either:

(a)     the return on debt for each regulatory year in the access arrangement period being the same; or

(b)     the return on debt (and consequently the allowed rate of return) being, or potentially being, different for different regulatory years in the access arrangement period.

(10)     Subject to subrule (8), the methodology adopted to estimate the return on debt may, without limitation, be designed to result in the return on debt reflecting:

(a)     the return that would be required by debt investors in a benchmark efficient entity if it raised debt at the time or shortly before the time when the AER's decision on the access arrangement for that access arrangement period is made;

(b)     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 access arrangement period; or

(c)     some combination of the returns referred to in subrules (a) and (b).

(11)     In estimating the return on debt under subrule (8), regard must be had to the following factors:

(a)     the desirability of minimising any difference between the return on debt and the return on debt of a benchmark efficient entity referred to in the allowed rate of return objective;

(b)     the interrelationship between the return on equity and the return on debt;

(c)     the incentives that the return on debt may provide in relation to capital expenditure over the access arrangement period, including as to the timing of any capital expenditure; and

(d)     any impacts (including in relation to the costs of servicing debt across access arrangement 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 access arrangement period to the next.

(12)     If the return on debt is to be estimated using a methodology of the type referred to in subrule (9)(b) then a resulting change to the service provider's total revenue must be effected through the automatic application of a formula that is specified in the decision on the access arrangement for that access arrangement period.

Rate of return guidelines

(13)     The AER must, in accordance with the rate of return consultative procedure, make and publish guidelines (the rate of return guidelines).

(14)     The rate of return guidelines must set out:

(a)     the methodologies that the AER proposes to use in estimating the allowed rate of return, including how those methodologies are proposed to result in the determination of a return on equity and a return on debt in a way that is consistent with the allowed rate of return objective; and

(b)     the estimation methods, financial models, market data and other evidence the AER proposes to take into account in estimating the return on equity, the return on debt and the value of imputation credits referred to in rule 87A.

(15)     There must be rate of return guidelines in force at all times after the date on which the AER first publishes the rate of return guidelines under these rules.

(16)     The AER must, in accordance with the rate of return consultative procedure, review the rate of return guidelines:

(a)     at intervals not exceeding three years, with the first interval starting from the date that the first rate of return guidelines are published under these rules; and

(b)     at the same time as it reviews the Rate of Return Guidelines under clauses 6.5.2 and 6A.6.2 of the National Electricity Rules.

(17)     The AER may, from time to time and in accordance with the rate of return consultative procedure, amend or replace the rate of return guidelines.

(18)     The rate of return guidelines are not mandatory (and so do not bind the AER or anyone else) but, if the AER makes a decision in relation to the rate of return (including in an access arrangement draft decision or an access arrangement final decision) that is not in accordance with them, the AER must state, in its reasons for the decision, the reasons for departing from the guidelines.

(19)     If the rate of return guidelines indicate that there may be a change of regulatory approach by the decision maker in future decisions, the guidelines should also (if practicable) indicate how transitional issues are to be dealt with.

Division 5A

87A     Estimated cost of corporate income tax

(1)    The estimated cost of corporate income tax of a service provider for each regulatory year of an access arrangement period (ETC_t) is to be estimated in accordance with the following formula:

ETC_t = (ETI_t × r_t) (1 – γ)

Where

ETI_t 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 reference services if such an entity, rather than the service provider, operated the business of the service provider;

r_t is the expected statutory income tax rate for that regulatory year as determined by the AER; and

γ is the value of imputation credits.

…

Division 6     Depreciation

88     Depreciation schedule

(1)    The depreciation schedule sets out the basis on which the pipeline assets constituting the capital base are to be depreciated for the purpose of determining a reference tariff.

(2)    The depreciation schedule may consist of a number of separate schedules, each relating to a particular asset or class of assets.

89     Depreciation criteria

(1)    The depreciation schedule should be designed:

(a)    so that reference tariffs will vary, over time, in a way that promotes efficient growth in the market for reference services; and

(b)    so that each asset or group of assets is depreciated over the economic life of that asset or group of assets; and

(c)    so as to allow, as far as reasonably practicable, for adjustment reflecting changes in the expected economic life of a particular asset, or a particular group of assets; and

(d)    so that (subject to the rules about capital redundancy), an asset is depreciated only once (ie that the amount by which the asset is depreciated over its economic life does not exceed the value of the asset at the time of its inclusion in the capital base (adjusted, if the accounting method approved by the AER permits, for inflation)); and

(e)    so as to allow for the service provider's reasonable needs for cash flow to meet financing, non-capital and other costs.

(2)    Compliance with subrule (1)(a) may involve deferral of a substantial proportion of the depreciation, particularly where:

(a)    the present market for pipeline services is relatively immature; and

(b)    the reference tariffs have been calculated on the assumption of significant market growth; and

(c)    the pipeline has been designed and constructed so as to accommodate future growth in demand.

(3)    The AER’s discretion under this rule is limited.

Note:

See rule 40(2).

90    Calculation of depreciation for rolling forward capital base from one access arrangement period to the next

(1)    A full access arrangement must contain provisions governing the calculation of depreciation for establishing the opening capital base for the next access arrangement period after the one to which the access arrangement currently relates.

(2)    The provisions must resolve whether depreciation of the capital base is to be based on forecast or actual capital expenditure.

CONSULTATION BY THE TRIBUNAL

19    The consultation process referred to in s 261(1)(b) of the NGL and s 71R(1)(b) of the NEL is an additional step which the Tribunal must take and, ideally, accommodate within the target time prescribed by s 260 of the NGL and s 71Q of the NEL. The Tribunal, having given leave to the applicants to apply for review, sought information from the AER as to groups or persons who might have an interest in the Tribunal’s review under s 261(1)(b) of the NGL and s 71R(1)(b) of the NEL.

20    The Tribunal then invited each of those groups or persons to indicate:

    whether they wished to consult with the Tribunal in relation to the Final Decisions;

    the nature of their proposed participation; and

    how the consultation might best be carried out.

21    Having determined a protocol for the consultation, the Tribunal issued a Consultation Agenda under which it provided for those who wished to speak to the Tribunal, either personally or on behalf of an organisation, to do so.

22    The Tribunal conducted the first consultation session on 6 October 2016 at the Federal Court of Australia in Melbourne. At that time, the Tribunal heard submissions by Ms R Sinclair, the Chief Executive Officer of Energy Consumers Australia Ltd (ECA), Ms T Jelenic for the Public Interest Advocacy Centre (PIAC) and Ms R Held for the Consumer Utilities Advocacy Centre (CUAC).

23    The ECA is an organisation formed by the COAG Energy Ministers Council as part of the energy market reform package of 2012. Its oral submissions were addressed first to the new limited merits review process itself; secondly, to the increased focus in that process on the long-term interests of consumers; and thirdly, to the need for the Tribunal’s decision to be a holistic, materially preferable decision focused on the long-term interests of consumers. The ECA also noted the revenue impacts which totalled $365.3 million, of which 60 per cent was said to be collected from residential and small business consumers. It was submitted that consumers did not think that the price increases that those revenue increases would lead to were in their long-term interests. The ECA had conducted an energy consumer sentiment survey in July 2016 and the survey showed that customers valued reliability and competition in the energy sector but were not at all happy with the value for money they were getting from energy services and they did not think that the sector had their interests at heart. The core question, it was submitted, was whether the businesses had truly considered the long-term interests of their customers in putting forward the revised proposals and now in seeking the Tribunal’s review of the AER decisions with respect to the matters raised. It was submitted that the long-term interests of consumers must be the Tribunal’s paramount consideration in determining that a materially preferable decision exists. Emphasis was placed on the terms of s 71P of the NEL. It was submitted that the long-term interests of consumers were not delivered by any one factor in isolation but rather required balancing of those factors. The preference for one outcome over another was to be determined with reference to the whole outcome and not the individual components of it and materiality was to be determined with reference to the national electricity and gas objectives and nothing else.

24    PIAC addressed in particular s 71R(1)(b) of the NEL, concerning statutory consultation requirements. PIAC sought to support and strengthen the Tribunal’s role in promoting consumer participation by, first, emphasising that s 71R(1)(b) represented an important precondition to the Tribunal’s powers of review and should be construed broadly. Secondly, PIAC drew the Tribunal’s attention to the difficulties faced by consumers in providing useful comment at a time before the parties’ submissions had been lodged and within restrictive time and page limits. Thirdly, PIAC addressed the need for the Tribunal to give adequate consideration to consumer submissions in its final decision. PIAC referred to the national electricity objective in s 7 of the NEL as guiding all decision-making by the AER and the Tribunal, being framed in terms of the long-term interests of consumers. PIAC also referred to s 71P(2)(a). PIAC also relied on its written submissions.

25    The CUAC’s submissions and supplementary submissions were directed to whether CitiPower and Powercor’s proposed growth rates for 2016 represented efficient costs of achieving the opex and capex objectives for the purpose of rr 6.5.6(c) and 6.5.7(c) of the NER. The Tribunal addresses this matter fully in Applications by CitiPower Pty Ltd and Powercor Australia Ltd [2017] ACompT 4.

26    The Tribunal conducted a second consultation session on 11 October 2016 at the Federal Court of Australia in Canberra. The only registered speakers at the Canberra consultation withdrew before the start of proceedings. Consequently, opportunity was provided for those who had not registered to participate to make submissions to the Tribunal. One person took that opportunity. It has not been necessary for the Tribunal to revisit those issues in the course of forming its decisions.

27    In order to address concerns about the brevity of time allocated to user and consumer groups and their inability to respond to the written submissions and replies of the applicants, on 18 October 2016 the Tribunal made a direction giving leave to the user and consumer groups to lodge written submissions by 10 November 2016 in relation to the written submissions lodged by the parties. This leave was taken up.

28    In addition to the organisations already referred to, the Tribunal received and considered written submissions from the St Vincent de Paul Society; the Consumer Action Law Centre; the Ethnic Communities’ Council of NSW Inc; UnitingCare Australia; the Queensland Electricity Users Network; and the New South Wales Irrigators’ Council and Cotton Australia.

SUMMARY OF APPLICANTS’ GROUNDS

29    ActewAGL summarised its grounds as follows.

30    ActewAGL submitted that Pt 8 of the NGR governed the AER’s decisions about whether to approve access arrangements by service providers.

31    It submitted that r 76 of the NGR adopted a building block approach to the determination of total revenue to be derived from access arrangements. The building blocks relevant to the present grounds of review were: r 76(a) (return on debt); r 76(b) (forecast inflation); and r 76(c) (gamma). We have set these out at [18] above.

Return on debt

32    Specifically in relation to return on debt, ActewAGL submitted, in summary as follows.

33    The issue was whether the historical trailing average (HTA) approach should be applied immediately to estimate the return on debt (which both ActewAGL and Jemena proposed) or whether, as the AER decided, the trailing average approach should be phased in over a decade (the transition to trailing average (TTA)).

34    ActewAGL argued that the AER’s decision was predicated on a misconstruction of r 87 of the NGR, equivalent to s 6.5.2 of the NER. The AER’s construction embodied a so-called zero net present value (NPV) criterion over the ‘life of the Regulatory Asset Base (RAB)’; interpreted ‘efficient financing costs’ as relating solely to ‘prevailing’ interest rates (but did not require that those rates be prevailing at any particular time); and did not require any consideration of the financing practices of the benchmark efficient entity (BEE).

35    ActewAGL contended that those concepts could not be reconciled to the text of r 87 of the NGR. Rather, it appeared that the AER had reached what it considered to be the ‘right’ outcome, and then had sought to fit that conclusion into that rule, something ActewAGL argued was not a proper basis for statutory construction.

36    Similarly, Jemena argued the NPV = 0 criterion in the AER’s Final Decision was incorrect. It had no explicit basis in r 6.5.2 and could not override the requirements of that rule. Jemena also argued that the AER’s application of the condition did not reflect a proper NPV = 0 analysis in any event, because the AER’s approach confused the RAB with debt instruments, and because the on-the-day approach only ever produced a random mismatch with the actual cost of debt, and this random mismatch did not produce some consistent value over time.

37    Further, ActewAGL argued that the AER’s financial theory (ie, that the trailing average approach should only be adopted if there was a transition that preserved “revenue neutrality” over the “life of the RAB” with the on-the-day approach) depended on its conclusion that the on-the-day approach necessarily satisfied the allowed rate of return objective (ARORO). That conclusion was untenable. Accordingly, the theoretical premise of the AER’s decision (even leaving aside its construction issues) was flawed.

38    ActewAGL contended that the phrase “efficient financing costs” in r 87(3) of the NGR, equivalent of r 6.5.2(c) of the NER, related to all matters that bore upon the financing costs that the BEE, acting efficiently, would incur during the regulatory period, including its efficient financing practices. It argued this construction was consistent with extrinsic material, the text of the clause of the NGL, and the decisions in Applications by Public Interest Advocacy Centre Ltd and Ausgrid [2016] ACompT 1 (Ausgrid) and Application by EnergyAustralia [2009] ACompT 8 (EnergyAustralia).

39    Jemena also argued that efficient financing costs are actual costs incurred by the benchmark business under its debt instruments (facilities and bonds) and, because the business had contractual commitments to pay interest under those instruments, there was a component of future debt costs which was historically predetermined. Consequently, Jemena argued the AER’s construction of the term “efficient financing costs” as the prevailing cost of debt (at the time of making the distribution determination, or alternatively, at any one particular point in time) was incorrect. Rather, the term “efficient financing costs” in the allowed rate of return objective was a reference to costs that a benchmark service provider would incur adopting efficient financing and risk management practices, including practices that might be expected in the absence of regulation, that is, the costs facing an efficient business. Jemena contended the AER’s construction was therefore inconsistent with the adoption of a trailing average approach, which was not a prevailing cost approach.

40    Jemena further argued that, because interest rates payable on existing debt instruments held by the benchmark business were not at the prevailing rate for new debt, the prevailing rate could not sensibly be applied as a measure for the entire cost of debt and did not represent the efficient financing costs of that business. Alternatively, and unlike both the on-the-day approach and the AER’s transition approach, the trailing average approach calculated the efficient financing costs of a benchmark business for Jemena.

41    ActewAGL had no debt. Therefore, it argued, the change in debt methodology to the trailing average approach had no ‘impact’ on it for the purposes of r 87(11)(d), the NGR equivalent of r 6.5.2(k)(4) of the NER. Given that it was common ground that a BEE would hold a staggered portfolio of debt, and finance (or refinance) 10 per cent of that portfolio each year, the trailing average approach should therefore be applied without a transition.

42    Similarly, although Jemena did have debt, it did not have hedge contracts. In addition to arguing that neither the on-the-day approach nor the AER’s transition produced a measure of the efficient financing costs of a BEE in its position, Jemena contended that r 6.5.2(k)(4) provided a transition for businesses that had changed their position in an irrecoverable way on the basis of the previous regulatory regime (such as by entering into hedge contracts). Jemena was not such a business, consequently, it argued, the AER’s transition was not in accordance with r 6.5.2 of the NER.

43    ActewAGL argued the issues raised by the AER about the proper construction of s 71 of the NEL/s 246 of the NGL regarding the nature of the grounds claimed by the applicants did not arise because the AER had misconstrued the NGR equivalent of r 6.5.2 and had therefore exercised its discretion incorrectly.

44    Jemena further argued that the other reasons provided by the AER for a transition (the avoidance of perceived bias and measurement difficulties of the trailing average approach) were not valid reasons for a failure to apply the ARORO.

45    Jemena also argued that the AER’s change in reasoning was not foreshadowed to the service providers, in contravention of r 11.60.4(k) of the NER and s 16(1)(b) of the NEL.

Gamma

46    In respect of this ground, ActewAGL relied upon submissions at that time advanced jointly on behalf of all network applicants. However, by letter dated 21 July 2017, all applicants other than ActewAGL and Jemena abandoned their grounds of review in respect of gamma. ActewAGL and Jemena contended for a value of 0.25 for gamma rather than the value of 0.4 applied by the AER. We set out their submissions on this topic at [240]-[260] below.

Forecast inflation

47    Specifically in relation to forecast inflation, ActewAGL submitted, in summary as follows.

48    A forecast of inflation was a component of the depreciation building block in r 76(b) of the NGR.

49    Forecast inflation on the capital base must be deducted from the depreciation allowance in each regulatory year to avoid “double counting” of inflation (given that the return on capital building block was calculated by reference to the inflation adjusted capital base).

50    ActewAGL proposed a forecast of inflation of 2.19 per cent (later updated to 1.96 per cent) based on the Breakeven Forecasting Methodology, which provided a market-based estimate of inflation.

51    The AER was required to accept ActewAGL’s forecast of inflation (and the methodology used to create it) unless the AER was positively satisfied that ActewAGL’s forecast: (a) was not arrived at on a reasonable basis; and (b) did not represent the best forecast available.

52    In the Final Decision, the AER imposed a forecast of inflation of 2.18 per cent based on the AER’s Forecasting Methodology, which relied on RBA forecasts for the first three years of the 2015-21 period, and then the mid-point of the RBA’s long-term inflation target range for a further seven years.

53    There was expert evidence before the AER that Australia had been experiencing very low inflation, and in prevailing low interest rate conditions, the RBA had little capacity to use monetary policy to lift inflation.

54    ActewAGL submitted that the AER did not attempt a meaningful analysis of these issues. Rather, the AER assumed that its Forecasting Methodology should be adopted unless persuaded otherwise. Further, the AER concluded that any change to its forecasting methodology should only occur after industry wide consultation outside of the decision-making procedures in the NGR.

55    ActewAGL submitted that the AER misconceived its statutory task by making a decision otherwise than in accordance with the NGR. The issues raised by the AER about the proper construction of the grounds in s 71 of the NEL/s 246 of the NGL did not arise. On any view, the AER had exercised its discretion incorrectly.

56    ActewAGL submitted it was not precluded from pursuing its grounds of review in relation to forecast inflation under r 60 of the NGR.

