Origin Energy Electricity Ltd & Anor v Queensland Competition Authority & Anor [2012] QSC 414

ADMINISTRATIVE LAW - JUDICIAL REVIEW - GROUNDS OF REVIEW - DECISION NOT AUTHORISED BY THE ENACTMENT - where the Minister made a delegation to the first respondent to determine regulated electricity tariffs - whether the terms of the delegation were inconsistent with the power to delegate - whether the delegation was invalid for the purposes of making a price determination under the Act

ADMINISTRATIVE LAW - JUDICIAL REVIEW - GROUNDS OF REVIEW - RELEVANT AND IRRELEVANT CONSIDERATIONS - where pricing entity makes price determination of regulated electricity tariffs - where the Act requires the decision-maker to have regard to 'the actual costs of making, producing or supplying the goods or services' - whether the decision-maker bound to take into account costs of producing electricity acquired through the NEM - whether the decision-maker bound to take into account costs of power purchasing agreements - whether the decision-maker bound not to take into account estimation of costs of a 'representative retailer' in making its determination

Acts Interpretation Act 1954 (Qld), s 14A

Administrative Decisions (Judicial Review) Act 1977 (Cth),

s 5

Electricity Act 1994 (Qld), s 3, s 26, s 30, s 33, s 38, s 40, s 40A, s 47, s 48A, s 48C, s 48D, s 48E, s 48F, s 48H, s 48I,

s 52, s 55D, s 89A, s 90, s 90A, s 90AA, s 329

Electricity Price Reform Amendment Act 2011 (Qld), s 7,

Electricity Regulation 2006 (Qld), s 104, s 107, s 124

Judicial Review Act 1991 (Qld), s 20, s 23, s 43,

Queensland Competition Authority Act 1997 (Qld), s 10

AGL Energy Ltd & anor v Queensland Competition Authority [2009] QSC 90, considered

Alcan (NT) Alumina Pty Ltd v Commissioner of Territory Revenue [2009] HCA 41; (2009) 239 CLR 27, cited

Attorney-General for the Northern Territory v Minister for Aboriginal Affairs [1989] FCA 159; (1989) 23 FCR 536, cited

Australian Retailers Association and ors v Reserve Bank of Australia [2005] FCA 1707; (2005) 148 FCR 446, referred

Bendixen v Coleman Scott and Croft [1943] HCA 40; (1943) 68 CLR 401, considered

Chandra v Webber and anor [2010] FCA 705; (2010) 270 ALR 393, referred

East Australian Pipeline Pty Limited v Australian Competition and Consumer Commission [2007] HCA 44; (2007) 233 CLR 229, cited

Foster v Minister for Customs and Justice [2000] HCA 38; (2000) 200 CLR 442, cited

Houssein v Seretary of Industrial Relations and Technology (NSW) [1982] HCA 2; (1982) 148 CLR 88, considered

Kahn v Minister for Immigation and Ethnic Affairs [1987] FCA 457; (1987) 14 ALD 291, considered

Kioa v West [1985] HCA 81; (1985) 159 CLR 550, cited

McCormack v Deputy Commissioner of Taxation [2001] FCA 1700; (2001) 114 FCR 574, referred

Minister for Aboriginal Affairs v Peko-Wallsend Ltd [1986] HCA 40; (1985-1986) 162 CLR 24, cited

Minister for Immigration and Citizenship v SZJSS & Ors [2010] HCA 48; (2010) 243 CLR 164, cited

Minister for Immigration and Multicultural Affairs v Yusuf [2001] HCA 30; (2001) 206 CLR 323, considered

Neat Domestic Trading Pty Limited v AWB Pty Limited [2003] HCA 35; (2003) 216 CLR 277, considered

Plaintiff M47-2012 v Director General of Security [2012] HCA 46; (2012) 292 ALR 243, referred

Rathborne v Abel (1964) 38 ALJR 293, considered

R v Hunt; ex parte Sean Investments Pty Ltd [1979] HCA 32; (1979) 180 CLR 322, considered

R v Lyon [1906] ArgusLawRp 35; (1906) 3 CLR 770, considered

R v Toohey; parte Meneling Station Pty Ltd [1982] HCA 69; (1982) 158 CLR 327, referred

Tobacco Institute of Australia v National Health and Medical Research Council [1996] FCA 1150; (1996) 71 FCR 265, cited