Materially preferable designated NGO decision requirement

57    As to the “materially preferable designated NGO decision requirement” in s 259 of the NGL, ActewAGL submitted:

(a)    a decision that was made according to law was properly construed as being a “materially preferable designated NGO decision” when compared to a purported decision that was not in accordance with law; that was to be contrasted with a decision that was “preferable”, in the sense that, if there were a range of decisions that were correct in law, the decision that was ultimately achieved was the best that could have been made;

(b)    the term “likely” in s 259(4a)(c) meant that there was a real chance or possibility, rather than that it was “more likely than not”, that varying or remitting the decision will result in a materially preferable designated NGO decision;

(c)    there was no requirement for the Tribunal to engage in “crystal ball gazing”, whereby it must consider the decision as made by the AER, and compare that with a counterfactual in which a different decision was made by the AER on remittal with a different result, in order to determine whether the requirement was met. Rather, a simple comparative exercise was called for, whereby the grant of relief was compared with the existing regulatory decision;

(d)    if its return on debt and forecast inflation grounds were made out, then the AER had made a decision about ActewAGL’s allowable revenue on a basis that was not authorised by the NGR. A decision that did not apply the NGR, and which the AER was not authorised to make, was incapable of furthering the NGO. A decision which did apply the NGR, on the other hand, was more likely to further the NGO, and was therefore likely to be materially preferable to a decision which did not. Accordingly, either a remitted or varied decision, which resulted in a decision being made in accordance with law, will necessarily be materially preferable;

(e)    the AER’s errors otherwise resulted in the Final Decision not furthering the NGO, in that the Final Decision did not apply the building block methodology in the NGR in accordance with its terms, and therefore did not allow ActewAGL a reasonable opportunity to recover its efficient costs of providing services pursuant to the access arrangement. The AER had not asserted (nor was ActewAGL aware of) any offsetting interrelationships or generosities in the Final Decision. Thus, there would likely be material adverse consequences for the long-term interests of consumers absent error correction.

58    The Tribunal turns now to consider in more detail the submissions of the parties in respect of each of these subject matters.

RETURN ON DEBT

Debt Transition

59    The principal issue here is whether the historical trailing average (HTA) approach to estimating the return on debt for use in the determination of the allowed rate of return for the 2015-16 year and the 2016-21 access arrangement period (together, 2015-21 period) should be applied immediately (which ActewAGL and Jemena proposed in their Revised Access Arrangement Proposal and Revised Regulatory Proposal, respectively) or whether there should be a transition to the trailing average (TTA) approach, with the HTA phased in from the previous on-the-day approach over a decade (which the AER decided in both cases). Both of those applicants had originally proposed a hybrid transition.

60    It was common ground between the parties that the HTA approach would assume that the BEE holds a staggered portfolio of fixed rate debt and refinances 10 per cent of that debt each year at prevailing 10-year interest rates. The interest rate used to calculate the allowance for debt in each regulatory year under that approach is a simple average of the current and previous nine years’ prevailing interest rates.

61    The TTA approach is a transition from the previous on-the-day approach that fixed the interest rate used to calculate the allowance for debt at the rate prevailing at, or close to, the start of each regulatory period for the whole of that period (currently five years). The transition lasts ten years, so that the prevailing/on-the-day rate applies in the first year of the new regulatory period; a weighted average of 90 per cent of the initial on-the-day rate and 10 per cent of the second-year prevailing rate applies in the second year; a weighted average of 80 per cent of the on-the-day rate and 10 per cent of each of the second-year and third-year prevailing rate applies in the third year; and so on until the tenth year when the rate applying in that year is a simple average of the current and previous nine years’ prevailing interest rates, as would apply with the HTA approach. This may be called a “fully-implemented” HTA.

62    Both TTA and HTA approaches “roll in” 10 per cent of the new prevailing rate at the beginning of each regulatory year. Consequently, the practical difference between the TTA and HTA approaches is that the TTA approach substitutes the interest rate prevailing at the beginning of the first regulatory period for all the historical rates prior to that time that would have been averaged into the HTA approach.

63    The AER’s imposition of a transition to the HTA approach is essentially the same issue as was raised in relation to debt by the parties in Ausgrid and, more narrowly focussed on only the debt risk premium (DRP) component of debt costs, in Application by SA Power Networks [2016] ACompT 11 (SAPN). In Ausgrid the Tribunal’s focus was on the rationale for transition, particularly the proper definition of the BEE and interpretation of r 6.5.2(k)(4) of the NER (equivalent to r 87(11)(d) of the NGR). The Tribunal in that matter concluded that the AER had misinterpreted the definition of a BEE as a regulated entity with hedged debt that required transition to the HTA approach. The AER subsequently adopted a different rationale for applying the TTA approach in the recent determinations for SA Power Networks, ActewAGL and Jemena, relying on an implied requirement of the NEO and revenue and pricing principles (RPP) to ensure a “zero NPV” condition was satisfied “over the life of the RAB”. That is, over time, the net present value of the allowance for the return on capital would just cover the cost of the regulated asset base. This rationale was contested in SAPN, but was accepted by the Tribunal in that matter. Since then, the Full Court has decided an application for judicial review of the Tribunal’s Ausgrid decision, as discussed in what follows.

Regulatory framework

64    This issue was raised by ActewAGL under the NGL and Jemena under the NEL. As we have noted, there are substantial parallels in the relevant provisions of both the NGL and the NGR and the NEL and the NER. The discussion to follow refers to both sets of provisions where necessary.

65    The framework governing the cost of capital is set out in r 87 of the NGR; the return on debt specifically is dealt with in rr 87(8) to (12). These rules are the result of substantial recent revision, as explained in the AEMC’s Rule Determination: National Electricity Amendment (Economic Regulation of Network Service Providers) Rule 2012; National Gas Amendment (Price and Revenue Regulation of Gas Services) Rule 2012, 29 November 2012 (2012 Rule Amendments).

66    Rule 87(1) of the NGR requires a calculation of the allowed rate of return in accordance with the terms of r 87 itself. Rules 87(2) and (3) are the main provisions. We have set them out at [18] above.

67    Prior to the 2012 Rule Amendments, the revenue allowance for the cost of debt was explicitly set as the sum of measures of the risk-free rate and the DRP, the sources for both of which were specified in some detail. The effect was to apply, at the beginning of a regulatory period, an estimate of the long-run cost of debt based on interest rates prevailing during a period close to the start of the regulatory period.

68    Market interest rates constantly change with economic conditions, so the prevailing rate (for instance, for suitably rated (BBB+), long-term (10-year) debt) at the start of a regulatory period is unlikely to be the rate that will apply to new debt over the course of that period. In the lead-up to the 2012 Rule Amendments governing the costs of capital, the AEMC noted concerns that the then mandated on-the-day approach encouraged risk management practices by service providers that would not occur in the absence of such regulation. It considered that the then current approach may lead to various mismatches between the regulatory estimate allowed by the regulator and the actual interest rate exposures of those service providers that employ debt management practices that are not closely aligned with the benchmark assumptions.

69    The 2012 Rule Amendments addressed this by giving the AER the discretion to choose between the on-the-day approach, an HTA approach, or some combination of these two approaches. It described the rule as setting out at a very broad level the characteristics of these three approaches to estimating the return on debt that could reasonably be contemplated by a regulator. The AEMC intended that the regulator could adopt more than one approach to estimating the return on debt, having regard to different risk characteristics of benchmark efficient service providers. The regulatory revenue allowance for debt is now required to be estimated so as to contribute to the ARORO (NGR r 87(8), NER s 6.5.2(h)), that is, to provide a service provider with a rate of return commensurate with the efficient financing costs of a benchmark efficient entity with a similar degree of risk as the service provider in respect of the provision of standard control services.

70    Under the new rules, if the AER estimates the return on debt using a methodology that results in different or potentially different rates in each regulatory year (r 87(9)(b)), the AER’s choice of method must be “effected through the automatic application of a formula” specified in the AER’s final decision at the start of the regulatory period.

71    As summarised in the AEMC 2012 Rule Amendments, at page 8:

2.4.3 Rate of return

The final rule introduces a new framework for determining the rate of return. It provides that the allowed rate of return for a NSP must meet an objective related to the efficient financing costs of a benchmark efficient NSP with a similar degree of risk as that which applies to the NSP subject to the decision. The final rule provides the regulator with sufficient discretion on the methodology for estimating the required return on equity and debt components but also requires the consideration of a range of estimation methods, financial models, market data and other information so that the best estimate of the rate of return can be obtained overall that achieves the allowed rate of return objective.

The final rule also provides for the allowed rate of return to reflect changing circumstances so that the application of the framework should result in the best overall estimate of the rate of return in any case that is commensurate with efficient financing costs. This should ensure sufficient funds are attracted for network investment, while minimising costs for electricity consumers.

72    The 2012 Rule Amendments did not alter the task for the AER: to set an allowance for the cost of capital that was the product of the projected capital base (the RAB in the case of the NER) and a weighted average of the returns on equity and debt (r 87(4)(a)). Nor did they change the previously understood cost concepts - that the returns on equity and debt were estimates of the long-run returns on those components of capital costs. As noted by the AEMC in the quote at [71] above, the new rules “introduce a new framework”, granting the AER “discretion on the methodology for estimating the required rate of return”, but not changing the underlying cost concept, summarised by the AER as: “the economic understanding that the cost of capital (equity and debt) is a forward-looking opportunity cost”.

73    Besides the requirement (r 87(2)) for the overall rate of return on capital to achieve the ARORO (defined in r 87(3) that the rate of return is commensurate with the efficient financing costs of a BEE with a similar degree of risk), under r 87(5) the AER is required to have regard to:

(a)    relevant estimation methods, financial models, market data and other evidence;

(b)    the desirability of using an approach that leads to the consistent application of any estimates of financial parameters that are relevant to the estimates of, and that are common to, the return on equity and the return on debt; and

(c)    any interrelationships between estimates of financial parameters that are relevant to the estimates of the return on equity and the return on debt.

74    The AER is also required (r 87(11)) to have regard to:

(a)    the desirability of minimising any difference between the return on debt and the return on debt of a benchmark efficient entity referred to in the allowed rate of return objective;

(b)    the interrelationship between the return on equity and the return on debt;

(c)    the incentives that the return on debt may provide in relation to capital expenditure over the access arrangement period, including as to the timing of any capital expenditure; and

(d)    any impacts (including in relation to the costs of servicing debt across access arrangement 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 access arrangement period to the next.

What the AER decided

75    The AER decided in the Final Decisions that it would not depart from the return on debt methodology that it had proposed in Chapter 6 of its December 2013 Better Regulation Rate of Return Guideline (RoR Guideline) and its Draft Decision for ActewAGL and Preliminary Decision for Jemena. At page 10 of Attachment 3 of the Jemena Final Decision the AER said, with an equivalent statement in the ActewAGL Final Decision (omitting footnotes):

We are satisfied that the allowed rate of return … we determined achieves the allowed rate of return objective (ARORO). That is, we are satisfied that this allowed rate of return is commensurate with the efficient financing costs of a benchmark efficient entity with a similar degree of risk as that which applies to Jemena Electricity Networks (JEN) in providing standard control services.

and later at pp 11-12 of attachment 3:

Consistent with our preliminary decision, we agree there should be a transition from the on-the-day approach to the trailing averaging approach to estimating the return on debt. However, we disagree with the hybrid form of transition proposed in JEN's (initial) regulatory proposal.

In its revised proposal, JEN departed from its initial position to apply a transition to the trailing averaging approach at all. It now proposes to not apply a transition (that is, to immediately move to a trailing average approach). We disagree with JEN on this and a number of other components of the rate of return.

and further at p 13:

This gradual transition will occur through updating 10 per cent of the entire return on debt each year to reflect prevailing market conditions in that year (a full transition). This approach is consistent with the approached we proposed in the Guideline and adopted in the preliminary decision.

76    Discussing the ARORO, the AER said, at pages 17 and 18 (omitting footnotes):

We are to determine the allowed rate of return such that it achieves the ARORO. The 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 regulatory regime is an ex-ante (forward looking) regime. As such, we consider a rate of return that meets the ARORO must provide ex-ante compensation for efficient financing costs. This return would give a benchmark efficient entity a reasonable opportunity to recover at least its efficient financing costs. This is a zero net present value (NPV) investment condition, which can be described as follows:

The zero NPV investment criterion has two important properties. First, a zero NPV investment means that the ex-ante expectation is that over the life of the investment the expected cash flow from the investment meets all the operating expenditure and corporate taxes, repays the capital invested and there is just enough cash flow left over to cover investors’ required return on the capital invested. Second, by definition a zero NPV investment is expected to generate no economic rents. Thus, ex-ante no economic rents are expected to be extracted as a consequence of market power. The incentive for investment is just right, encouraging neither too much investment, nor too little.

Under our regulatory framework, a benchmark efficient entity’s assets are captured in its RAB. The return on capital building block allows a benchmark efficient entity to finance (through debt and equity) investment in its network. Because investments usually carry a degree of risk, to satisfy the zero NPV condition the allowed rate of return must be sufficient to compensate a benchmark efficient entity's debt and equity investors for the risk of their investment.

77    The AER went on to discuss, at pages 18-19, efficient financing costs:

A key concept in the ARORO is ‘efficient financing costs’. Because the market for capital finance is competitive, a benchmark efficient entity is expected to face competitive prices in the market for funds. Therefore, we consider efficient financing costs are reflected in the prevailing market cost of capital (or WACC) for an investment with a similar degree of risk as that which applies to a service provider in respect of the provision of regulated services.

…

We consider employing a rate of return that is commensurate with the prevailing market cost of capital (or WACC) is consistent with the zero NPV investment condition (see above). We also consider economic efficiency more generally is advanced by employing a rate of return that reflects rates in the market for capital finance. Similarly, Partington and Satchell interpret efficient financing costs as the opportunity cost of capital, which is a market rate of return for assets with a given level of risk.

78    At page 24, the AER said the following regarding the benchmark efficient entity (omitting a footnote):

We consider a benchmark efficient entity with a similar degree of risk as that which applies to the service provider in the provision of its regulated services would be ‘a pure play, regulated energy network business operating within Australia’ acting efficiently. To understand this position, it is essential to understand the relationship and distinction between risk and expected returns. All else being equal, we consider an unregulated monopoly will have higher risk and higher expected returns than a regulated monopoly. This is because regulation:

•    mitigates monopolies from being able to extract monopoly rents, thereby constraining potential profits

•    increases the certainty of the revenue stream, thereby reducing risk.

For clarity, regulation reduces both risks that are compensated through the rate of return (for example, demand risk) and risks that would not be compensated through the rate of return (for example, by allowing cost pass throughs for unsystematic risks such as industry-specific tax changes or geographic-specific natural disasters). We only focus on risks that are compensated through the rate of return (compensable risks).

79    In what it makes clear is a later addition to its thinking, the AER said at page 3-279, within Appendix H, which deals in detail with the return on debt approach (omitting footnotes):

It is important to note that a debate has now arisen, since the submission of proposals in the matters under consideration, as to whether a benchmark efficient entity would be unregulated. In their recent revised proposals, service providers submitted that a benchmark efficient entity with a similar degree of risk in respect of the provision of regulated services must be an unregulated business. This followed the Tribunal hearing in an application for review of revenue determinations by Networks NSW, and ActewAGL which resulted in the Tribunal recently forming the view that a benchmark efficient entity referred to in the ARORO is likely not a regulated entity.

and further at page 3-280:

Nevertheless, even if a benchmark efficient entity was necessarily unregulated, we do not consider this would affect our conclusions. Our approach to the cost of debt would be applicable to an unregulated firm if it had a similar degree of risk to the service provider in providing regulated services. Further, irrespective of whether a firm is regulated or not, efficient financing costs reflect the current (or prevailing) forward looking costs observed in capital markets.

80    The AER referred to the 2012 Rule Amendments in support of its imposition of a transition as follows in Appendix H, at pages 3-290 to 3-291 (footnotes omitted):

We note that when undertaking the rule change in 2012 the AEMC added in clause 6.5.3(k)(4) that states (emphasis added):

(k) 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.

This clause is explicit in requiring us to have regard to any impacts on a benchmark efficient entity that could arise as a result of a change of methodology. This would include having regard to any material changes in the present value of a benchmark efficient entity's regulated revenue purely due to changing the debt estimation methodology. If such changes increased a benchmark efficient entity's value, then this would benefit its equity holders at the expense of consumers. Conversely, if such changes decreased a benchmark efficient entity's value, then this would cost its equity holders but provide a short term financial benefit to consumers (which may not be a long-term benefit to the extent this results in underinvestment). As such, this methodological change may also have a material negative impact on the confidence in the predictability of the regulatory regime. We consider the AEMC’s guidance on the intent of this clause is consistent with our approach (emphasis added):

The purpose of the fourth factor 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.

81    The AER added:

We have taken this factor into account and consider our transitional approach is consistent with the intent of this factor. Nevertheless, we consider that irrespective of this factor, our transition approach meets the requirements of the ARORO, NEO/NGO and RPPs.

82    With this background, the AER argued in broad terms as follows.

83    It considered that the on-the-day approach could contribute to the ARORO and was therefore open to it. In particular, the present value of a benchmark efficient entity’s allowed revenues would be sufficient to compensate it for its efficient financing costs.

84    The trailing average approach, once fully implemented in the manner it proposed, would similarly compensate a benchmark efficient entity for its efficient financing costs. Compared to the on-the day approach, a trailing average approach would lead to less volatile cash flows (a smoother price path over the long run, across regulatory periods); would in the view of some stakeholders reduce service providers’ risk; and receive broad stakeholder support. Thus the AER favoured moving from the on-the-day approach to the trailing average approach. The question was how to implement the change.

85    The AER’s method of phasing in the trailing average approach would contribute to the ARORO because it would be approximately revenue neutral in the sense that it would not change the benchmark efficient entity’s present value. These conclusions about the three methodologies – on-the-day, fully implemented trailing average, and phased-in trailing average - were supported by a mathematical analysis of present value calculations.

86    By contrast, the AER argued, the applicants’ proposed methodology would over-compensate them. The allowed return on debt would be greater than the benchmark efficient entity’s efficient financing costs. This was because the current market cost of debt was considerably below the average market cost of debt over the past nine years. Implementing the trailing average approach immediately would effectively treat the applicants as if they now faced a cost of debt considerably higher than in fact they do face. It would thus not produce a rate of return that would be commensurate with achievement of the ARORO.

87    If, instead, the trend of interest rates had been up rather than down, immediate implementation of the trailing average approach would cause under-compensation of the benchmark efficient entity, contrary to the ARORO.

88    Allowing service providers to choose a methodology according to whether known historical data favours them potentially biases regulatory outcomes.

The submissions to the Tribunal

89    ActewAGL’s grounds for review were set out at [103]-[118] of its application for review, and Jemena’s grounds were set out at [68]-[82] of its application. The grounds covered essentially the same issues. As was the case throughout the hearing, counsel were asked to provide short written outlines of their oral submissions prior to making them. The two applicants’ outlines covered the same territory, but differed in the way they were expressed. The two applicants did not rely on a common set of submissions. Given the length of the elaboration of grounds for review in the applications, and the overlap in arguments made, the Tribunal has considered it best consider the grounds for review by reference to a melding of the outlines of oral submissions.

90    Jemena’s attack on the AER’s decision had a number of elements, as follows:

    The “efficient financing costs of a benchmark efficient entity with a similar degree of risk as that which applies to [Jemena]” (r 6.5.2(c)) are actual costs incurred by the benchmark business under its debt instruments (facilities and bonds). The business has contractual commitments to pay interest under those instruments. This business would hold a staggered portfolio of fixed rate debt. The trailing average approach calculates the efficient financing costs of a benchmark business for Jemena.

    The AER’s construction of “efficient financing costs” is that they are prevailing costs.

    The interest rates payable on existing debt instruments held by the benchmark business are not at the prevailing rate for new debt in 2016. Only new debt taken out by the business in 2016 is at the prevailing rate. Therefore, the prevailing rate cannot sensibly be applied as a measure for the entire cost of debt.

    Whilst the on-the-day approach was required under the previous rules, the rules were amended and now give primacy to the allowed rate of return objective.

    The “on-the-day” approach (which is continued by the AER’s transition, in decreasing weights over 10 years) is neither the prevailing cost of debt nor a measure of the “efficient financing costs” of Jemena. The on-the-day approach is only an appropriate measure for a business that finances (e.g. a new entrant) or refinances all of its debt requirements at the commencement of the regulatory control period. That is not Jemena, or an appropriate benchmark for Jemena. The on-the-day approach and the AER’s transition do not calculate the “efficient financing costs of a benchmark efficient entity” in the case of Jemena. Thus the AER’s transition is not in accordance with clause 6.5.2 of the NER.

    The AER’s new “NPV = 0” approach from its Final Decision is not found in r 6.5.2 and cannot override the requirements of r 6.5.2. In any case, the analysis is incorrect. For both the on-the-day approach and the transition to a trailing average, it applies an interest rate that does not represent the efficient financing costs of an existing business. Moreover, the notion of revenue neutrality is meaningless in relation to debt costs. It makes no sense to speak of debt costs over the life of assets.

    Rule 6.5.3(k)(4) provides a transition for businesses that have changed their position in an irrecoverable way on the basis of the previous regulatory regime such as by entering into hedge contracts. Jemena is not such a firm.

    Other reasons given by the AER for a transition – avoidance of bias and measurement difficulties – are not valid reasons to fail to apply the ARORO.

    The AER’s change of reasoning was not foreshadowed to the service providers in contravention of r 11.60(k) of the NER and s 16(1)(b) of the NEL.

91    In a similar vein but with some variations, ActewAGL submitted, in summary, as follows.

    The AER’s decision is predicated on a misconstruction of r 6.5.2 of the NER. The AER’s construction: embodies a so-called ‘NPV = 0’ criterion over the “life of the RAB”; interprets “efficient financing costs” as relating solely to “prevailing” interest rates (but does not require that those rates be prevailing at any particular time); and does not require any consideration of the financing practices of the BEE.

    Those concepts cannot be reconciled to the text of r 6.5.2. Rather, it appears that the AER has reached what it considered to be the ‘right’ outcome, and then had sought to fit that conclusion into r 6.5.2. That is not a proper basis for statutory construction.

    Further, the AER’s financial theory (ie, that the trailing average approach should only be adopted if there is a transition that preserves “revenue neutrality” over the “life of the RAB” with the on-the-day approach) depends on its conclusion that the on-the-day approach necessarily satisfies the ARORO. That conclusion is untenable. Accordingly, the theoretical premise of the AER’s decision (even leaving aside its construction issues) is flawed.

    ActewAGL contended that the phrase “efficient financing costs” in r 6.5.2(c) related to all matters that bear upon the financing costs that the BEE, acting efficiently, would incur during the regulatory period, including its efficient financing practices. This construction is consistent with the extrinsic material, the text of r 6.5.2 of the NER, and the Ausgrid and EnergyAustralia Decisions.

    ActewAGL has no debt. Therefore, the change in debt methodology to the trailing average approach has no “impact” on it for the purposes of r 6.5.2(k)(4) of the NER. Given that it is common ground that a BEE would hold a staggered portfolio of debt, and finance (or refinance) 10 per cent of that portfolio each year, the trailing average approach should therefore be applied without a transition.

    Given that it was common ground that the trailing average methodology was most likely to represent the debt portfolio of a business in a competitive market, then, consistently with the conclusions expressed by the Tribunal in Ausgrid, the efficient financing costs of ActewAGL for the purposes of r 87(3) of the NGR ought to have been determined on that basis. That is to say, ActewAGL should have been allowed a return on debt based on the trailing average approach without a transition.

92    The AER submitted:

    The meaning of the rules governing the allowed rate of return in the NER and the NGR is informed by the finance and economic principles that underpin the rules, and the extrinsic materials that explain the ‘problems’ sought to be overcome by the previous rules and the purposes of the amended rules.

    In its Final Decisions, the AER decided that:

(a)    either the on-the-day approach or the trailing average approach, consistently applied, would contribute to the achievement of the allowed rate of return objective;

(b)    switching immediately from an on-the-day approach to a trailing average approach would be unlikely to be revenue-neutral;

(c)    in the current market circumstances, an immediate adoption of a trailing average would lead to an excess positive return in favour of the service provider relative to the efficient return in the market.

    The on-the-day approach to estimate the return on debt is consistent with the allowed rate of return objective, as it reflects the prevailing cost of capital.

    The term “efficient financing costs” in the allowed rate of return objective encompasses both return on equity and return on debt. The words “financing costs” therefore refer to the cost of capital. ‘Efficient’ should be given its full meaning consistent with the NEO.

    The AER did not err in concluding that an immediate change from an on-the-day approach to a trailing average approach would lead to an excess positive return in favour of the service provider relative to the efficient return in the market.

    The AER did not err in concluding that NER r 6.5.2(k)(4) permitted it have regard to the potential for consumers and service providers to face a significant and unexpected change in costs or prices resulting from a change in the methodology of estimating the cost of debt.

    The AER did not err in having regard to the potential for regulatory bias and historical data issues that arise from immediately adopting a trailing average approach.

    No error arises by reason that the Final Decisions explained the AER’s reasons in greater depth than the Preliminary Decisions, including by reference to the underlying economic principles.

    Even if the AER’s decision involved error, the Tribunal should affirm the decision because the alternative would not result in a materially preferable NEO decision.

The Full Court’s decision

93    As mentioned above, after the hearing by the Tribunal in the present proceedings, the Full Court’s decision on its judicial review of the Tribunal’s Ausgrid decision was handed down, on 24 May 2017. That decision is Australian Energy Regulator v Australian Competition Tribunal (No 2). The participants in this Tribunal’s reviews were given the opportunity to make further submissions.

94    ActewAGL relevantly made the following submissions in respect of the Full Court’s decision.

•    First, the Court concluded that the required benchmarking for the purposes of cl 6.5.2(c) extends to the debt management practices of the BEE. In particular:

a)    at [531]-[533], in the context of discussing the 2012 Rule Amendments, the Court implicitly accepted that the purpose underlying cl 6.5.2 was to provide an incentive for service providers to adopt efficient financing practices, and that to that end, ‘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’; and

b)    at [571] the Court concluded that: cl 6.5.2(c) is an ‘efficiency yardstick’ of the costs that would be incurred in a workably competitive market; that considered against that yardstick, ‘the service provider’s debt management practices will either be efficient or inefficient’; and that for the purposes of considering whether an ‘impact’ of the kind referred to in cl 6.5.2(k)(4) exists, it is relevant to consider whether the service provider has ‘already implemented a debt structure that satisfies a required aspect of the intended benchmark efficiency’ in cl 6.5.2(c) (that ‘required aspect’ being the service provider’s debt management practices) [emphasis added].

•    The effect of these conclusions is that the expression ‘efficient financing costs’ in cl 6.5.2(c) is not (as the AER now contends) confined solely to interest rates.

•    Secondly, the Court concluded that the service provider’s actual debt management practices are relevant to considering whether an ‘impact’ of the kind referred to in cl 6.5.2(k)(4) exists that should be taken into account when transitioning from one methodology for estimating return on debt to another. In particular:

a)    at [557]-[560], the Court acknowledged that the AEMC’s purpose in promulgating the 2012 Rule Amendments (including cl 6.5.2(k)(4)) was to address the debt management practices that service providers had entered into in reliance on previous methodologies for estimating return on debt deployed by the AER;

b)    at [571], as discussed above, the Court reasoned, when considering whether an ‘impact’ of the kind contemplated by cl 6.5.2(k)(4) exists, it is relevant to consider whether the particular service provider's actual circumstances, including its debt management practices, are efficient when assessed against the benchmark.

•    Those conclusions cannot be reconciled with the AER’s current construction of cl 6.5.2(k)(4) as being concerned only with maintaining revenue neutrality for the BEE between two methodologies over the ‘life of the RAB’ or 10 year term of the BEE’s debt portfolio. Not only does the AER’s construction not allow for any consideration of debt management practices (as it is focussed solely on interest rates), but it also does not allow for any consideration of the individual circumstances of the service provider (i.e. its debt management practices) in considering the ‘impact’ of a change in debt methodology.

•    Thirdly, the Reasons are directed to the assessment of efficient costs during the current regulatory period, and the ‘impact’ of a change in debt estimation methodology from the previous regulatory period to the current regulatory period (see especially at [571]-[573]). The Reasons do not contemplate assessment of efficient financing costs over multiple past and future regulatory periods. The Reasons therefore cannot be reconciled with the AER’s current construction and application of the ARORO, which is that the assessment of ‘efficient financing costs’ is to be assessed over the ‘life of the RAB’, which spans an indeterminate number of regulatory periods (or alternatively, the 10 year term of the BEE’s debt portfolio, which spans the two previous regulatory periods).

•    Based on its construction of cll 6.5.2(c) and 6.5.2(k)(4), the Court concluded that no impacts of the kind referred to in cl 6.5.2(k)(4) exist that are 'apposite to the benchmark efficient entity' for any NSW DNSP or ActewAGL (at [572] and [573]). The effect of this conclusion is that the trailing average approach must be applied without a transition for ActewAGL because '…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 transition simply did not exist' (at [573]).

•    The Court's conclusions about ActewAGL are express and emphatic in their terms. They apply to the matter before the Tribunal at present. There is no basis to distinguish the Court's reasons. It follows that the AER has made a reviewable error in applying a trailing average approach with transition, in that it misconstrued r. 87 of the NGR (which mirrors cl 6.5.2).

95    Jemena made the following submissions:

•    First, the Full Court has adopted a construction of cl 6.5.2 of the NER whereby, in the case of debt, it is concerned with the costs of the entity in question, but measured by reference to an efficiency yardstick, being the benchmark efficient entity, where different entities can have different benchmarks: at [529]-[532], [536]-[537], [539], [543], [571]. At [571], the Full Court observed:

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.

•    The Full Court’s position is therefore consistent with the position adopted by JEN in its submissions in reply at [54]. As explained by JEN in the hearing before the Tribunal, the trailing average approach, for an entity in the position of JEN, has the benefit of matching the costs that would be incurred by an efficient business in the position of JEN, both under existing facilities and under new facilities as they are taken out over the regulatory control period. The on-the-day approach does not. As it happens, the trailing average approach also provides an appropriate signal for efficient investment for a business in the position of JEN because it provides the current rate for new debt, whereas the on-the-day approach does not because it provides a single rate for all debt (both existing and new) across the whole 5 year period, and indeed for 10 years (in decreasing weighting) under the AER’s transition.

•    More generally, the decision of the Full Court is not consistent with a construction of “efficient financing costs” in cl 6.5.2 as meaning “prevailing rates in the market”, or any such combination of prevailing rates in the market and costs of the (efficient benchmark) business as the AER may choose in its discretion.

•    Secondly, the Full Court has adopted a construction of cl 6.5.2(k)(4) of the NER which is consistent with JEN’s construction, but inconsistent with the AER’s construction: at [560], [570]-[574]. The Full Court concluded that cl 6.5.2(k)(4) was concerned with impacts arising from a change of methodology on a business that had entered into commitments on the basis of the previous methodology. Thus the Full Court concluded, at [572]-[573], that in the case of the Networks NSW businesses, who did not have hedge contracts, and in the case of ActewAGL, which did not have debt, there was no scope for the application of cl 6.5.2(k)(4) because there were no “impacts”. As the Full Court observed at [573], “the occasion or need for a transition simply did not exist”. The Full Court therefore adopted the approach to the construction of cl 6.5.2(k)(4) advanced in JEN’s submissions in chief at [80]-[85], and rejected the approach advanced in the AER’s submissions at [61]-[66].

•    This is significant, because cl 6.5.2(k)(4) was the sole clause that the AER could point to as providing textual support for considering wider NPV=0 issues over the life of the asset (whatever that means). In particular, the Full Court’s construction of cl 6.5.2(k)(4) does not support: (a) the AER’s approach of not permitting a business to recover its efficient debt costs on the basis that the perpetuation of a mismatch is consistent with some broader NPV=0 consideration, or (b) the AER’s approach of considering the effect of changing a methodology on “expectations” at the time of entry into debt facilities.

•    The position of JEN is relevantly indistinguishable from the position of ActewAGL, which was the subject of the conclusion of the Full Court at [573].

96    The AER made the following submissions (footnotes omitted):

•    The conclusions of the Full Court in AER v Australian Competition Tribunal (No 2) bear upon, but are not determinative of, the grounds of review raised by the Applications of Jemena (under the NEL) and ActewAGL (under the NGL) with respect to the return on debt topic.

•    The AER acknowledges that the ultimate conclusion it reached in its Final Decisions with respect to the return on debt topic for Jemena and ActewAGL was materially the same as the conclusion it reached in its Final Decisions for the NSW/ACT service providers, the subject of the Tribunal decision in Re Public Interest Advocacy Service and Ausgrid [2016] ACompT 1 (the Ausgrid Tribunal decision) and related decisions and the Full Court’s decision in AER v Australian Competition Tribunal (No 2).

•    However, the reasoning in the two sets of decisions differs materially. Having received the adverse decision from the Tribunal in the NSW/ACT matters, the AER reconsidered the proper approach to the applicable rules on the return on debt topic for the Jemena and ActewAGL Final Decisions, including the economic considerations relevant to the applicable rules and the AER’s decision, and explained its reasons for decision in greater depth with reference to the economic principles that underlie the NER. In the Jemena and ActewAGL Final Decisions, the AER considered that the previously applied on-the-day methodology had provided ex-ante compensation for efficient financing costs (referred to as the NPV=0 condition), consistent with the ARORO (and the NEO/NGO and the revenue and pricing principles). The AER concluded that compliance with the ARORO requires the change of methodology to be implemented in a forward-looking manner so that the change was revenue neutral and did not affect the net present value of the future cash flows of the service providers through the PTRM (avoiding windfall gains or losses to the service provider and consumers). That conclusion was independent of, but consistent with, the considerations arising under NER cl 6.5.2(k)(4).

•    As recorded by the Tribunal in the Ausgrid Tribunal Decision, the primary errors asserted by the service providers on the part of the AER were adopting the concept of the benchmark efficient entity as a regulated entity and adopting a “one size fits all” benchmark efficient entity for the purposes of each of its decisions. The Tribunal found that once the step is taken of starting with a benchmark efficient entity that has the characteristics of a participant in a competitive market, the AER’s approach to transitioning the return on debt estimate, and the application of clause 6.5.2(k)(4), must be reconsidered.

•    In AER v Australian Competition Tribunal (No 2), the Full Court found no error in those conclusions of the Tribunal. The Full Court concluded that NER cl 6.5.2(c) does not, in terms, posit the benchmark efficient entity as either a regulated entity or an unregulated entity, although the degree of risk to be attributed to the benchmark efficient entity may be affected by the fact that the service provider’s provision of the relevant services is regulated. The Full Court concluded that clause 6.5.2(k)(4) contemplates the possibility that there may not be a single benchmark efficient entity; the Full Court also considered that, in light of the AER’s conceptualisation of efficient financing costs of a benchmark efficient entity in its reasons in the NSW/ACT matters, the AER had not identified any impacts from changing methodologies that would activate clause 6.5.2(k)(4).