Trawl Industries of Australia Pty Ltd v Effem Foods Pty Ltd (1992) 27 NSWLR 326, considered

Upper Hunter County District Council v Australian Chilling and Freezing Co Ltd [1968] HCA 8; (1967) 118 CLR 429, considered

Vardon v The Commonwealth [1943] HCA 30; (1943) 67 CLR 434, considered

Weal v Bathurst City Council [2000] NSWCA 88, cited

P A Freeburn SC and A Scott for the second respondent

[1] Jackson J: The validity of a decision made under an enactment is challenged by judicial review. The enactment is 90(1) of the Electricity Act 1994 (Qld) ("the Act"), as affected by s 90(5). The decision is the price determination for the tariff year commencing 1 July 2012 and ending 30 June 2013, made by the first respondent, Queensland Competition Authority ("QCA") ("the Determination").

[2] The Determination was made pursuant to the delegation in writing made by the second respondent Minister ("the Minister") on or about 8 May 2012 ("the Delegation") amending an earlier delegation made by the prior minister on or about 22 September 2011.

[3] The challengers are the applicants Origin Energy Electricity Ltd ("Origin") and Sun Retail Pty Ltd ("Sun") which is a related corporation of Origin.

[4] The challenge is made under the Judicial Review Act 1991 (Qld) ("JR Act"). Procedurally, the applicants invoke the pathways of a statutory order of review, under s 20, or a prerogative order or declaration, under s 43. However, it was not suggested in argument that the scope for relief was greater for an order by way of prerogative order or declaration than in the case of a statutory order of review, so it is unnecessary to consider a prerogative order or declaration separately.

[5] In substance, the applicants advance two legal grounds for a statutory order of review: first, that the Determination was not authorised by the enactment under which it was purported to be made because the Delegation was invalid ("the Delegation ground"), under s 20(2)(d) of the JR Act;[1] secondly, that the making of the Determination was an improper exercise of the power conferred by the enactment under which it was purported to be made, because QCA failed to take a relevant consideration into account or took an irrelevant consideration into account in the exercise of the power to make the Determination ("Relevancy grounds")[2], under ss 20(2)(e), 23(a) and 23(b) of the JR Act.[3] There are a number of ways in which the applicants deploy these two legal grounds of challenge.

[6] Part of the evidence tendered by the parties identified the material which was before QCA as decision-maker or taken into account by it. It is axiomatic in a proceeding for judicial review based on the Relevancy grounds that only that kind of evidence is receivable, in general. As it was succinctly put in the leading case on the point, Attorney-General for the Northern Territory v Minister for Aboriginal Affairs:[4]

"The admissibility of evidence not before the decision-maker depends upon the grounds of review on which the applicant relies before the Court**.** In the case of some grounds of review (for example, if the decision-maker failed to take into account a relevant consideration) (s5(2)(b) of the Judicial Review Act) or took into account an irrelevant consideration (s5(2)(a)) it is difficult to see the relevance of material not before the decision-maker. Other grounds of review (for example, unreasonable exercise of the power (s5(2)(g)) will generally, lead to the evidence consisting primarily of the material before the decision-maker.

Where the ground relied upon is error of law (s5(1)(f)) the trend of judicial opinion is that the evidence before the Court is confined to the material before the decision-maker..."

[7] Notwithstanding this, the parties tendered evidence which was not confined to that which was before QCA. Some of it was justified or sought to be justified by the proposition that it was tendered in explanation of matters of a technical nature in order to understand the matters relevant to consideration of the questions raised for decision. Regrettably, there was no serious attempt to limit the exercise to that purpose in the reports of the experts which were tendered on either side. Much of the evidence given in the reports was potentially additional material as to the author's views about particular matters.

[8] Even in the evidence of the lay witnesses, there was evidence of facts which were not before the decision-maker which ought not to have been tendered. For example, the applicants sought to tender further evidence as to the calculation of actual costs of their business operations, for the purpose of seeking to prove amounts to be used in comparison to the amounts determined by QCA in the tariffs for the price determination, so as to support a factual contention that the tariff amounts were too low. At no stage was any rational basis given for seeking to do so. Nor does one appear from the circumstances of the case. For example, there was no contention by the respondents that relief should not be granted as a matter of discretion under the JR Act, because any error, if made, was trivial.

[9] It is no surprise, therefore, that numerous objections were notified as proposed to be taken, on either side, as to the admissibility of large parts of the evidence. In this state of play, the parties dealt with some of the objections as the witnesses were called, but otherwise informed me that they were content for the objections to be resolved at the time of addresses or decision. Neither party orally addressed any particular objection in final addresses. The applicants referred to a case which illustrated the admissibility of evidence which serves to explain the existing evidence. As I have said, regrettably, the parties did not confine themselves to tender of evidence of that limited kind.

[10] Having now read all the affidavits and reports which were tendered on that footing, I have resolved not to decide all the notified objections for the purpose of deciding the case. Separating the admissible from the inadmissible in the expert reports, in particular, would be an inutile endeavour. Instead, I have taken the approach which accords with the law as I understand it, and had regard only to that which was before the decision-maker for the purpose of deciding the case and avoided any factual assertions or opinions contained in the affidavits or reports which were not before the decision-maker or taken into account in the decision making process.

(1) The Minister must, for each tariff year, decide (a price determination) the prices, or the methodology for fixing the prices, that a retail entity may charge its non-market customers for all or any of the following -

(b) charges or fees relating to customer retail services;

• credit card surcharges for payments for the services

(c) other goods and services prescribed under a regulation."

[12] The central concepts of "a retail entity", "non-market customers" and "customer retail services" require explanation, which unfortunately must be more than brief.

[13] The objects of the Act are set out in Ch 1 Pt 2. Section 3 provides:

(a) set a framework for all electricity industry participants that promotes efficient, economical and environmentally sound electricity supply and use; and

(b) regulate the electricity industry and electricity use; and

(c) establish a competitive electricity market in line with the national electricity industry reform process; and

(d) ensure that the interests of customers are protected; and

(e) take into account national competition policy requirements."

[14] Chapter 2 of the Act is divided into four parts. Part 1 defines the "electricity industry" and "electricity entities". The electricity industry is "the industry involved in generating, transmitting, supplying and selling electricity in the State". The participants in the electricity industry are defined as "generation entities", "transmission entities", "distribution entities" and "retail entities".

[15] Part 2 of Ch 2 identifies customers and their types, so that a "customer" is a person who "receives or wants to receive a supply of electricity for premises from an electricity entity" and a "non-market customer" for premises is a customer "other than a [customer prescribed under a regulation to be a market customer] for the premises".

[16] Part 3 of Ch 2 of the Act defines a "generation entity" as a person who holds a "generation authority". By s 26, a generation authority authorises its holder "to connect the generating plants stated in the authority to the transmission grid or supply network stated in the authority" and "to sell electricity ... through the spot market in accordance with the National Electricity Rules ...".

[17] Part 4 of Ch 2 of the Act defines a "transmission entity" as a person who holds a "transmission authority". By s 30, a transmission authority authorises its holder "to operate the transmission grid stated in the authority", but by s 33 it is a condition of a transmission authority held by a transmission entity that operates a regulated transmission grid that the transmission entity "must not buy or sell electricity directly or indirectly".

[18] Part 5 of Ch 2 of the Act defines a "distribution entity" as a person who holds a "distribution authority". By s 38, a distribution authority authorises its holder to supply electricity using its supply network within its distribution area. By s 40, a customer who owns or occupies premises may make an application for the provision of "customer connection services" to the premises and by s 40(2) the application may be made by retail entity for the customer. Section 40A provides for an obligation on the part of the distribution entity to provide services described as the "connection obligation".

[19] Part 6 of Ch 2 of the Act defines a "retail entity" as a person who holds a "retail authority". By s 47, a retail authority authorises its holder to provide "customer retail services" under the terms of the authority. "Customer retail services" is defined in the Dictionary, as follows:

"customer retail services, for premises, means the sale of electricity to the premises."

[20] By s 48A of the Act, a retail authority authorises its holder to provide customer retail services either in a particular area stated in the authority or to any customer in the State, subject to certain restrictions or conditions. By s 48C, the customer who owns or occupies premises may make an application to a retail entity for the provision of customer retail services to the premises. By s 48D or s 48E, in some circumstances, a retail entity must provide the customer retail services applied for to the premises. Section 48F describes those obligations as the "retail obligation" which is made subject to s 48H and s 48I.

[21] Division 3 of Pt 6 of Ch 2 of the Act provides for a "retail contract" which is any contract under which a retail entity agrees to provide customer retail services to a customer's premises. Under sub-division 2, if there is no "negotiated retail contract", a retail contract is made the subject of standard terms.