•    The issues for determination in the present matter differ materially from the issues determined in the Ausgrid Tribunal Decision and AER v Australian Competition Tribunal (No 2). That is because the AER’s reasoning for its conclusion in the Jemena and ActewAGL Final Decisions differed materially from the NSW/ACT matters. As a consequence, the Applications for Review of Jemena and ActewAGL, while worded differently, raise the following common issues for determination:

a)    the proper construction of the expression “efficient financing costs” within the allowed rate of return objective in NER clause 6.5.2(c) and “efficient costs” within the revenue and pricing principles in s 7A of the NEL;

b)    the economic reasons for using prevailing rates (also referred to as the “on the day” methodology) in determining a return on debt (and the cost of capital more generally) and the economic reasons for changing to a trailing average methodology;

c)    the economic consequences of changing from an “on the day” methodology to a trailing average methodology and whether the change, without a forward looking transition, would conflict with the NPV = 0 condition which underpins the building block framework prescribed by the NER;

d)    the proper construction of NER cl 6.5.2(k)(4);

e)    whether the AER was correct to have regard to the difficulties associated with historical debt data and the potential for bias in regulatory decision making;

f)    whether the AER failed to consult with Jemena and ActewAGL with respect to its further reasoning for its return on debt decision.

The benchmark efficient entity issue

•    ActewAGL’s Application (but not Jemena’s Application) raises the issue of the proper characterisation of the benchmark efficient entity. ActewAGL contends that the AER erred because it characterised the benchmark efficient entity as a regulated entity, while it ought to have characterised it as an unregulated entity.

•    The AER acknowledges that the effect of the decision in AER v Australian Competition Tribunal (No 2) is that it was incorrect for the AER to characterise the benchmark efficient entity as regulated. However, that does not mean that ActewAGL has established a ground of review that would require the AER’s decision to be set aside. That is for the following interrelated reasons.

•    First, ActewAGL’s ground of review is itself incorrect. The Full Court did not conclude that the benchmark efficient entity should be characterised as unregulated, as propounded by ActewAGL. Indeed, the Full Court recognised that the risks of regulation may affect the degree of risk that the benchmark efficient entity bears.

•    Second, the question whether the benchmark efficient entity is characterised as regulated or unregulated was not material to the AER’s Final Decision with respect to ActewAGL (or, for that matter, the other service providers before the Tribunal in the current matters). As explained above, the AER’s decision to adopt a forward-looking transition arose because this was the only option that it considered met the ARORO and the NEO/NGO. This decision did not depend on the conception of the benchmark efficient entity.

•    The AER specifically considered the character of the benchmark efficient entity in section H.1.2 of its reasons. In that section, the AER found that the relevant benchmark efficient entity was ‘a pure play, regulated energy network business operating within Australia’. The AER noted that a debate had now arisen with the service providers as to whether the benchmark efficient entity should be unregulated. The AER found that whether the benchmark efficient entity was regulated or unregulated made no difference to its ultimate conclusion on debt transition:

…even if a benchmark efficient entity was necessarily unregulated, we do not consider this would affect our conclusions. Our approach to the cost of debt would be applicable to an unregulated firm if it had a similar degree of risk to the service provider in providing regulated services. Further, irrespective of whether a firm is regulated or not, efficient financing costs reflect the current (or prevailing) forward looking costs observed in capital markets.

•    Third, before granting any relief after a ground of review is established, the Tribunal must consider the materially preferable test set out in NGL s 259(4a). The Tribunal ought not be satisfied that a materially preferable NGO decision would be achieved by setting aside the AER’s decision and remitting the matter to the AER to take into account the correct characterisation of the benchmark efficient entity. Having regard to the AER’s expressed reasons as extracted in the preceding paragraph, the Tribunal can be satisfied that a change in the characterisation of the benchmark efficient entity would not alter the AER’s decision.

NER cl 6.5.2(k)(4) and NGR cl 87(11)(d)

•    Both the Jemena and ActewAGL Applications raise as an issue the proper construction of NER cl 6.5.2(k)(4) and NGR cl 87(11)(d).31

•    In AER v Australian Competition Tribunal (No 2), the Full Court considered the manner in which the Tribunal in the NSW/ACT matters had interpreted NER cl 6.5.2(k)(4) and concluded that there was no error in the Tribunal’s approach. However, for the following reasons, the Full Court’s decision does not determine the issues raised by Jemena and ActewAGL, nor the outcome of the “materially preferable test”.

•    First, the Full Court was not required to provide, and did not provide, an exhaustive analysis of the meaning of clause 6.5.2(k)(4). Its task was narrower: to determine whether the Tribunal had erred in its construction of the clause in the light of the facts and issues raised before it.

•    Second, the Full Court was not required to consider, and did not consider, the specific issue raised in this matter: whether clause 6.5.2(k)(4) includes changes to revenue (positive or negative) that arise solely from a change in methodology and which would deviate (positively or negatively) from the recovery of efficient financing costs in accordance with the NPV=0 condition. While this matter was raised in grounds of review before the Tribunal in the Ausgrid matter, those grounds were not ultimately addressed by the Tribunal because it found error at an earlier point (the characterisation of the benchmark efficient entity). Accordingly, the issue was not considered by the Full Court.

•    Third, the AER’s decision to adopt Option 2 (the transition) had regard to, but did not depend on, NER cl 6.5.2(k)(4) and NGR cl 87(11)(d). As noted above, the AER based its decision on the ARORO, and considered that that conclusion was consistent with NER cl 6.5.2(k)(4) and NGR cl 87(11)(d).

The Tribunal’s analysis

97    The requirement of r 6.5.2(h) is to estimate the return on debt for a particular service provider such that it contributes to the achievement of the allowed rate of return objective. First, the Tribunal notes that in estimating the return on debt, the AER was faced with:

    revised proposals by the applicants for immediate implementation of an HTA of interest rates (the initial proposals having been for a form of transition to that result);

    the fact that in previous regulatory periods, the on-the-day approach had been applied (and indeed mandated);

    a new (as of 2013) Rule which provides a discretion, without limitation, to choose between the on-the-day approach, an HTA, or some combination of the two, subject to r 6.5.2(h) and via it, r 6.5.2(c).

In short, it had a choice, subject to compliance with r 6.5.2(h), and with the NEL and NER more generally.

98    The AER was thus in the position contemplated by s 16(1)(d) of the NEL. This section was considered by the Full Court, which said (at [159]):

…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.

Extrapolating from that requirement, the Tribunal considers that the AER must, in applying r 6.5.2(h), consider which of a choice of estimates, or estimation approaches, would contribute, or be likely to contribute, to the achievement of the ARORO to the greater or greatest degree. This has implications for the manner in which the Tribunal has assessed the applicants’ claims.

99    In the Tribunal’s view, the applicants’ submissions are best dealt with in the following categories of claims:

1.    That the immediate implementation of the trailing average would (uniquely for the applicant DNSPs) contribute to the ARORO.

2.    That the AER was in error in considering that its transition to a fully implemented HTA could contribute to the ARORO in the case of the applicant DNSPs.

3.    That the AER could only decide to impose a transition under r 87(11)(d) and was not in fact empowered to do so by that rule.

4.    That the AER’s decision miscarried because of a failure to foreshadow its change of reasoning.

In respect of Claims 1 and 2, both applicants emphasised that the construction of “efficient financing costs of the BEE” was the key issue, and that on which the AER was in error.

100    The essence of Jemena’s case was that a service provider ought to be compensated for its efficient debt costs, with efficiency measured by reference to a BEE; and that efficient debt costs would be those associated with a staggered portfolio of debt, refinanced 10 per cent each year. Senior Counsel for Jemena described debt obligations as costs of the business like other costs of the business, and stated that existing businesses have a staggered portfolio of debt. ActewAGL’s case was presented slightly differently, with the starting point being the AER’s wrong construction of “efficient financing costs” as concerning only interest rates, whereas they related to all matters that bear upon the financing costs that the benchmark efficient entity acting efficiently would incur during the regulatory period, including its efficient financing practices. Since the efficient debt portfolio of a benchmark efficient entity, which was not otherwise regulated by a debt methodology imposed on it by the AER, would be a staggered portfolio of debt as at the commencement of this access arrangement period, then that that is how ActewAGL’s return on debt should be assessed.

101    The Tribunal considers first the construction of “efficient financing costs”.

102    The applicants, in their submissions made subsequent to the Full Court’s 2017 decision, relied on statements there regarding efficient financing costs as set out above. It is worth setting out a number of the Full Court’s statements. The following points are particularly relevant:

•    In promulgating the 2012 Rule Amendments, 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.

•    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. The correct approach to estimating the return on debt is by reference to the efficient financing costs of a benchmark efficient entity, rather than the actual financing costs of the particular service provider.

•    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.

•    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. 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.

•     “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 … this does not mean that the benchmark efficient entity must be characterised as a regulated entity.

•    Thus … 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).

•    If one service provider’s risk was different to another service provider’s risk, … 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…. [T]here 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).

•    In the context of a consideration of r 6.5.2(k)(4): …[T]he undoubted function of r 6.5.2(c) is to provide … 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.

103    The applicants put particular weight on the last of those passages. The Tribunal notes that it is essential to read the Full Court’s reasons in the context of the questions it was considering. Those questions were: whether the Tribunal erred in law 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; and whether the Tribunal erred in its construction of r 6.5.2(k)(4) of the NER. However, the question immediately facing this Tribunal is the construction of “efficient financing costs” for the purpose of giving effect to r 6.5.2(b) and r 6.5.2(h). While the passage above deals with the application of r 6.5.2(h), it does so in the context of construing r 6.5.2(k)(4).

104    In making that point, the Tribunal does not seek to distinguish the words of the Full Court in any of the passages quoted above. Indeed, they have been quoted because of their relevance. But it also considers that the construction of “efficient financing costs” needs to be considered in the context of what is required by r 6.5.2 as a whole, in the context of the NEO and the NGO, and in the light of the specific claims by the applicants about how the return on debt should be estimated.

105    It is true that among the attributes of debt costs is that they constitute obligations to make payments, and in that regard are similar in many ways to other obligations to meet costs of doing business, such as paying wages and salaries, and other components of operating expenditure (opex). But the regulatory scheme treats debt very differently.

106    First, there is no building block for debt. It is not included in opex, and there is no explicit, separate allowance (in dollar terms) for the cost of debt. Rather, there is a building block for return on capital. The return on capital implicitly – through the manner in which it is estimated – contains components for a return on equity and a return on debt. But the return on capital is not calculated separately for equity and debt. The return on capital (in dollar terms) is found by multiplying the rate of return (on capital) by the regulatory asset base (electricity) or the projected capital base (gas).

107    In the economic theory on which the regulatory scheme is based, the rate of return on an investment depends on the risk characteristics of that investment. It is a required rate of return: required ex ante by the investor. It is also called the cost of capital. Perhaps counter-intuitively, the cost of capital is not determined by the financial cost of raising debt and equity, but by the risk of the investment. That is because, as usually in economics, it is an opportunity cost. The question is how it can best be estimated.

108    As with so much of the regulatory framework governing the rate of return, there is a large body of economic theory underlying the Rules. That economic theory is broadly accepted by all the stakeholders, but there is disagreement about some of the details, and in particular how the theory should be applied. It is entirely uncontroversial in conventional economics that the required rate of return on an investment is determined by the risk of the investment. Under the usual, generally-accepted, assumptions, in a workably competitive market that rate of return will be just sufficient to give the investment’s stream of cash flows a net present value of zero.

109    This is an ex ante concept. That is, the expectation of the investor at the time of making the investment is that the positive cash flows, discounted at the required rate of return, will equal the cost of the investment. Obviously, investors will seek an NPV higher than zero, but competition will prevent that occurring. So the theory goes.

110    It is no leap to consider that a company’s required rate of return is determined by the risk of its investments, i.e of its assets, and can be regarded similarly. All sorts of things might change after the investment is made. It is the possibility of different outcomes that gives rise to the risk in the first place. The investment might turn out better or worse than expected. Of course, in normal circumstances, the debt-holders will expect to get paid, and the equity-holders bear the risk, but debt also has its risks.

111    Moreover, the required rate of return is independent of the financing of the investment or, in the case of a firm’s assets, of its capital structure. That is, the required rate of return depends on the risk characteristics of the firm’s assets, not on the nature of its liabilities (debt and equity).

112    How is that required rate of return to be estimated? More specifically, how is the required rate of return for a regulated electricity or gas service provider to be estimated? Broadly, the idea – again uncontroversial – is that the regulator seeks to estimate the required rate of return (or cost of capital) of a company that has a similar degree of risk to the regulated company but is operating in a workably competitive market. That (opportunity) cost of capital will be efficient in the economic sense, which goes beyond the sense of “with a minimum of waste”. It will lead to the regulated company charging prices for its services that promote efficient investment in, and efficient operation and use of electricity and gas services for the long-term interests of consumers. That is, the required rate of return, properly estimated, will serve the NEO or NGO.

113    The Rules, as now amended, guide the regulator towards the estimate of the required rate of return through the ARORO. It is obvious, then, that the phrase “efficient financing costs” must be construed consistently with the efficiency thrust of the NEO and NGO, ie in terms of the accepted notions of economic efficiency.

114    The allowed rate of return has two components: the return on equity and the return on debt. The allowed rate of return is to be determined such that it achieves the ARORO. The components of the allowed rate of return cannot individually achieve the ARORO. They are to be determined so as to contribute to the achievement of the ARORO. Achievement of the ARORO consists of the rate of return being commensurate with the efficient financing costs of the BEE.

115    In the Rules, “return on equity” and “return on debt” are rates, ie percentages. The allowed return on equity and return on debt in dollar terms are never calculated. Instead, the estimated rate of return on equity and rate of return on debt are averaged using equity and debt weights to generate the estimate of the overall rate of return on capital, also known as the weighted average cost of capital. Of course, arithmetically, one could instead calculate the dollar allowed return on equity, apply the appropriate weight to it, and add it to the weighted dollar allowed return on debt. The fact that the Rules do not take that approach reflects the underlying theory.

116    The rate of return on capital has primacy. The use of the rates of return on equity and debt is a means to the end of estimating that overall rate of return, and to then determining the return on capital building block. From its total allowed revenue, the service provider must meet its debt obligations along with its opex, capex and other expenditures. The Rules are careful not to imply that there is a guaranteed return to the shareholders. They get what is left. If the service provider can operate more efficiently than expected – including no less in its debt management than its management of opex, for example – the shareholders may receive a higher rate of return than was expected in the forward-looking framework.

117    This is reflected in the fact that the ARORO relates to the overall rate of return – the rate of return is to achieve the ARORO – while the rate of return on debt and the rate of return on equity are to be determined so as to contribute to the achievement of the ARORO. At this point a stark difference might be noted in the degree of prescription in the Rules for estimating the return on equity compared to the return on debt. The return on equity is to be estimated such that it contributes to the ARORO, and in making the estimate regard must be had to the prevailing conditions in the market for equity funds. That is the sum of the guidance given. By contrast, the approach to the estimation of the return on debt is prescribed in far more detail, as has been set out above.

118    And yet in each case it is the ARORO to which reference is made to govern the estimation. It can be seen that the ARORO has a lot of work to do in the case of the return on equity. The AER has very wide discretion in deciding on an approach. In fact, the practice is to apply the Capital Asset Pricing Model (CAPM), as explained in the RoR Guideline, which spends considerably more space explaining the return on equity than the return on debt. The CAPM is a highly simplified and elegant framework embedded in a broader finance theory that assumes that an investor holds a fully diversified portfolio and abstracts greatly from the day-to-day realities of equity markets (such as transactions costs). This framework is consistent with the forward-looking rate of return.

119    We add that the use of the word “commensurate” in the ARORO might be regarded as an acknowledgement that the efficient financing costs of a BEE are not something that can be estimated with precision. The injunction to achieve the ARORO, while unequivocal, has an unavoidably aspirational aspect to it. The determinations of the returns on equity and debt are, equally unavoidably, at a remove from the achievement of the ARORO; hence the requirement that their determination “contribute”.

120    Whether or not the Rule acknowledges an imprecision in the task facing the regulator, it is certainly the case that practitioners would never claim that a required rate of return can be estimated with precision.

121    The point of mentioning the return on equity is to note that in construing the ARORO, and in particular “efficient financing costs of a BEE with a similar degree of risk as that which applies to the DNSP”, it is clear that all the constituent terms, and indeed individual words, are redolent with the economic theory – generally accepted by stakeholders – of the whole investment framework given effect to by the Rules. Ultimately that framework is employed in the regulatory scheme to serve the NEO and the NGO for the long-term interests of consumers. Those objectives embody notions of productive, allocative and dynamic efficiency. It would be a mistake to think of efficient costs in the context of return on debt purely in terms of efficient debt management practices. The broader objectives need to be kept in mind.

122    The Tribunal sees no tension between this view and the statements by the Full Court, set out at [102] above. In particular, the term “efficient financing costs” embodies the ex ante, forward-looking, expectations-based framework briefly described above. This does not seem to be in contention when expressed in those broad terms.

123    A question arises whether the term “efficient financing costs of a BEE” needs to be interpreted differently in estimating the return on debt from how it is interpreted in determining the return on equity. In the case of the return on debt, the AER has a somewhat narrower discretion. The Rules give more guidance than in the case of the return on equity. Nevertheless, the discretion is still considerable. In particular, although attention is given to the time at which a BEE raises debt – at the time of or shortly before the making of the determination, or over a historical period prior to the commencement of a regulatory year, or a combination – that Rule is expressed in non-mandatory terms (“the methodology may be designed…”) and is without relevant limitation.

124    The Tribunal concludes that “the efficient financing costs of a BEE” has the same meaning whether applied to the estimation of the return on equity or the return on debt, and should not be seen as carrying deeper or more specific implications in one case than the other. However, there is no avoiding the fact that estimating the return on debt is a very different process from estimating the return on equity. We shall return to what the AEMC had to say about that difference.

125    With that background, we turn to the four claims by the applicants.

Claim 1: The immediate implementation of the trailing average approach would contribute to the achievement of the ARORO

126    There were several assertions of what was common ground between the parties. The Tribunal can only be confident that all parties seem to agree that it is a reasonable assumption that a BEE for either of the applicants would, in due course, ten years after implementation of the HTA approach (either immediately or with a transition), hold a staggered portfolio of debt, refinanced 10 per cent each year. Clearly the AER does not accept that the BEE would already, at the start of the regulatory period to which its Final Decisions apply, have such a staggered portfolio in place. The applicants assert that the BEE in their cases would. Jemena concedes that the BEE for another firm, such as a new entrant, might not have a staggered portfolio in place.