[22] Division 4 of Pt 6 of Ch 2 of the Act provides for conditions of a retail authority, including by section 55D(a) that the retail entity must consider both the demand side and supply side options to provide, as far as technically and economically practicable, for the efficient use of electric energy. There are a number of other conditions stated in s 55DA, s 55DB, s 55DC, s 55E, s 55G and s 55GA.

[23] Part 6A of Ch 2 of the Act provides for a coordination agreement between a distribution entity and a retail entity.

[24] Part 2 of Ch 4 of the Act provides for pricing by the mechanism of a price determination made under s 90(1).

[25] Under s 89A of the Act, the "pricing entity" referred to in s 90(1) may be either the Minister or, if the Minister delegates the function of the Minister under s 90(1), QCA. By s 90(2), the price determination must be in the form of a tariff schedule. By s 90(3), a determination may be made from time to time, not just once a year, and may include network charges, but cannot be made for distribution non-network charges, such as a disconnection fee, a reconnection fee or a meter test fee.[5]

[26] Section 90(5) is central to the arguments in the case. It provides:

"(5) In making a price determination, the pricing entity -

(i) the actual costs of making, producing or supplying the goods or services;

(ii) the effect of the price determination on competition in the Queensland retail electricity market;

(iii) if QCA is the pricing entity - any matter the pricing entity is required by delegation to consider; and

(b) may have regard to any other matter the pricing entity considers relevant."

[27] The Delegation ground of invalidity is based on the requirements of s 90AA. That section provides:

(1) The Minister may delegate to QCA all or any of the Minister's functions under section 90(1).

(2) The delegation may state the terms of reference of the price determination.

(3) The terms of reference may specify the following -

(a) the period for which the price determination is to apply;

(b) the time frame within which QCA is to make and publish reports on the price determination;

(c) the particular policies or principles QCA is to consider when making the price determination;

(d) the matters QCA must consider when working out the notified prices and making the price determination;

the particular methodology to be used to determine the prices

(e) the consultation requirements QCA must comply with before making the price determination.

(a) apply generally to all tariffs or be limited in its application by reference to specified exceptions or factors; or

(b) apply differently according to different factors of a specified kind."

[28] The Delegation was preceded by a number of consultative processes involving both the Minister and QCA, which informed the Minister's exercise of power or purported exercise of power in making the Delegation. They included, under the 22 September 2011 iteration of the delegation, that QCA was required to publish a paper outlining its draft methodology for calculating the R (energy and retail costs) component no later than December 2011 and a report on its draft determination of regulated retail electricity tariffs on 30 March 2012.

[29] However, because the only question for the court on judicial review is whether or not the Delegation was invalid as one "not authorised by the enactment under which it was purported to be made", it is unnecessary to recount or explore those steps in any detail at this point. The Delegation in its final form must stand, or fall, according to its own terms.

[30] Relevantly, it provided that, pursuant to s 90AA(1), the Minister referred to QCA the determination of regulated electricity tariffs, excluding the general supply residential tariff, to apply from 1 July 2012 to 30 June 2013 in accordance with the requirements which were set out in the terms of reference. The terms of reference were divided into three parts headed "1 Matters to be considered", "2 Consultation" and "3 Timing".

[31] The opening paragraph, under matters to be considered, directed that "in calculating the delegated regulated electricity tariffs ... [QCA] should ensure its price determination has regard to ... the actual costs of supplying electricity ...".

[32] Under the subheading "Methodology for calculating regulated retail tariff prices" the Delegation directed QCA "to the extent possible, [to] base its determination on a Network (N) plus Retail (R) costs build up approach to setting notified prices, where N (network cost) is treated as a pass-through and R (energy and retail cost) is determined by" QCA.

[33] As to energy costs, the Delegation provided that "the energy cost component of each regulated retail tariff should include the cost of purchasing energy, environment and renewable energy costs, energy losses and National Electricity Market fees". The Delegation also directed that "in calculating the energy cost component, [QCA] must consider ... the cost of energy ...".

[34] The challenge to the validity of the Delegation is made in the applicants' written submissions on two bases, namely that:

(a) "on the proper construction of the Delegation it purports to state exhaustively the matters that QCA should consider in making its determination"; and

(b) "the terms of the Delegation are more limiting than those provided for in s 90(5) in that they require consideration of the costs of supplying electricity and the costs of purchasing electricity, but do not contain any requirement to consider the costs of producing electricity".

[35] In support of the first of those arguments, the applicants submitted: "that appears to be the way in which ACIL Tasman and the QCA understood the Delegation". However, that contention is irrelevant to whether the Delegation is invalid, which does not depend on how anyone, in fact, interpreted the Delegation. It depends on whether the Delegation, properly construed, is not authorised by the power contained in s 90AA.

[36] The applicants submitted that the question was whether the delegated instrument was repugnant to the Act which confers the power to make it because it varies or departs from the provisions of the Act.[6] Accepting the test of repugnancy between an Act and delegated legislation is to be applied in this context, the question of inconsistency turns on the Act's requirements and the terms of the Delegation.

[37] The requirements of s 90 of the Act in relation to a price determination are extensive. They include the matters identified in s 90(5), but they also include the requirements and permissions contained in ss 90(2), 90(3) and 90(6). For example, s 90(3)(d) prohibits the making of a price determination for distribution non-network charges. But there is no mention of that restriction in the Delegation.

[38] As that example shows, in my view, the drafter of the Delegation did not purport to state all of the requirements of s 90 in respect of a delegated price determination. As well, there is other non-exhaustive language in the text. The direction to base the Determination on a Network (N) plus Retail (R) cost build up approach is qualified by words which recognises it may not be possible to do so fully - "to the extent possible" - which permit another unstated approach to be taken if QCA finds it necessary. The direction as to the energy cost component is that it "should include the cost of purchasing energy" and other matters, not that no other matters may be considered.

[39] In my view, there is no basis in the text of the Delegation for the conclusion that the Delegation purports to state exhaustively the matters that QCA should consider.

[40] It is important to identify at the outset that the second basis of argument is not that the Delegation expressly prohibits consideration of the actual costs of producing electricity. It is that because "actual costs of making or producing electricity" are not expressly mentioned, the Delegation impliedly excludes the actual costs of producing electricity from the matters to which QCA must or may have regard in making the price determination.

[41] The only language which could arguably have that effect is the direction that the QCA should ensure it has regard to "the actual costs of supplying electricity". But, in my view, neither as a matter of ordinary language nor in the context of the Delegation under the Act, is that a direction made to exclude QCA from having regard to the cost of producing electricity in making the price determination.

[42] It follows, in my view, that the challenge to the validity of the Delegation fails.

[43] As will appear, on the Relevancy grounds, the debate skirted around what the QCA as decision-maker was bound to take into or not take into account as a matter of construction of the statute, on the one hand, and whether the challenge was mounted in a way that "slid into impermissible merit review",[7] on the other hand.

[44] However that may be, it is necessary to determine a number of arguments about the operation of s 90(5). It is convenient to explore the operation of the section and its meaning, to some extent, before dealing with some of the specific arguments.

[45] It is of some importance that s 90(5) in its present form was introduced as part of specific measures intended to alter the statutory provisions for the method of deciding a price determination of tariffs for customer retail services.

[46] The sub-section in its present form was introduced by s 7 of the Electricity Price Reform Amendment Act 2011 (Qld) ("the amending Act"), commencing on 13 September 2011. The prior form of the legislation is relevant context to the interpretation of s 90(5) of the present Act.

[47] Until the relevant sections of the amending Act commenced, s 90(5) required that the pricing entity must, in deciding the prices under s 90(1), "comply with any indexation required under division 3". Division 3 of part 2 of Ch 4 provided for a methodology which utilised what was called the "benchmark retail cost index" ("BRCI"). BRCI was a mathematical function of "total benchmark retail cost".[8] Total benchmark retail cost was the total of a number of elements, including "the cost of energy, as worked out under s 92".[9] The cost of energy, as worked out under s 92, was required to "reflect the pricing entity's view of the likely total of the costs to be incurred... to purchase energy to supply all of the NEM[10] load of the State...". However, s 92(2) then required that "the view must be based on ...the most recent estimate of the long run marginal cost of energy..." As well, s 92(6) required that "in estimating the long run marginal cost, the pricing entity must comply with any methodology prescribed under a regulation". Regulations 104 to 107 of the Electricity Regulation 2006 (Qld) provided that "the prescribed methodology for estimating the long run marginal cost of energy... is a theoretical framework that complies with this Division...".