127    Effectively the claim is that a BEE for Jemena or ActewAGL would have adopted that financing practice ten years ago (or earlier) because it is the efficient debt management practice in a workably competitive market.

128    The Tribunal has a number of concerns. The first is that it cannot be said that the efficient debt management practice of any particular BEE would be to have had in place a staggered portfolio of fixed-rate ten-year debt (although reference was made to the apparent acceptance by the Tribunal of that proposition in Ausgrid at [865]). Given the vast array of financing options in modern capital markets, it would be heroic indeed to try to specify what the debt financing arrangements of a BEE would look like. In fact, neither Jemena nor ActewAGL has any debt; both are 100 per cent equity-financed. That is itself indicative of the range of efficient actual debt management practices available. Many strategies are used in workably competitive markets to raise debt including, indeed, hybrids of debt and equity.

129    It is unclear what debt management practices a BEE with similar risk to the applicant service providers would adopt. The only characteristic that the Rules specify for the BEE is that it have similar risk to the company for which it is the benchmark (specifically, risk in respect of the provision of reference services). No evidence was presented to suggest that this degree of risk is uniquely associated with – or in any way associated with – the fact that the company was not a new entrant but had been operating under the regulatory regime for some years. Yet this was the only characteristic proposed as making it almost self-evident that the BEE would have a staggered portfolio of debt. To emphasise the point: the BEE is not to be supposed to ape the characteristics of the company for which it is the benchmark in extraneous respects such as debt management arrangements. It is only risk that is germane. This is all the more clear when the BEE is specific to a company that has no debt, as in the case here.

130    In fact, no persuasive evidence was adduced that staggered debt portfolios are the norm for firms with any particular risk characteristics in a workably competitive market.

131    The AEMC did not make the 2012 Rule Amendments because there was evidence that in a workably competitive market firms adopt a staggered portfolio of ten-year fixed-rate debt. Rather its concern was that an efficient debt management practice for DNSPs might be to manage debt in that way, and that the on-the-day approach was preventing or discouraging it; and that the on-the day approach was forcing inefficient debt-management practices on some service providers.

132    The 2012 Rule Amendments should be seen not as changing the method of estimating the cost of debt so as to be in accord with a debt management practice that had been in place for ten years, but as removing an impediment to that practice where it was efficient. Importantly, the Rule change did not mandate the new approach but allowed it, without limitation, as a regulatory option along with the on-the-day approach or a combination of the two. That is very far from making a finding, or adopting a fiction, that a BEE would have had a staggered portfolio of debt in place for ten years.

133    But more than that, the Tribunal does not consider that the ARORO can be interpreted such that the term “efficient financing costs” should generate and govern a search for supposedly superior actual debt management practices in the market. Such a search would have to consider the risk/return trade-off in various debt instruments. As the Full Court made clear, a BEE is a construct which, while specific to a given DNSP, is simply a means to the end of judging whether financing costs are efficient. And that exercise is undertaken to assess whether the overall rate of return on capital is commensurate with efficient financing costs.

134    This is consistent with the AEMC’s statements:

The final rule provides the regulator with sufficient discretion on the methodology for estimating the required return on equity and debt components but also requires the consideration of a range of estimation methods, financial models, market data and other information so that the best estimate of the rate of return can be obtained overall that achieves the allowed rate of return objective.

and:

…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 risks of creating distortions in the service providers’ investment decisions.

and:

… the new rules allow the regulator (and the appeal body) to focus on whether the overall rate of return meets the allowed rate of return objective, which is intended to be consistent with the NEO, the NGO and the RPP.

There is no hint that the AEMC considered that there was a single right answer to estimating the return on debt, or a range of single right answers for different service providers. Rather, its whole emphasis was on giving the AER a discretion and a guiding light.

135    The HTA approach is consistent with an assumption that the BEE has refinanced, and will continue to refinance, one-tenth of its debt each year. This is a simplifying assumption. It is far from a finding that the efficient financing costs of a BEE would in fact involve such a debt structure and debt instruments. That is asking too much of the ARORO and the concept of efficient financing costs. The objective is more modest. It is not to specify efficient debt management practices but, by reference to a workably competitive market, to ascertain whether the return on capital (overall) is commensurate with efficient financing costs. As part of that, it is to have the return on debt contribute to that exercise.

136    In the case of Jemena and ActewAGL there is little difficulty in accepting the artificial construct of a BEE that would have refinanced the whole of its debt at the start of the regulatory period because that would be efficient given the risk attributed to it, being a similar degree of risk as applying to the relevant DNSP in the provision of standard control services, and noting the Full Court’s comment that the degree of risk may be affected by the fact that the service provider’s provision of the relevant services is regulated. To see this, we need to consider what is involved in such an attribution of degree of risk. Little was said about this in submissions to the Tribunal, so the Tribunal will not attempt a definitive analysis.

137    However, it is conventional that a firm’s risk relates to the variability of the cash flows generated by its assets. In the case of a firm regulated under the Rules, its cash flows are partly determined as allowable revenue. Under the on-the-day approach, a firm that is able to borrow at around the time its allowed revenue is set is guaranteed revenue – a cash flow that is less variable than in a workably competitive market – that includes a component very close to what is needed to just cover its actual (not opportunity) borrowing costs. For such a firm, the BEE with a similar degree of risk could thus reasonably be constructed as one that borrows “on the day”.

138    This is not to characterise the BEE as a regulated entity. It is simply to examine how to attribute the similar degree of risk to it.

139    The applicants argue that firms in their circumstances could not adopt that approach. But since they do not borrow at all, it is hard to see why the construct should be considered out of court. In any case, a BEE for either of them would have had the risk characteristics of a service provider whose allowed revenue included a return on its capital base calculated, in part, on a return on debt estimated by the on-the-day approach. It is not clear to the Tribunal why one should assume that such a BEE would have had a staggered portfolio in place over the last ten years.

140    To the extent that ActewAGL’s argument rested on the assumption that the BEE is an unregulated entity, which it considered to have been the Tribunal’s position in Ausgrid, that argument must be rejected in the light of the Full Court’s decision that the BEE is neither regulated nor unregulated, per se.

141    The Tribunal therefore does not accept that the immediate implementation of the HTA would necessarily contribute to the achievement of the ARORO in the case of Jemena or ActewAGL.

142    So far we have passed over the objection that the AER’s analysis is necessarily in error because it took the BEE to be a regulated entity, contrary to the subsequent findings of the Full Court. We note that in its Final Decisions the AER stated that its approach to the cost of debt would be applicable to an unregulated firm if it had a similar degree of risk to the service provider in providing regulated services; and that irrespective of whether a firm is regulated or not, efficient financing costs reflect the current (or prevailing) forward looking costs observed in capital markets. We accept the AER’s approach to be consistent with what we have said above about the efficient financing costs of the BEE, with the proviso that our discussion of the relationship between forward-looking costs and prevailing market conditions is presented later in these reasons.

143    We come now to the AER’s conclusion that immediate implementation of the HTA would not contribute to the achievement of the ARORO. That conclusion was based on the fact that the current market cost of debt is considerably below the average over the last nine years. Implementing the HTA immediately would, in the AER’s view, provide a rate of return higher than needed to compensate the applicants for their efficient financing costs.

144    The Tribunal considers that unless the applicants did have legacy debt as described – and they did not – it is indisputable that providing a return on debt partially based on the much higher interest rates of some years ago would give the applicants a windfall gain. This would be inefficient. The allowed revenue would be higher than needed to cover their costs. They do not have debt borrowed at the higher legacy interest rates, or any debt. There is no need – and it would be inappropriate – to adopt a fiction that a BEE would have had such debt obligations. Nothing in the ARORO could support an outcome so starkly at odds with the long term interests of consumers. It would simply be a transfer from consumers to the service providers. And the Full Court said, at [571]: “… 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.”

145    The applicants submitted that the Full Court’s reasons do not contemplate an assessment of the efficient financing costs over multiple past and future regulatory periods, and that this invalidates the AER’s reasoning in respect of windfall gains (and the use of the NPV = 0 analysis). The Tribunal considers that the Full Court’s reasons in respect of the construction of r 6.5.2(k)(4) are not relevant to that part of the AER’s reasoning.

146    The AER also considered that, had the trend in interest rates been up rather than down, immediate implementation of the HTA would under-compensate the service providers, and the Tribunal agrees. It cannot be doubted that no service provider would in those circumstances propose an immediate implementation of the HTA. The Tribunal considers it a relevant consideration whether an approach to estimating the return on debt is neutral with respect to past interest rates.

Claim 2: The AER was in error in considering that its transition to a full trailing average (TTA) could contribute to the ARORO

147    The AER concluded that its phased implementation of the HTA could contribute to achievement of the ARORO. Both applicants argue that it is a step in the AER’s reasoning to that conclusion that the on-the-day approach could contribute to achievement of the ARORO, and that it could not. Hence, the AER was in error. Embedded in these claims is a dispute about the prevailing cost of capital.

148    The applicants claim that the AER had incorrectly equated the concept of efficient financing costs in the ARORO with prevailing debt interest rates, ignoring other factors that influence the financing decisions of a BEE. For instance, the AER stated (ActewAGL Final Decision at page 3-17):

A key concept in the ARORO is ‘efficient financing costs’. Because the market for capital finance is competitive, a benchmark efficient entity is expected to face competitive prices in the market for funds. Therefore, we consider efficient financing costs are reflected in the prevailing market cost of capital (or WACC) for an investment with a similar degree of risk as that which applies to a service provider in respect of the provision of regulated services.

149    In reaching this conclusion the AER relied upon the advice of Partington and Satchell (Report to the AER: Discussion of the allowed cost of debt, 5 May 2016 (Partington and Satchell)). In particular, Partington and Satchell stated that the WACC should be calculated on the basis of prevailing interest rates:

The theory of finance (and common practice) is that in computing the weighted average cost of capital for use in NPV calculations it is the current required returns on debt and equity that should be used for the WACC. Thus with respect to the cost of debt it is the current cost of debt (as currently required in the market) that should be used in the WACC, not the historic cost of debt. (Partington and Satchell, page 14)

150    The AER summarised this and other advice in that report and consequently concluded:

(Partington and Satchell) interpret efficient financing costs as the opportunity cost of capital, which is a market rate of return for assets with a given level of risk. They advise the opportunity cost of debt is generally measured using the (appropriately benchmarked) yield to maturity. They also consider our use of a benchmark BBB+ credit rating and ten year term is appropriate.

We consider that productive, allocative and dynamic efficiency are advanced by employing a return on debt that reflects prevailing rates in the market for funds. This will also promote the long term interests of consumers in line with the National Electricity Objective/National Gas Objective (NEO/NGO). (ActewAGL Final Decision p.3-279, footnotes excluded)

151    A similar conclusion was included in the Jemena Final Decision, at page 3-278.

152    There was dispute over when the rates were or are “prevailing”, with it even being asserted that the AER never said when they were prevailing. The Tribunal considers it clear that Partington and Satchell are referring to rates in existence at each point in time. At any given point in time, those are the rates that should be used. This is not only consistent with the plain meaning of “prevailing” and the passages above (“current required returns”) but is also consistent with – and the only interpretation that is consistent with – the forward-looking nature of the cost of capital, remembering that it is an opportunity cost.

153    Consequently, the Tribunal concludes that the AER’s statement in the Final Decisions that “we consider efficient financing costs are reflected in the prevailing market” does not constitute an error as the applicants claim. The question is what guidance that provides for estimating a return on debt for incorporation in the rate of return on capital, given that market interest rates fluctuate, and how the AER applied the principle of a current required rate of return. That is the essence of the dispute about the NPV = 0 condition. The problems with the on-the-day approach that motivated the legislative changes arise largely from its use of the rate prevailing at the start of the regulatory period rather than at each point of time during the regulatory period.

154    As noted previously, the AER took the view that the return on debt must satisfy a zero NPV condition that does not appear explicitly in the NGL or NGR (or the NEL and NER). It states that:

... we consider a rate of return that meets the ARORO must provide ex-ante compensation for efficient financing costs. This return would give a benchmark efficient entity a reasonable opportunity to recover at least its efficient financing costs. This is a zero net present value (NPV) investment condition ... (ActewAGL Final Decision page 3-16; Jemena Final Decision page 3-18).

155    The Tribunal agrees that, as expressed above, the zero NPV condition is consistent with the NGO and RPP and considers that, consistent with achievement of the ARORO, it represents a criterion that should be satisfied, to the greatest degree practicable, by the approach used to assess the rate of return (cost of capital) and the returns on equity and debt. The AER then considered the zero NPV investment condition by reference to Partington and Satchell:

The zero NPV investment criterion has two important properties. First, a zero NPV investment means that the ex-ante expectation is that over the life of the investment the expected cash flow from the investment meets all the operating expenditure and corporate taxes, repays the capital invested and there is just enough cash flow left over to cover investors’ required return on the capital invested. Second, by definition a zero NPV investment is expected to generate no economic rents. Thus, ex-ante no economic rents are expected to be extracted as a consequence of market power. The incentive for investment is just right, encouraging neither too much investment, nor too little. (Partington and Satchell, p.14).

156    Translated into a property of the revenue allowance for a regulated service provider, this condition also restates the cost recovery requirements in the RPP, although limited to a specific investment over its life span. It would require that a regulated entity be allowed a stream of revenues over the life of the asset that, at the time the investment is made, have a net present value, measured at the prevailing WACC, equal to the cost of the asset.

157    The Tribunal considers that to be entirely conventional economic thinking in perfect accord with the Rules and is potentially a test that the AER could use, together with the other criteria specified in the NGR, to establish its choice between approaches. We disagree with the applicants that the AER elevated the criterion above the Rules.

158    At this stage we need to summarise. The objective is to implement the requirement that the ARORO be achieved by the determination of the allowed rate of return. That requirement comes from a view that the outcomes in workably competitive markets are to be emulated by reference to the efficient financing costs of a benchmark efficient entity for the service provider in question. The conventional view (which the Tribunal accepts) is that that involves a forward-looking cost of capital, which in turn implies the use of prevailing, or current, rates in the market; and that that in turn implies an NPV = 0 test. And all that comes back to use of the prevailing cost of capital leading to prices and revenues that provide the right incentives for the BEE to invest efficiently and for its services to be used efficiently, as they would be in a workably competitive market.

159    But this is about the BEE, not the service provider. The service provider is not free to make investment decisions based on its WACC. Rather, the Rules govern in detail what capital expenditure is allowed to be added to the capital base and thus generate revenue – as part of the return on the capital base building block – at the allowed rate of return. (The Rules relating to capital expenditure differ between the electricity and gas regulatory regimes.)

160    It bears reiterating that this whole process is targeted on making a forward-looking estimate of the overall rate of return. Thinking in terms of an allowance for debt is likely to lead to error, especially if the allowance is thought of as compensation for borrowing costs incurred in the past at interest rates then prevailing.

161    The applicants argued that their proposal was superior to the on-the-day approach because it updated one-tenth of the return on debt to the rate prevailing at the start of each regulatory year. But the transition to the HTA decided upon by the AER has this same property. The idea seems to be that new investment decisions during the regulatory period are given the correct signal because they will, by assumption, be funded by the one-tenth of total debt that is assumed to be borrowed in that year at the rate prevailing at the start of that year.

162    Thus, it is said that the rate of return that the firm has regard to in considering new investment is not its WACC, but the one-tenth of its WACC that is based on the latest debt interest rate. (In fact the argument was generally put ignoring that it is the rate of return that is relevant to investment decisions, not the return on debt, as if the investment were fully debt financed. Perhaps it was even assumed that the new investment is fully financed by debt.)

163    The Tribunal sees no support for that proposition in the economic theory of investment, and it is contrary to the advice from SFG Consulting, with which the Tribunal agrees, to the AEMC, which is discussed below. The proposition is part of a view that the return on debt can and should provide an allowance to compensate for past borrowings and simultaneously provide the signal for efficient new investment decisions. Indeed, it was claimed that there was no tension between those aims. The Tribunal disagrees. It is a misunderstanding of the role of the opportunity cost of capital.

164    The question arises: if the rate of return on capital is a forward-looking concept, how can historical interest rates on debt figure in the estimation of the return on debt, which requires prevailing rates? That question can equally be asked of the HTA and the on-the-day approach, because each incorporates past interest rates rather than current interest rates, at least as the regulatory period progresses. (For example, by the second year, in the on-the day-approach the rate used is “out of date”.) The answer, so it appears to the Tribunal, is that adherence to the prevailing rate principle cannot be achieved in practice, and nor can full consistency be achieved between the estimation of the return on equity and the return on debt.

165    This issue is set out starkly in the report to the AEMC by SFG Consulting at the time of the Rule change proposals:

13.    … The rule change … that the cost of debt allowed by the regulator should reflect interest costs on debt issued prior to the regulatory period … would represent a fundamental change to the principles which underpin the regulated rate of return. It would also embed a clear inconsistency between the principles which underpin the cost of debt and equity components of the regulated rates of return. Specifically, the cost of equity component would reflect the return equity holders require to encourage them to commit to a new project today; while the cost of debt component would reflect the return debt holders required to entice them to commit funds in a prior period.

14.    In all likelihood, the historical indexing approach … will produce cost of debt estimates which exhibit lower variation over time from one determination to the next. If this time-series stability promotes a regulatory objective then it should be considered. However, it should also be made clear that this computation does not represent an estimate of the prevailing cost of funds at the time of the determination.

15.    In short, the regulated rate of return will not be an estimate of the cost of capital. The principle which underpins the regulatory framework in Australia is to estimate a price which equates the present value of expected cash flows to the regulated asset base. If the regulated rate of return is set at a rate other than the cost of capital this will no longer hold.