[48] All of this was done away with by the amending Act, when s 90(5) was amended to provide that in making a price determination the pricing entity must have regard to, inter alia, the "actual costs of making producing or supplying the goods or services". In particular, the references to a theoretical framework for estimating long run marginal cost and any requirement to have regard to long range marginal cost were deleted from the matters to which the pricing entity must have regard in making a price determination. In this case, and in the Determination and surrounding documents, long range marginal cost was abbreviated to "LRMC".

[49] These amendments did not occur in a vacuum. In June 2009, QCA received a ministerial direction notice under s 10(e) of the Queensland Competition Authority Act 1997 (Qld) directing it to examine the BRCI methodology and alternative price setting methodologies for reflecting the costs of supplying electricity and to examine Queensland's existing retail electricity tariffs and alternative tariff structures. The Review of Electricity Pricing and Tariff Structures was completed in 2009 in two stages ("2009 Review").

[50] By its Final Report on the Review of Electricity Pricing and Tariff Structures - Stage 1 dated September 2009, QCA "consider[ed] that a N+R cost built up approach which will allow full cost reflectivity to be achieved (over time if the difference between notified prices and efficient pricing is significant) should be adopted".[11] It also considered two broad ways to assess the energy cost component of the R component of that approach and identified the "market based approach, sometimes referred to as the energy purchase cost approach, [as] aimed at estimating the wholesale energy costs involved in supplying electricity at prevailing market prices over a given period. This approach considers the best mix of financial instruments a retailer would enter into to meet the required load of electricity in the period".[12] QCA believed that there were sufficient reasons to "move... to a completely market based energy purchase cost approach".

[51] On 11 May 2011, and in response to QCA's recommendations, QCA received a second ministerial direction notice under s 10(e) requiring it to investigate and report on an alternative retail electricity pricing methodology for the determination of cost components under an N (network) + R (energy and retail) approach and an alternative set of retail electricity tariffs, based on an N + R approach, which could be applied from 1 July 2012.

[52] The Minister's 2011 direction notice was a transitional measure to allow the 2012-2013 pricing review process to commence while amendments were being made to the Electricity Act and the Electricity Regulation to remove the BRCI approach.

[53] The explanatory notes to the Bill for the Amending Act referred to the policy rationale for the Bill including:

(a) the 2009 direction notice to QCA and the 2009 Review, including QCA's proposal that an alternative price setting methodology based on a Network (N) + Retail (R) cost build up approach be introduced; and

(b) the Minister's 2011 direction notice to QCA to develop an alternative methodology and an alternative set of retail electricity tariffs.

[54] The explanatory notes as to the electricity pricing provisions of the Bill described the amendments as intended to achieve the policy objectives, by removal of the overly prescriptive nature of the (then) current legislation and all references to the BRCI, and continued:

"In its place, a more enabling legislative framework is proposed to be inserted which specifies broad Government policy objectives and parameters for the determination of notified electricity prices. The amendments will also provide sufficient flexibility to deal with any unidentified policy changes or market upheavals. In particular, this approach will ensure that the regulatory framework is flexible enough to allow any future carbon costs or additional costs associated with environmental obligations to be captured and passed through to end use customers."

[55] On 13 September 2011, on the commencement of the amending Act, including s 90(5) in its amended form, s 329 of the Act was also inserted. It applied to the Minister's 2011 direction notice and provided that "any investigation or report submitted by QCA under the QCA Act s 10(e) is taken to be a valid part of the price determination process under chapter 4, part 2 for the relevant tariff year".

[56] The document submitted by QCA entitled "2011 Issues Paper - Review of Regulated Electricity Tariffs and Prices" dated June 2011 ("2011 Issues Paper") addressed the Minister's 2011 direction notice and constituted an investigation or report within the meaning of s 329.

(a) summarised the 2009 Review process and outcomes;

(b) reiterated the 2009 Review's identification of the possible inclusion in the energy cost component of the cost of purchasing energy through the NEM;

(c) reiterated the 2009 Review's conclusion that a market-based approach offered the best method for assessing the wholesale energy costs likely to be faced by retailers;

(d) stated that a market-based methodology involved establishing the level of energy purchase costs that a representative retailer would incur in supplying the regulated customer load;

(e) stated that establishing the energy purchase costs would involve consideration of the various financial products and hedging strategies that a representative retailer would use to mitigate its potential exposure to high NEM spot market prices;

(f) identified that a significant proportion of the hedging undertaken by retailers is in the form of bilateral agreements with generators for which there is no publicly available data;[13]

(g) sought stakeholders' views on its proposed hedging based model, the appropriate mix of hedging contracts, how (if at all) bilateral agreements should be taken into account and whether there were any other factors QCA should consider in relation to this issue;

(h) discussed the use of LRMC as a floor price mechanism in other jurisdictions and noted the argument that regulated retail prices should not be used to attempt to correct concerns about the long term investment in electricity generation; and

(i) sought stakeholders' views as to whether energy costs should include LMRC as a floor price.