166    This strong view is not reflected in the AEMC’s Rule Determination, where the dilemmas involved are discussed to a degree, but not entirely satisfactorily. The AEMC said, at page 44, in relation to the rate of return on capital:

It was determined that a robust and effective rate of return framework must be capable of responding to changes in market conditions. If the allowed rate of return is not determined with regard to the prevailing market conditions, it will either be above or below the return that is required by capital market investors at the time of the determination. The commission was of the view that neither of these outcomes is efficient, nor in the long-term interests of energy consumers.

167    Then, dealing with the question of consistency between the estimation of the return on equity and the return on debt, at page 64, it said:

If a historical average is used to estimate the allowance for the return on debt, there is no requirement for the resulting estimate, or any part of it, to constrain the estimate of the allowance for the required return on equity. The return on equity must reflect the prevailing conditions in the market as it is a forward-looking financial concept. The estimation of an historical average is a different exercise altogether, and one estimate does not constrain the other.

…

The final rule also requires that the return on equity estimation to have regard to the prevailing conditions in the market for equity funds. There is no corresponding requirement on the allowance for the return on debt, since the final rule (and the draft rule) allows the return on debt to be estimated with reference to an historical average as well to the prevailing conditions in the market.

The final rule makes clear the desirability of using an approach that leads to the consistent application of any estimates of financial parameters that are relevant to the estimates of, and that are common to, the return on equity and the return on debt.

and:

The Commission has decided not to reference “prevailing conditions in the market for funds” in the overall rate of return objective. This is because the objective would then be in conflict with the allowance for the return of debt where it is estimated on a trailing average basis. The Commission considers that it is clear from the return on equity provisions that the “prevailing conditions” concept continues to apply there, and that it also does apply, to the extent possible, for the return on debt estimation.

and, at page 66:

Also, references to “prevailing conditions” are not meant to exclude from any consideration historical or realised returns. However, the Commission does not consider that any drafting changes to the draft rule are required for the final rule to give effect to this. The final rule (as with the draft rule) is drafted in sufficiently broad terms to allow for such financial concepts to be taken into account for the purposes of developing appropriate estimation methodologies.

168    It seems to the Tribunal that there is a tension between the AEMC’s concerns that:

    first, if the overall allowed rate of return is not determined with regard to the prevailing market conditions then inefficient outcomes not in the long term interests of consumers would ensue; and secondly

    the return on equity must reflect prevailing conditions in the market as it is a forward-looking concept; and thirdly

    the estimation of the return on debt might not recognise prevailing conditions, and (therefore) could not be allowed to constrain the estimation of the return on equity; and fourthly

    the “prevailing conditions” concept continues to apply to the overall rate of return but could not be specified because there would then be a conflict with the estimation of the return on debt; and finally

    nevertheless the “prevailing conditions” concept does apply, to the extent possible, to the return on debt estimation.

The final passage quoted above deals with a different issue: the fact that making an estimate that reflects prevailing conditions may require the use historical data. That is certainly the case in estimating the return on equity. The situation in respect of the estimation of the return on debt is different. Current interest rates are set by competitive interactions between borrowers and lenders in the market. There is no need to have reference to past interest rates. In short, interest rates can be directly observed; the return on equity cannot.

169    The passage “the ‘prevailing conditions’ concept … does apply, to the extent possible, for the return on debt estimation” is telling. In the Tribunal’s view it properly reflects the fact that use of the HTA approach represents a departure from forward-looking rates of return, as set out in the SFG Consulting report. That approach was made possible by the Rule changes as an option to overcome difficulties with the application of the on-the-day approach, at least to some service providers. And as noted above by the Tribunal, the on-the-day approach also represents a far less than perfect application of the forward-looking concept, even where it can be implemented by a service provider with minimal difficulty through the use of capital market instruments.

170    These comments are consistent with those of the Tribunal in SAPN, at [239]:

Efficient investment and the regulatory approach to debt costs

NER cl 6.5.2.(k)(3) requires that in setting the cost of debt consideration be given to “the incentives that the return on debt may provide in relation to capital expenditure over the regulatory control period, including as to the timing of any capital expenditure”. While the historical trailing average approach is more compatible with actual debt management practices, it might be thought to be less consistent with determination of a cost of capital, and allowable revenue and prices, required for efficient investment decisions.

171    At this stage we remind ourselves why the HTA was introduced into the Rules as an option to be considered by the AER: to provide a smoother price path over the long run, across regulatory periods, possibly to reduce service providers’ risk, and because it received broad stakeholder support. It was not introduced because the AEMC considered that, in the normal case, it can be assured of representing the efficient financing costs of a BEE with similar risk, as the applicants claim. The departure from even the imperfect reliance on prevailing conditions embodied in the on-the-day approach needs to be kept in mind.

172    Furthermore, the AEMC placed weight on advice from SFG Consulting that:

For a given benchmark (in particular, the assumed credit rating and term to maturity), whether the regulatory allowance is based on the spot rate or an average of that spot rate over time, the average cost of debt will be the same over the long run. That is, changing to an averaging approach will not, in itself, systematically reduce the allowed return on debt.

and further:

Given a particular benchmark for the cost of debt, the long-run average regulatory allowance for the cost of debt will be the same regardless of which method is adopted in the Rules. This means that, over time, consumers will pay the same amount to provide a return to debt holders regardless of which approach is adopted in the Rules;

The Tribunal interprets those passages to mean that if the on-the-day approach were to be applied permanently, or the HTA were applied permanently, the average rate of return on debt would be the same in the long term. As we observed above (also quoting from an SFG report to the AEMC), the HTA would not, when incorporated in the WACC, provide an estimate of the opportunity cost of capital; and the on-the-day estimate would only be consistent with an opportunity cost of capital at the start of the regulatory period. Nor are the comments relevant to the issue of a transition.

173    We turn now to how the AER applied the NPV = 0 criterion. The key issue is as follows. The NPV = 0 criterion relates to the overall rate of return on capital. What can be deduced from it for the return on debt?

174    The AER relied in its Final Decisions on a mathematical analysis to show, it claimed, that the on-the-day approach and its transition to an HTA for the return on debt satisfy the NPV = 0 criterion, while immediate implementation of the HTA does not. This is in Attachment H. The applicants did not refer in detail to Attachment H. Rather, their attack was mounted at a level that did not require an assessment of the actual mathematical formulation. The AER followed the same approach in defending its conclusions. We largely take the same approach in the following discussion, although we have considered it desirable to examine Attachment H in detail.

175    First, even applying the NPV = 0 criterion using the overall rate of return presents difficulties. The RAB of a BEE will comprise a large number of assets of different values, ages and remaining operating lives, and will change in value and composition over time. Applying the Partington and Satchell zero NPV investment condition to those assets individually would suggest a stream of revenues based on all the initial valuations and prevailing rates at their time of acquisition. The aggregate revenue allowance for a return on capital building block implied by that stream could not be precisely implemented within the Rules because the framework requires a single rate of return to be applied each year to a separately defined RAB. Immediately, some approximation is inevitable.

176    The AER’s mathematical analysis of the zero NPV condition treated the RAB as if it was a single asset, as set out in the following passage:

We consider a rate of return that meets the ARORO must provide ex-ante compensation for efficient financing costs (we refer to this as ex-ante efficient compensation).

We consider ex-ante efficient compensation should result in the ex-ante allowed return on capital cash flows having a present value equal to the present value of the ex-ante efficient cost of capital cash flows required to finance the regulatory asset base (RAB). This means we must set, ex-ante, an allowed rate of return for a benchmark efficient entity such that the return on its investment (in its RAB) equals its efficient cost. This is a zero net present value (NPV) investment condition, which is a forward looking concept that shows a benchmark efficient entity is provided with a reasonable opportunity to recover at least efficient financing costs over the life of its investment (in its RAB). (footnotes omitted) (ActewAGL Final Decision page 3-282; Jemena Final Decision page 3-281)

177    The period “over the life of its investment (in its RAB)” is unknown, since the capital stock of a service provider has no fixed lifespan while the provider continues to operate. The AER’s mathematical analysis assumes a finite life stretching over multiple regulatory periods. For simplicity it assumes a fixed value of the RAB throughout the “life of the assets” (with capex exactly offsetting depreciation in each regulatory period and NPV calculations that collapse each regulatory period into a single period with a constant rate of return within that period, reset at the start of each regulatory period).

178    The AER commented:

We note the mathematical explanation in this section is a simplification of reality. We use it to demonstrate the principle that the allowed rate of return should be set (and periodically reset) such that the ex-ante allowed return on (and of) capital cash flows equals the ex-ante cost of a benchmark efficient entity's investment in its RAB (in present value terms). This gives service providers a reasonable opportunity to recover at least efficient financing costs over the term of the RAB.

179    The trouble is that the AER went close to assuming what it sought to prove algebraically. If the BEE can borrow for the regulatory period at the rate prevailing at the beginning of the regulatory period, then the NPV = 0 condition holds, in a more or less trivial way. If the rate of return facing the BEE varies during the regulatory period, as it surely must, then the analysis is at best an approximation. The Rules require that the return on equity be fixed for the regulatory period, and allow for that return on debt to be fixed, but do not require it.

180    The same shortcoming applies, in the Tribunal’s view, to the AER’s mathematical analysis purporting to show that its transition to the HTA satisfies the NPV = 0 condition.

181    The present value analysis necessarily deals with the (overall) rate of return. Translating insights from that analysis to precise results for the return on debt is not to be expected. The Tribunal considers, indeed, that it is intuitively obvious that without strong simplifying assumptions, the NPV = 0 criterion for the rate of return cannot be converted into a precise condition to be applied to the return on debt. The algebra would be intractable. In the Tribunal’s view, the AER’s analysis does not produce proofs of the sort claimed, viz. that the on-the-day approach and fully implemented trailing average approach ensure that the present value of the benchmark efficient entity is unchanged over the regulatory period or the life of the assets.

182    The Tribunal concludes that there are inconsistencies in the AER’s analysis, and that they tend to vitiate some of the reasoning it used to reach its conclusions. The next question is whether they render the AER’s conclusions amenable to the limited merits review available to parties under the NEL and the NGL.

183    First, the fact that the AER’s NPV = 0 analysis of the on-the-day method is unconvincing does not show that the approach could not contribute to the achievement of the ARORO. The Tribunal regards the NPV = 0 condition as a conceptual tool rather than a formula from which a precise value of the return on debt can be extracted. It is no more than another way of saying that the search required by the ARORO provisions in the Rules is for an estimate of the rate of return that provides ex ante compensation for efficient financing costs.

184    There is ample support for the AER’s conclusion that the on-the-day approach to estimating a return on debt can contribute to the ARORO. That is explicit in passages quoted above from the SFG Consulting report and from Partington and Satchell.

185    It is also consistent with the following reasoning of the Tribunal in SAPN, at [240]:

By using an on-the-day approach to determine cost of capital and thus allowable revenues, the outcome – in theory at least – is that the DNSP can expect a revenue stream consistent with what a competitive market would generate. Capital investment decisions made at that date which meet a positive NPV condition using the allowed rate of return would be efficient, based on current conditions in the market for funds. In contrast, if the DNSP assessed capital investment options by reference to the rate of return incorporating debt costs allowed from the trailing average approach, inefficient investment decisions could result.

186    Thus, despite the inconsistencies in its analysis and in some of its assertions, the AER was correct, the Tribunal finds, to take the view that the on-the-day approach could contribute to the achievement of the ARORO. It follows that the Tribunal does not accept that the AER’s view that its transition could contribute to the achievement of the ARORO must be in error because, as the applicants submit, it relies on the view that the on-the-day approach could do so.

187    Secondly, the fact that the AER’s NPV = 0 analysis of the on-the-day method is flawed does not imply that the immediately implemented HTA is better. It implies nothing for that approach. We have explained above why we consider that the AER was correct to consider that the immediately implemented HTA would be less likely than the on-the-day approach to contribute to the ARORO.

188    Thirdly, the AER considered, and this is not in contention, that a fully implemented HTA would, in due course, contribute to the ARORO. It sought to get the point of having the HTA approach fully implemented. That seems to be a widely-accepted goal of stakeholders following the AEMC’s Rule change process, although the AEMC did not consider it to be the only correct approach. The Tribunal has expressed a degree of scepticism about the HTA approach because it is a clear departure from an opportunity cost of capital, but accepts that no approach that does apply a true opportunity cost of capital is available, and the HTA approach may have certain advantages compared to the on-the-day approach.

189    In those circumstances, in the Tribunal’s view, the AER was bound to consider alternatives. Setting aside continuation of the on-the-day approach – because the widely-held view is that in due course a fully implemented HTA will be superior – the clear option is a combination of the on-the-day approach and a fully implemented HTA. The transition that the AER decided upon is such a combination. Both the on-the-day approach and the transition are specifically provided for in the Rules and it is clear that the AEMC considered that both could contribute to the ARORO in certain circumstances. It is not clear what the AEMC considered those circumstances would be, or how they would differ from each other, or even if they would differ. Rather, it considered that it was a matter for the AER in each particular case. No compelling argument was ever put why, if the fully implemented HTA approach can contribute to achievement of the ARORO, the transition to that outcome could not also contribute to the ARORO.

190    The Tribunal does not consider that the AER improperly had regard to extraneous matters, but in any case considers that those concerns played no substantial part in its reasoning.

Claim 3: The AER could only decide to impose a transition under Rule 87(11)(d) and was not in fact empowered to do so by that Rule.

191    As can be seen in the consideration above, the AER’s reasons for a transition to the HTA do not involve resort to r 6.5.2(k)(4).

Claim 4: The AER’s decision miscarries because of a failure to foreshadow its change of reasoning

192    The AER made preliminary decisions when the initial proposals put to it involved a transition to the HTA. Following those decisions, the applicants put revised proposals that involved immediate implementation of the HTA. The AER necessarily addressed the newly proposed approach in explaining why it was not changing its position from that of the RoR Guideline and the preliminary decisions. It had not used the NPV = 0 analysis until the Final Decisions.

193    If this was a departure from the procedural requirements of the Law and Rules, the Tribunal considers that that does not of itself constitute an error providing a ground for review. The grounds of review comprise only the three types of error set out at [11] above.

194    The AER’s failure to provide service providers with an opportunity to advance arguments against its reasoning could deprive it of useful inputs to its reasoning, thus making it more liable to error. Indeed, the Tribunal considers that this is the likely purpose of the provisions in question. But any error, if there is one, must exist apart from matters of procedure.

Conclusion

195    This ground of review fails. It is not necessary for the Tribunal to consider the issue of “materially preferable”.

GAMMA

Introduction

196    The building block approach and the post-tax revenue model (PTRM) have been described by the Tribunal in SAPN in terms which we adopt for introductory purposes. The Tribunal in that case said, at [113ff]:

The NER (r 6.3) specify that the DNSP’s revenue requirement is to be calculated using the “building block” approach. Rule 6.4 provides detail on the building block components. Two components are relevant in this matter. One is the allowed rate of return on capital which is to be calculated in accordance with cl 6.5.2. The NER (cl 6.5.2(d)(2)) specifies that 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 clause 6.5.3.” The second component (cl 6.4.3(b)(4)) is that “the estimated cost of corporate income tax is determined in accordance with clause 6.5.3”.

197    As the Full Court said in Australian Energy Regulator v Australian Competition Tribunal (No 2) at [752ff], the context is the determination of a regulated return using a PTRM based on a nominal vanilla WACC. The Rules require consistency in the way the relevant building blocks interact, that is, a post-company tax and pre-personal tax and personal costs basis. The nature of gamma is an estimate to be used in a model. We also note that, at [756], the Full Court said:

… it was not therefore a reviewable error for the AER to prefer one theoretical approach to considering the determination of gamma over another. This means that it is not an error of construction for the AER to focus on utilisation rather than on implied market value. This is not to say that the AER’s approach to gamma is immune from (limited) merits review. For example, if it erred in principle and thereby omitted to take into account a mandatory relevant consideration or arrived at a conclusion which meant that its decision was unreasonable within the meaning of s 71C(1)(d) of the NEL, the Tribunal may intervene. It may also be that the AER would err within s 71C(1)(c) or (d) in acting on economic learning outside the mainstream of that discipline, at least if it did so without explaining the basis for so doing.

198    Under the heading “Matters relevant to the making of building block determinations”, r 6.5.2 of the NER provided:

6.5.2     Return on capital

Calculation of return on capital

(a)    The return on capital for each regulatory year must be calculated by applying a rate of return for the relevant Distribution Network Service Provider for that regulatory year that is determined in accordance with this clause 6.5.2 (the allowed rate of return) to the value of the regulatory asset base for the relevant distribution system as at the beginning of that regulatory year (as established in accordance with clause 6.5.1 and schedule 6.2).

Allowed rate of return

(b)    The allowed rate of return is to be determined such that it achieves the allowed rate of return objective.

(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).

(d)    Subject to paragraph (b), 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.

(e)    In determining the allowed rate of return, regard must be had to:

(1)    relevant estimation methods, financial models, market data and other evidence;

(2)    the desirability of using an approach that leads to the consistent application of any estimates of financial parameters that are relevant to the estimates of, and that are common to, the return on equity and the return on debt; and

(3)    any interrelationships between estimates of financial parameters that are relevant to the estimates of the return on equity and the return on debt.

Return on equity

(f)    The return on equity for a regulatory control period must be estimated such that it contributes to the achievement of the allowed rate of return objective.

(g)    In estimating the return on equity under paragraph (f), regard must be had to the prevailing conditions in the market for equity funds.

Return on debt

(h)    The return on debt for a regulatory year must be estimated such that it contributes to the achievement of the allowed rate of return objective.

(i)    The return on debt may be estimated using a methodology which results in either:

(1)    the return on debt for each regulatory year in the regulatory control period being the same; or

(2)    the return on debt (and consequently the allowed rate of return) being, or potentially being, different for different regulatory years in the regulatory control period.

(j)    Subject to paragraph (h), the methodology adopted to estimate the return on debt may, without limitation, be designed to result in the return on debt reflecting:

(1)    the return that would be required by debt investors in a benchmark efficient entity if it raised debt at the time or shortly before the making of the distribution determination for the regulatory control period;

(2)    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; or

(3)    some combination of the returns referred to in subparagraphs (1) and (2).

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

(1)    the desirability of minimising any difference between the return on debt and the return on debt of a benchmark efficient entity referred to in the allowed rate of return objective;

(2)    the interrelationship between the return on equity and the return on debt;

(3)    the incentives that the return on debt may provide in relation to capital expenditure over the regulatory control period, including as to the timing of any capital expenditure; and

(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.

(l)    If the return on debt is to be estimated using a methodology of the type referred to in paragraph (i)(2) then a resulting change to the Distribution Network Service Provider's annual revenue requirement must be effected through the automatic application of a formula that is specified in the distribution determination.

Rate of Return Guidelines

(m)    The AER must, in accordance with the distribution consultation procedures, make and publish guidelines (the Rate of Return Guidelines).

(n)    The Rate of Return Guidelines must set out:

(1)    the methodologies that the AER proposes to use in estimating the allowed rate of return, including how those methodologies are proposed to result in the determination of a return on equity and a return on debt in a way that is consistent the allowed rate of return objective; and

(2)    the estimation methods, financial models, market data and other evidence the AER proposes to take into account in estimating the return on equity, the return on debt and the value of imputation credits referred to in clause 6.5.3.

(o)    There must be Rate of Return Guidelines in force at all times after the date on which the AER first publishes the Rate of Return Guidelines under these Rules.

(p)    The AER must, in accordance with the distribution consultation procedures, review the Rate of Return Guidelines:

(1)    at intervals not exceeding three years, with the first interval starting from the date that the first Rate of Return Guidelines are published under these Rules; and

(2)    at the same time as it reviews the Rate of Return Guidelines made under clause 6A.6.2.

(q)    For the avoidance of doubt, nothing prevents the AER from publishing the Rate of Return Guidelines made under this clause 6.5.2 in the same document as the Rate of Return Guidelines made under clause 6A.6.2.

199    We set out below, at [209], the provisions of r 6.5.3 of the NER.

200    Until 1987 Australia had what is called a classical taxation system under which company profits were subject to company tax, and dividends paid to shareholders were taxable as part of their income (eg as personal income tax). This double taxation of company profits created an incentive to retain profits rather than pay dividends, although the incentive depended on the nature of taxation on capital gains. Equivalently, there was a preference – now removed – for debt financing.

201    Under the current imputation tax system, shareholders can be given franking credits that attach to dividends and which recognise that company tax has been paid. If a dividend of $1 is fully franked (requiring the company in question to have paid sufficient tax), the franking credit is worth 43 cents. This reflects the fact that, to pay a dividend of $1, the company needed to have taxable income of $1.43, on which it paid 30 per cent company tax, i.e. 43 cents. (These figures are slightly rounded.) The 43 cents paid by the company to the Australian Taxation Office (ATO) is effectively given back to the shareholder by the ATO when the shareholder lodges a tax return and then taxed in the hands of the shareholder in the same manner as cash dividends. This can result in a tax refund to the shareholder.

202    The system of dividend imputation can be seen as a simple recognition that the company tax paid by the company was paid on behalf of its owners; indeed, pre-paid, where company tax may be thought of as a withholding tax. It removes the double taxation of company profits.

203    In discussions of this sort, it is often assumed that the shareholder pays personal income tax, but, in truth, the shareholder could be a company, or a trust, or a superannuation fund, or a charity. Ultimately, returns accrue to natural persons. The route by which they do so is one of the many, often unstated, complications in considering returns on investment.

204    Clearly the system of tax imputation affects the returns on an investment. The return on investment is central to the Australian (and many another) system of economic regulation (i.e. regulation concerned with the prices charged, revenues received and profits made by certain enterprises subject to such regulation, typically those exhibiting natural monopoly characteristics).

205    As explained by the AER in its Final Decisions (footnotes omitted):

The National Electricity Rules/National Gas Rules (NER/NGR) recognises that a service provider’s allowed revenue does not need to include the value of imputation credits. Under the NER/NGR, service providers are to recover revenue that compensates them for their efficient costs in providing regulated services. This includes, among other things, a return to be provided to investors (return on equity) that is required to promote efficient levels of investment. The more that imputation credits are valuable, the less return that investors require from dividends and capital gains. However, 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. Specifically, the NER/NGR require that the estimated cost of corporate income tax be determined in accordance with a formula that reduces the estimated cost of corporate tax by the 'value of imputation credits' (represented by the Greek letter, , 'gamma'). This form of adjustment recognises that it is the payment of corporate tax which is the source of the imputation credit return to investors.

206    Gamma is a ratio between nought and one: it is the value per dollar of imputation credits created. A dollar of imputation credits created cannot be worth more than a dollar or less than nothing. This part of the Applications for review largely revolves around the meaning of the phrase “value of imputation credits”.

207    It is useful to reproduce here a table provided by the applicants which sets out five categories of investor who may value imputation credits at less than face value, leading to gamma being less than one. The other columns in the table refer to types of evidence for estimating gamma, and are explained later.

(Footnote omitted).

208    We have set out at [18] above r 87A of the NGR which corresponds with r 6.5.3 of the NER.

209    Rule 6.5.3 was as follows:

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:

ETC_t = (ETI_t ×r_t) (1 – γ)

where:

ETI_t 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.

210    In addition, rr 87(4)(b)/6.5.2(d)(2) provide that 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 rr 87A/6.5.3.

211    The Tribunal set out the historical context of this part of the NER in Ausgrid as follows, at [1021]-[1029]:

Gamma was first introduced into the Australian regulatory context in 1998 by the ACCC in the first version of the National Electricity Code (the Code) under Part VII of the Trade Practices Act 1974 (Cth), in the context of calculating the weighted average cost of capital under the imputation tax system. It was defined as the “value of franking credits or imputation factor”. This definition of gamma was then introduced as part of the weighted average cost of capital formula in Ch 6 of the NER, which commenced operation on 1 July 2005.

In 2006, Ch 6A of the NER relating to transmission services was introduced, which broadly aligned the provision in Ch 6 relating to distribution services as part of the determination of the corporate income tax building block. It defined gamma as “the assumed utilisation of imputation credits, which is deemed to be 0.5”. From 1 January 2008, the definition of gamma as “the assumed utilisation of imputation credits” was adopted in Ch 6.

The current versions of the Rules are the result of a series of amendments culminating in the 2012 Rule Amendments. Unlike the Code which was initially implemented, and as the AER has observed (in Attachment 4 to each of the JGN Final Decision at p 4-6, the ActewAGL Final Decision at p 4-7, and the Networks NSW Final Decisions p 4-7), without footnotes) under the Rules:

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.

As the AER observes, the 2012 Rule Amendments effectively restore the wording of the definition of gamma that appeared in the first version of the NER (incorporating Schedule 6.1 of the Code), as the “value of franking credits”.

As observed, it is agreed between the parties that the change in the definition of gamma in the Rules from the “assumed utilisation of imputation credits” to the “value of imputation credits” does not change the meaning of gamma. Instead, the dispute concerns what that meaning is.

The Rules continue 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 provides 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.

[emphasis in bold added]

Rule 87(2)(4) of the NGR is in similar terms.

As noted elsewhere in these reasons, the 2012 Rule Amendments also introduced a requirement for the AER to periodically publish the RoR Guideline: r 6.5.2(m) of the NER and r 87(13) of the NGR. As required, the RoR Guideline sets out (among other things) the estimation methods, financial models, market data and other evidence the AER proposes to take into account in estimating the value of imputation credits under r 6.5.3 of the NER and r 87A of the NGR: r 6.5.2(n) of the NER and r 87(14) of the NGR. Also as noted elsewhere in these reasons, while the RoR Guideline is not binding on the AER in relation to making individual determinations, if the AER makes a decision that is not in accordance with the RoR Guideline, it must state its reasons for departing from the them: r 6.2.8(c) of the NER and r 87(18) of the NGR.

In accordance with the Rules, as indicated above, the AER published the RoR Guideline in December 2013, setting out (amongst other things) guidelines for estimating imputation credits. In keeping with accepted interpretation and practice, the AER’s gamma decision calculated the value of imputation credits as the product of the distribution rate for imputation credits and the utilisation rate of distributed imputation credits. The estimates outlined at p 23 of the RoR Guideline were an estimate of 0.5 for the value of imputation credits, based on a distribution rate of 0.7 and a utilisation rate of 0.7.

212    Contrary to the position in Ausgrid, in the present proceedings there was a difference between the parties as to the significance of the change in the definition of gamma in the Rules.

213    At [1083] the Tribunal stated: “It is … clear that the intent of the changes to the NER and the NEL was to allow the AER greater flexibility to adopt a more sophisticated approach to the cost of capital than previously envisaged by the NER and the NEL.”

214    The comments of the Tribunal in SAPN at [118]-[120] are also relevant:

The dispute over gamma needs to be placed in the context of changes made to the NER in November 2012 which changed the terminology used to describe gamma. Prior to that change, gamma was defined as “the assumed utilisation of franking credits”. Subsequently it has been defined (NER cl 6.5.3) as “the value of imputation credits”. No specific definition of what that term means, or how it is to be estimated, is given, other than that it needs to be consistent with the “vanilla WACC” approach.

Those rule changes also gave increased flexibility to the regulator in determination of the value to be chosen for gamma. “The current prescription of the gamma value of 0.5 in clause 6A.6.4 has also been removed to allow the regulator the ability to estimate an appropriate value that reflects the best available evidence at the time of a decision and would therefore result in a rate of return that meets the overall objective.” (AEMC, Rule Determination: National Electricity Amendment (Economic Regulation of Network Service Providers) Rule 2012, p 68).

Other than this statement, there appears to be no other explanation given for the change in terminology from “assumed utilisation” to “value”, although the latter term (even though not specifically defined) is more consistent with the flexibility given to the regulator and requirement (NER cl 6.5.2(e)) that:

In determining the allowed rate of return, regard must be had to:

(1)    relevant estimation methods, financial models, market data and other evidence;

(2)    the desirability of using an approach that leads to the consistent application of any estimates of financial parameters that are relevant to the estimates of, and that are common to, the return on equity and the return on debt; and

(3)    any interrelationships between estimates of financial parameters that are relevant to the estimates of the return on equity and the return on debt.

215    It might be noted that in 2010, the Tribunal had proposed that an expert study be undertaken to estimate theta, following which in May 2011 it decided (Energex No 5) that the appropriate value for theta was 0.35 and for the distribution rate was 0.7, giving a value for gamma of 0.25. That value of gamma was applied in regulatory determinations by the AER until it published the RoR Guideline.

What the AER decided

216    The AER’s Final Decisions set a value for gamma of 0.4, having rejected the applicants’ proposed value of 0.25. The effect was to reduce the revenue allowed for the cost of corporate income tax below the amounts sought by the applicants (or equivalently, as explained above, reduce the allowed rate of return on equity).

217    It is common ground that the value of imputation credits, gamma, can be considered to be the product of two ratios: the distribution rate (F), which is the proportion of imputation credits generated that is distributed to investors, and the utilisation rate (theta), which is the per dollar value to investors of imputation credits distributed.

218    The AER considered that it was not required to determine specific values for the distribution rate and utilisation rate separately. Rather, in each case it considered ranges, and consequently a range of values for gamma, from which it made a choice of what it considered the most appropriate estimate.

The distribution rate

219    The AER considered that a reasonable estimate of the distribution rate for listed equity was 0.75, and for all equity was 0.70.

The utilisation rate

220    As will be seen, the crux of the disagreement between the AER and the relevant applicants is about the utilisation rate. The AER stated that (footnotes omitted) “[t]here is no consensus among experts or regulators on the value of imputation credits or the techniques to use to estimate it” and that:

•    The role of the value of the imputation credits in the revenue building block framework suggests that the value of imputation credits is intended to reflect the value of imputation credits to investors in the benchmark efficient entity.

•    The framework developed in a 1994 paper by Officer is widely recognised as providing the basis for the value of imputation credits in the building block framework. A key implication of Officer's framework is that the value of imputation credits should be estimated on a before-personal-tax and before-personal-costs basis. This is consistent with a rate of return determined on a nominal vanilla (that is, a post-company tax pre-personal tax and costs) basis. Therefore, we view the value of imputation credits as the proportion of company tax returned to investors through the utilisation of imputation credits.

221    It considered that (footnote omitted):

•    or an ‘eligible’ investor, each dollar of imputation credit received can be fully returned to the investor in the form of a reduction in tax payable or a refund. Therefore, we consider that eligible investors have a utilisation rate of 1. Conversely, ‘ineligible’ investors cannot utilise imputation credits and have a utilisation rate of 0. It follows that the utilisation rate reflects the extent to which investors can utilise the imputation credits they receive to reduce their tax or obtain a refund.

•    This means imputation credits expected to be utilised should be valued at full face value on a pre-personal tax basis.

222    Elsewhere the AER said that “one dollar of claimed imputation credits has a post (company) tax value of one dollar to investors, before personal taxes and personal transaction costs.”

223    The Tribunal notes that there is a degree of looseness in these statements between “imputation credits expected to be utilised” and “claimed imputation credits”; and as to whether face value applies on a “pre-personal tax basis” or “before personal taxes and personal transaction costs”.

224    In reaching its views, the AER relied upon expert reports from Handley and Lally. Among other things these dealt with the development of the Officer framework by Monkhouse and Lally and van Zijl to interpret Officer’s seminal paper. One aspect of that development was that the utilisation rate is the weighted average, by wealth and risk aversion, of the utilisation rates of individual investors.

225    The AER stated (footnote omitted):

We do not contend that the Officer Paper is a “statute or a code”. However, as the Officer Paper underpins the inclusion of gamma in the corporate income tax formula in NER 6.5.3 and NGR 87A, it is fundamental to a coherent understanding of the role of gamma in the regulatory scheme.

The Officer Paper specifically identified gamma in its WACC formulae to be the “proportion of tax collected from the company which gives rise to the tax credit associated with a franked dividend”: It directly supports an interpretation of gamma which is focused on the utilisation or redemption of imputation credits, and an approach to theta which seeks to identify the proportion of investors that are eligible to utilise distributed imputation credits. So much is confirmed by Handley, who states:

It is clear from Monkhouse (1996) that the second parameter refers to the utilisation value of a distributed imputation credit. This parameter is commonly denoted and called theta. It is also clear from the post-tax basis of the regulatory framework (and the Officer and Monkhouse WACC frameworks) that the item of interest is more precisely described as the after-company-before-personal-tax utilisation value of a distributed imputation credit.

226    In the Final Decisions the AER identified three types of evidence for estimating the utilisation rate:

    the proportion of Australian equity held by domestic investors (the ‘equity ownership approach’)

    the reported value of credits utilised by investors in Australian Taxation Office (ATO) statistics (‘tax statistics’)

    studies that seek to infer from market prices the value to investors of distributed imputation credits (‘implied market value studies’).

227    For reasons it explained at length, the AER said that it placed most reliance on the first, less reliance on the second, and even less on the third. This judgement was supported by Handley and Lally.

228    The equity ownership approach – using the proportion of equity held by Australian investors – reflects the fact that foreign investors are unable to utilise imputation credits (because they do not pay tax in Australia at the personal or equivalent level). The AER placed significant reliance on the equity ownership approach because:

    it is well aligned with the definition of the utilisation rate in the Monkhouse framework

    it employs a relatively simple and intuitive methodology

    it uses a reliable and transparent source of data

    it provides estimates of the utilisation rate for investors in both all equity and only listed equity.

229    It went on to say (footnotes omitted):

We recognise the equity ownership approach does not take into account the existence of some domestic investors that do not hold their shares for 45 days at risk over the ex-dividend date (the 45 day rule). However, we consider it is unlikely to have a material impact on the utilisation of imputation credits by domestic investors. Importantly, no data has been presented that demonstrates a material impact.

We also recognise the equity ownership approach produces a range of estimates of the utilisation rate, and the upper end of this range is higher than the estimate produced from tax statistics (see discussion under ‘Tax statistics’ below). However, we do not consider this means the equity ownership approach is incorrect. This is because tax statistics, like the equity ownership approach, produce an estimate of the utilisation rate. These estimates are uncertain and dependent on the quality of the underlying data (this is a particular issue for tax statistics). As such, the true (unknown) value could be higher or lower. Following this, we do not consider the 45 day rule explains the difference in utilisation rate estimates between tax statistics and the equity ownership approach. We consider the difference is likely driven by estimation error in the taxation data. This view is supported by Lally. We also note Lally's statement that:

Thus, if foreign investors are recognized and absent any information on the terms other than the value weights on the RHS of equation (8), the equity ownership approach is not an upper bound on theta but an unbiased estimate.

230    The AER considered equity ownership proportions for both all equity and listed equity only, with the matching distribution rates. This gave corresponding ranges for gamma of 0.40 to 0.47, and 0.28 to 0.41, respectively, from which it selected 0.4, which is within the overlap of 0.40 to 0.41.

231    Tax statistics, which apply to all equity but not the subset of listed equity, gave a lower estimate of gamma of 0.34. The AER considered that the tax statistics were neither complete (ie did not include some credits that were in fact claimed) nor accurate, and placed less reliance on them. In this it relied on advice from Hathaway that “conclude[s] that the ATO statistics cannot be relied upon for making conclusions about the utilisation credits” and he would “urge all caution in using ATO statistics for any estimates of parameters concerned with franking credits”.

232    Hathaway’s advice was referred to and was reinforced by Handley and Lally.

233    In particular, Hathaway showed why the tax statistics based in franking account balances were reliable, but those based in dividends were not, with a very large discrepancy.

234    The only market study considered, provided by the applicants, gave a value of gamma of 0.26, but the AER considered that if that study and approach were to be accepted, the estimate would need to be adjusted upward, to 0.30. The market study necessarily applied only to listed equity. The AER’s rejection of that study was supported by Handley and Lally.