[58] There has already been reference to the NEM. Some explanation of the NEM and its underpinning legislation also serves as context for the questions of construction of s 90(1) and s 90(5) which are to be decided. It is convenient to start the description from another case in which Origin challenged the validity of a price determination made under s 90(1) before the amending Act commenced, from the reasons of Philip McMurdo J:[14]

"[4] In 1996, there was established a so-called national electricity market covering Queensland and most other States and the Australian Capital Territory. In each of these jurisdictions there is uniform legislation providing for this market, the Queensland statute being the Electricity - National Scheme (Queensland) Act 1997 (Qld). It provides for the application in this State of the National Electricity Law which is set out in the schedule to the National Electricity (South Australia) Act 1996 (SA) as in force for the time being.

[5] This national scheme has relevantly two elements. The first is an interconnected power system between the jurisdictions known as the national grid. Under the Act, this term takes its meaning from the so-called National Electricity Rules which are made under the National Electricity Law, and by those Rules the national grid is:

the sum of all connected transmission systems and distribution systems within the participating jurisdictions.

[6] Transmission systems or "networks" are the plant and equipment which transport relatively large quantities of electricity at high voltages from major electricity generators. The term is defined in the National Electricity Rules by reference to certain levels of voltages. A distribution system or network is plant and equipment by which electricity is transported but which is not a transmission network. These are lower voltage networks which are typically used to transport electricity from connection points with transmission networks to consumers such as households and most businesses.

[9] The other presently relevant element is the spot market for electricity. This is conducted by the National Electricity Market Management Company ("NEMMCO"). The sellers in this market are typically electricity generators and the buyers are typically electricity retailers or, in some cases, large consumers. The market is compulsory in that generators across the participating jurisdictions are required to transact through it with NEMMCO as the counterparty. Electricity generators are required to submit offers detailing how much power they are willing to sell and at what price. NEMMCO matches this information with the requirements of buyers with the objective of ensuring that total supply equals total demand and at the lowest feasible cost to the buyer, and by this means spot prices in the market are derived. The market is divided into geographic regions which presently correspond with the respective participating jurisdictions. Wholesale spot prices are determined for each region." (footnotes omitted)

[59] By the time of the events of this case, there were some changes to the system as just described. The National Electricity Law and National Electricity Rules were amended to replace NEMMCO with Australian Energy Market Operator Ltd ("AEMO").

[60] AEMO's functions are prescribed in the National Electricity Law while procedures for market operations, power system security, network connection and access, pricing for Network Services and national transmission planning are all prescribed in the National Rules.

[61] Wholesale trading in electricity is, as stated above, conducted as a spot market where supply and demand are instantaneously matched in real-time through a centrally coordinated dispatch process. Generators offer to supply the market with specific amounts of electricity at particular prices. Offers are submitted every five minutes of every day. From all offers submitted AEMO determines the generators required to produce electricity based on the principle of meeting prevailing demand in the most cost efficient way. AEMO then dispatches these generators into production.

[62] The volatility of this pricing mechanism in the NEM is recognised in the market generally. In a publication entitled "An Introduction to Australia's National Electricity Market" dated July 2011 AEMO acknowledged the risk of volatility and the market's response thus:

"Participants in the NEM require a means of managing the financial risks associated with the significant degree of volatility that occurs during trading periods. They typically achieve this by using financial contracts that lock in a firm price for electricity that will be produced or consumed at a given time in the future. These contracts serve to substantially reduce the financial exposure of market participants and contribute to spot market stability. They are known as derivatives, and include swaps or hedges, options and futures contracts."