235    The AER justified its rejection of (or giving least weight to) drop-off studies, with references to expert advice, on the basis that:

    The results of these studies might not be reflective of the value of imputation credits to investors in the market as a whole. For instance, in dividend drop-off studies the value of imputation credits is determined by the marginal investor that trades around the ex-dividend date. There is no reason to assume this reflects the value that long term investors who provide capital to a benchmark efficient entity place on imputation credits in aggregate.

    It is only the value of the combined package of dividends and imputation credits that can be observed in the market. However, there is no consensus among experts on how to separate the value of dividends from the value of imputation credits.

    Dividend drop-off estimates of the utilisation rate do not produce a post-tax (pre-personal tax) estimate and therefore do not produce estimates consistent with the Officer post-tax framework. The dividend drop-off estimate is influenced by differential personal taxes, given that investors value capital gains more than dividend income due to preferential tax treatment of capital gains.

The errors claimed by the applicant parties to have been made by the AER

236    The applications for review in respect of gamma are substantially identical for the relevant applicants. They made joint submissions.

237    The alleged errors extended to about seven pages of text. In some instances, the errors were expressed to “cover all the bases”, involving one or more errors of fact, an incorrect exercise of discretion and unreasonableness, with the Tribunal apparently invited to make a choice.

238    The outline of oral submissions provided by Senior Counsel in making the joint submissions, while not addressing the specific NEL categories of error giving rise to grounds of review, crystallised the applicants’ claims.

239    The relevant applicants say that the AER erred in that, in respect of the distribution rate:

    It should have considered only the figure for all equity, not that for listed equity, because listed equity firms are not a good proxy for the benchmark efficient entity, because the benchmark efficient entity has 100 per cent Australian income. The receipt of foreign income had the potential to significantly alter the distribution rate.

In respect of the utilisation rate:

    The AER’s approach of assessing gamma as the utilisation rate or face value of the credits was not consistent with the ordinary meaning of the term “value of imputation credits”, with the assessment of the rate of return parameters, or with the national electricity objective.

    The AER was incorrect in concluding that the assessment on a pre-personal-tax and pre-personal-costs basis supports face value. Rather, the value of imputation credits needed to reflect their value to investors taking into account the effects of personal taxes and costs.

    The equity ownership approach did not measure the value to the investor of imputation credits. It merely identified and quantified one class of investor who could not utilise credits. It was even above a proper measure of the utilisation rate, because it did not account for Australian resident investors who could not or did not utilise credits. Equity ownership figures were therefore above an upper bound.

    Tax statistics also did not measure the value of imputation credits to investors, as they did not capture matters that reduce that value. They were therefore an upper bound. However, this upper bound was below the AER’s figure for gamma. Contrary to the AER’s position, the tax statistics provided a reliable upper bound for gamma. By contrast, the AER’s demonstration of the limited effect of the 45 day rule was based on an inaccurate part of the tax office database.

The applicants’ submissions

240    The two remaining applicants contended for a distribution rate of 0.7 and an utilisation rate of 0.35, which multiplied together and rounded gave their contended-for value of 0.25 for gamma. The distribution rate was taken from statistics for all equity. The utilisation rate came from a dividend drop-off study conducted by the consulting firm SFG for the energy networks. The study sought to estimate the value placed by the market on imputation credits by investigating the amount by which share prices fell when they go ex-dividend, ie when ownership of the share no longer conferred a right to a dividend and associated imputation credits.

241    The theory and practice of using dividend drop-off studies to estimate theta was considered in detail by the Tribunal in Energex (No 2) at [70]-[76] and again at [100]-[144]. The study replied upon by the applicants updates a previous SFG study, reported and relied upon in Application by Energex Limited (Gamma) (No 5) [2011] ACompT 9, which was produced in response to the Tribunal’s concerns with previous studies as expressed in Energex (No 2).

The distribution rate

242    The applicants submitted that the benchmark efficient entity was assumed to operate solely in Australia. But the distribution rate for listed equity was likely to be skewed by the practices of multinational firms with significant foreign earnings. Therefore, estimation of the distribution rate should use data for all equity.

The utilisation rate

243    The applicants accepted that the value of imputation credits should reflect the value of imputation credits to investors in the BEE, and that the Officer framework implied that, consistent with the rate of return being estimated on a nominal vanilla basis, the value of imputation credits should be estimated on a before-personal-tax and before-personal-costs basis. However, they disagreed with how the AER interpreted “before-personal-tax and before-personal-costs”. Senior Counsel for the applicants acknowledged the AER’s reliance on the advice of its experts, and submitted that the advice was wrong.

244    In general, the applicants’ submissions were not directed towards identifying whether the AER's decision exhibited one or more errors of fact, an incorrect exercise of discretion, or unreasonableness. Rather, they sought to show that the AER had reached wrong conclusions, or wrongly analysed an issue, and that the applicants’ proposed approach was correct and superior.

245    The applicants submitted that imputation credits should not be valued at face value but at market value, reflecting the value that an investor actually placed on the imputation credits. Using the table at [207] above, they pointed to five reasons why some investors did not place full face value on imputation credits and how the three estimation approaches dealt with them: only the market value studies columns had ticks in each box.

246    In support of their contention that imputation credits should be given a market value, the applicants submitted that:

    the “value of imputation credits” in r 6.5.3 of the NER should be given its ordinary meaning. This reflected its conceptual role – to ensure that investors were provided with an appropriate rate of return.

    the consistency required by r 6.5.2 required market value: other rate of return parameters were assessed using market value.

The submission regarding market value was supported by Professor Gray of SFG, who had provided the dividend drop-off study on which the applicants relied.

247    The applicants further submitted that the above matters were correctly identified in Ausgrid and that approach should be followed. Contrary to the observations in SAPN, the resolution of the issues for determination did not depend upon a distinction between a marginal investor approach and an average investor approach. Both were merely analytical tools for explaining the setting of an equilibrium price. The “utilisation rate” approach of the AER was not an average investor approach. This issue was a distraction from the correct issue, which was to identify what was required to be measured, and which of the AER’s sources of evidence measured that thing.

248    Moreover, the “issues” identified by SAPN with dividend drop-off studies were not weaknesses. The existence of the 45-day rule was correctly taken into account by the analysis, as was the value to imputation credits (and returns expected) by short, medium and long term traders.

249    At this stage it is necessary for the Tribunal to give some information about the two previous decisions by the Tribunal, referred to above, on essentially the identical issue before this Tribunal, the value of imputation credits.

250    In Ausgrid the Tribunal concluded, at [1100]:

We consider that, by … effectively defining the utilisation rate as the proportion of distributed imputation credits available for redemption, the AER has adopted a conceptual approach to gamma that redefines it as the value of imputation credits that are available for redemption. This is inconsistent with the concept of gamma in the Officer Framework for the WACC which underlies the Rules, 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.

251    That Tribunal also accepted the view of Professor Gray of SFG for the applicants and concluded, at [1073], that the AER’s reasoning (in that decision, but the reasoning is the same in this decision) ignored the fact that the other parameters in the WACC calculations were market values that already incorporated the effects of the differences in investors’ tax positions and transactions costs.

252    In SAPN, a differently constituted Tribunal found that the AER was not in error. The applicants contend at the hearing before the present Tribunal that that decision, which is subject to a judicial review application in the Federal Court, was in error. Essentially, the applicants submitted that the Tribunal there wrongly treated the issue as one of weighing evidence, and did not properly address the meaning of “value of imputation credits”.

253    In their written submissions, the applicants argued that a detailed debate between experts about how particular statements by Officer should be interpreted had distracted the AER’s attention from the proper inquiry as a matter of economic regulation; and that “Officer’s papers are not a statute or a code. Seeking to parse and analyse Professor Officer’s papers is not a substitute for a proper consideration of the role of imputation credits pursuant to the regime established by the NER.”

254    Nevertheless, the applicants then quoted a footnote in Officer’s paper to support their interpretation:

Where there is a market for tax credits one could use the market price to estimate the value of γ for the marginal shareholder, i.e. the shareholder who implicitly sets the price of the shares and the price of γ and the company’s cost of capital at the margin, but where there is only a covert market, estimates can only be made through dividend drop-off rates.

255    The applicants placed significance on the 2012 Rule Amendments. They said that the amendments brought the language of the rule – “value of imputation credits” in place of “assumed utilisation of imputation credits” – into line with regulatory practice. They saw some irony in the fact that, while the NER now referred to the “value of imputation credits”, rather than the “assumed utilisation of imputation credits”, the AER’s approach in applying the NER had gone in the opposite direction.

256    The applicants argued, by reference to the table reproduced at [207] above, that the equity ownership approach failed to take account of three of the reasons why some investors may not value imputation credits at face value; and the use of tax statistics failed to take into account one of the reasons. Therefore, both those approaches could only give estimates that were upper bounds on the best estimate.

257    The applicants also submitted that the AER erred in using equity ownership figures over the period since July 2000 (when the tax laws changed) rather than current figures; and that, in contrast with their position with the distribution rate, the AER should have used only figures for listed equity rather than also considering all equity. The argument was that the proportion of foreign equity in listed companies was more likely to reflect the proportion in the BEE; all equity includes small, privately held companies, which are not as accessible to foreign investors as the BEE (or as electricity and gas service providers in Australia, a number of which are privately and foreign owned).

258    The applicants also questioned the AER’s concerns about the completeness and reliability of tax statistics.

259    In their oral submissions, the applicants submitted that tax statistics could be used to avoid separate consideration of the distribution rate and the utilisation rate.

260    Using figures from Hathaway’s paper, relied on by the AER and referred to above, the applicants derived an estimate of gamma by dividing credits redeemed by credits created, giving a figure of 0.3, or 0.34 using more recent data, noting that in their contention this was an upper bound. According to the applicants, the figures for credits created and credits redeemed were reliable, and were considered to be so by Hathaway. They pointed to a reference to that effect.

The AER’s submissions

261    In its written submissions, the AER largely repeated the reasoning in its Final Decisions. In that reasoning the AER relied on expert economic advice, in particular from Handley and Lally. For example, Handley’s advice was that “the per dollar value of an imputation credit γ gamma should be measured prior to any personal tax on the credit and prior to any personal costs associated with the receipt of the credit”. (The AER noted that “[a]lthough the term ‘personal’ is used, we note that classes of investors other than individual persons can value imputation credits (for example, superannuation funds and charities). Therefore, an alternative characterisation might be ‘before-investor-tax’ and ‘before-investor-costs’.”)

262    The AER pointed to a series of statements by the Tribunal over some years to the effect that there was no one way of measuring gamma, and that the estimation methodology was subject to continual change. Internal consistency was a key requirement, and its approach achieved that, being supported by a number of economic experts, and was not found to be in error by the Tribunal in SAPN.

263    The weighting of evidence – from the three approaches mentioned above – was, in the AER’s contention, a matter of regulatory judgement. It could not be concluded, the AER submitted, that the AER’s position was wrong and that the applicants’ – which relied entirely on a dividend drop-off study – was right.

264    However, in oral submissions, Counsel for the AER used an argument that seemed to be different from the reasoning in the Final Decisions, and to break new ground. Certainly it was characterised that way by Senior Counsel for the applicants. It amounted to an argument about double counting.

265    Counsel for the AER accepted that dividends and imputation credits can be valued by those entitled to them at less than face value, and indeed adopted the applicants’ table, reproduced at [207] above.

266    Using a post-tax valuation framework, or revenue framework, both the cash flows and the cost of capital were estimated before investor taxes and before investor costs. The observed returns based on stock prices reflected the full range of investor taxes and investor costs that affected the investment. To the extent that they diminished the value of an equity investment, that would be reflected in the asset price and the required return be higher as a result.

267    The AER contended that because those costs were capitalised into share prices and reflected in the return on equity, investors were compensated for those costs through the allowed return on equity, which were already in the allowed cash flows under the building block framework.

268    The applicants’ approach, as characterised by Counsel for the AER, was to provide compensation for the same costs a second time through the allowance for company tax.

269    The submissions referred to so far deal with the in-principle and practical case against a reliance, much less a complete reliance, on the dividend drop-off study put forward by the applicants. The AER also responded to specific points raised by the applicants regarding the equity ownership and tax statistics approaches.

270    The AER argued that neither the equity ownership approach nor the use of tax statistics provided point estimates. They generated ranges. Therefore, a particular estimate within the range could not be considered an upper bound. This was consistent with expert economic advice.

271    It pointed out that the applicants argued for the use only of all equity (and not listed equity) data in estimating the distribution rate, and the use of only listed equity (and not all equity) in estimating the utilisation rate. In both cases, the applicants’ argument was by reference to the BEE. The AER pointed out that in each case it considered both listed equity and all equity, and matched the corresponding estimates.

272    Regarding the period over which it estimated the proportion of foreign equity ownership (since July 2000), the AER argued that there was no reason why the most recent period would provide the best estimate for the forthcoming regulatory period. The period it chose was consistent with the period used in the applicants’ own dividend drop-off study.

273    The AER argued that it had specifically considered the category of investors who were unable to redeem imputation credits because of the 45-day rule, but were not captured in equity ownership approach estimates, and concluded that it was not material.

274    Regarding the use of tax statistics, the AER argued that it was clear from the Hathaway report that the figure for credits redeemed in fact came from the unreliable dividend data rather than the reliable franking account balance data: a table in Hathaway’s report appeared to show that. This would be consistent with Hathaway’s clear warnings against using the ATO data for drawing any conclusions about the utilisation of franking credits.

275    The AER received advice from Lally on the Tribunal’s Ausgrid decision. In that advice Lally stated, in relation to the value of imputation credits, “…the NER is not the arbiter on this matter. Nor is the ACT … Instead, one must look to the relevant economic literature.” He then went on to analyse Officer’s paper and later derivations of gamma under that framework.

276    The AER received Lally’s most recent advice too late in the piece for the network distributors to have had an opportunity to respond to it before the AER made its Final Decisions. However, the applicants made submissions on it.

277    The AER drew attention to the decision of the Tribunal in SAPN that the AER did not err in regard to gamma. Counsel relied more on that Tribunal’s conclusions than on its detailed reasoning.

The Full Court’s decision

278    As mentioned above, after the hearing in these proceedings, the Full Court’s decision on its judicial review of the Tribunal’s Ausgrid decision was handed down. That decision is Australian Energy Regulator v Australian Competition Tribunal (No 2). The parties were invited to make submissions on the effect of the decision, and did so.

279    The applicants made the following points.

280    At [751]-[755], the Full Court concluded that the Tribunal had erred in law in its construction of r 6.5.3 of the NER, in failing to construe the phrase as a whole in its context and having regard to the subject matter of the exercise. The Full Court said that it accepted the AER’s submission that the context was the determination of a regulated return using a PTRM based on a nominal vanilla WACC, and accepted the AER’s submission that the Rules required consistency in the way the relevant building blocks interacted, that is, a post-company tax and pre-personal tax and personal costs basis. The Full Court observed that, as the nature of gamma was an estimate to be used in a model, the context related to a statutory model rather than the value of something which existed, and that the Tribunal was distracted by the apparent simplicity of the concept of market studies and data into mistaking what was to be estimated as real in a market rather than as estimates within a model. The Full Court went on to state, at [755], that:

… we accept the AER’s submission the Tribunal’s approach to gamma was underpinned by a misunderstanding on its part about how return to investors was conceptualised in a WACC framework. In our opinion the Tribunal assumed that other parameters in the WACC calculations were market values that already incorporated investors’ tax positions and transaction costs but that misconstrued the “post-tax” framework. The rules required gamma to be determined consistently with return on equity.

281    Applying that approach in the present case, one would need to consider the content of the model. In this regard, it was apparent that the Full Court had before it different information from the information which was before this Tribunal as to the nature of the model.

282    In the present case, the position adopted by the AER was that its rate of return model did not ignore, or fail to incorporate, personal transaction costs. The AER had conceded that the model needed to account for them in some way. Rather, the position adopted by the AER was that they were already taken into account in the assessment of the return on equity because they affected asset prices, which in turn affected the measurement of the required return on equity. This was explained by Counsel for the AER at pages 627-639 of the transcript. It was a lengthy explanation, but required reading in full to see how the operation of the model was explained.

283    The applicants submitted that the AER’s explanation was directly contrary to the observations of the Full Court at [755] as to the content of the model, extracted at [280] above.

284    That difference arose from the different material that was before the Full Court and this Tribunal. So even if the approach of the Full Court was applied (to treat imputation credits as a function of a model, rather than as something that was “real”), that directed attention to the content of the model and to the evidence on that model in the individual case.

285    Likewise, in the present case, Counsel for the AER, at pages 648-651 of the transcript, discussed the decision of the Tribunal in Ausgrid in terms that made it clear that the Tribunal was correct to assume that other parameters “already incorporated investors’ tax positions and transaction costs”, but stated that to account for them in relation to imputation credits would involve some measure of double counting.

286    After briefly restating the argument they had made in the hearing, the applicants submitted that the position before the present Tribunal was:

(a)    the AER accepted that matters affecting the value of imputation credits, such as transaction costs, are relevant and need to be taken into account – the model did not exclude them;

(b)    the AER said they were taken into account in the assessment of asset prices, and therefore did not need to be included in the gamma calculation;

(c)    however, the AER was incorrect in stating that transactions costs and other matters affecting the value of imputation credits were already reflected in the return on equity calculation, for the reasons developed by the applicants in oral reply.

287    In these circumstances, the applicants submitted, the decision of the Full Court did not determine the Tribunal’s assessment of the gamma topic. Rather, it emphasised that the correct approach to gamma depended upon the content of the model used to calculate the overall rate of return of the regulated business. That model, properly understood on the material before the Tribunal in the present case, required two adjustments for the value of imputation credits: to gross up post-1987 return on equity calculations to exclude the value of imputation credits (so as to produce an overall average market risk premium, and thus an overall return on equity, unaffected by imputation credits), and then to deduct the value of imputation credits. At each stage, matters affecting the value of imputation credits must be taken into account under the model.