[63] Provisions which require that a price be fixed by reference to "costs" have been the source of no small amount of trouble in the law. For example, in Vardon v The Commonwealth[15] it was held that "cost" in a provision that fixed the price of a tailor of suits at "cost of the goods or services plus 20% thereof", "cost" not being defined, was an "ambiguous and uncertain term". But in Bendixen v Coleman Scott and Croft[16] it was held that a provision which fixed the price of a liquor retailer as the "cost of... non-listed liquors... plus 25% of that cost" was not uncertain as the cost could be ascertained "by what he has paid or is liable to pay for the liquor to the person from whom he buys it".

[64] More recently, in a contractual context, it has been said that a price fixed by reference to costs is not uncertain. In Upper Hunter County District Council v Australian Chilling and Freezing Co Ltd[17] the contract provided that the price of the supply of bulk electricity was to be varied "if the supplier's costs shall vary". It was held that the clause was not void for uncertainty, Barwick CJ saying:

"I do not think there is any uncertainty or for that matter ambiguity in the expression "supplier's costs" in cl 5, however wide may be the area of possible disagreement as to its denotation in a particular case. A contract to build a bridge at cost could not, in my opinion, be held void for uncertainty; it could not properly, in my opinion, be said to be meaningless; nor is it, in my opinion, ambiguous. Endless might be the arguments pro and con as to whether or not in marginal cases some item of expenditure is as claimed a cost, or as to how much of an expenditure is a cost, of the particular activity. But to my mind, generally speaking, the concept of a cost of doing something is certain in the sense that it provides a criterion by reference to which the rights of the parties may ultimately and logically be worked out, if not by the parties then by the courts. There are no elements in the circumstances of this contract to deprive the concept of that certainty."

[65] Similarly, in Trawl Industries of Australia Pty Ltd v Effem Foods Pty Ltd,[18] Clarke JA said:

"So, in the instant case, there might be dispute whether the cost of some activity or some item of plant, or some category of wages, expended in ways and for purposes apparently not directly related to the performance by TIA of the contract, might nevertheless be a cost of catching, processing and delivering headed and gutted jack mackerel (together with the anatomical desiderata). But controversy of that sort in working out the financial implications of the pricing formula is not an argument in favour of declaring it to be uncertain, incomplete or ambiguous."

[66] None of these cases is directly relevant to the meaning of "costs" in s 90(5)(a)(i). There is no question in this case of uncertainty - an Act of Parliament is not legally uncertain. Nevertheless, these cases illustrate two points: "costs" is a word that has no fixed connotation in the context of a price fixing formula and there are often debates to be had about the extent of its denotation in such a context; secondly, the meaning is to be gleaned by a close consideration of the surrounding text and context.

[67] In s 90(5)(a)(i), the word "costs" is qualified by the adjective "actual" and the target costs are those of "making, producing or supplying the goods or services" which are the subject of the price determination. The provision does not create a price fixing formula. The determination is one to be made by the pricing entity. The pricing entity "must have regard to" a number of matters. The relevant costs are one matter. The "effect of the price determination on competition in the Queensland retail electricity market" is another. Any matter QCA is "required by delegation to consider" is a third. And "any other matter" QCA "considers relevant" is the last.

[68] In a context like s 90(5)(a)(i), "actual" used as an adjective may connote something which is "existing in fact [or] real" as opposed to something which is "potential [or] possible".[19]

[69] A question as to the meaning of "actual" in a relevant context arose in R v Lyon,[20] where a section required unfinished goods to be valued for customs duty by deducting from their finished value the "actual cost of labour or material used or expended in Australia". Griffith CJ said:[21]

"... it seems to me to admit of two meanings, one being 'such cost as is ascertained by actual disbursements already made,' in antithesis to 'estimated' or 'probable' cost, and the other the 'real direct cost' as measured by necessary disbursement for the sole purpose of completing the manufacture or putting up the article under a proprietary or trade name, in antithesis to notional or constructive cost..."

[70] Here, the pricing entity must have regard to the actual costs of the identified activities in making a determination of the prices that are to apply prospectively for a period of up to a year. At the time when s 90(5)(a)(i) was introduced, the next tariff year was to commence 9 months thence. By using the present participial form as a gerund - "of making, producing or supplying" - the sub-paragraph directs the reader, as a matter of ordinary grammar, to the continuing actions of those kinds. That is, both as a matter of ordinary grammar, and in the context of s 90(5) and s 90(1), the costs in question are those of the tariff year to come which are to be the subject of the price determination.

[71] The assessment of "actual costs" is one which must occur before most of those costs will be incurred. Necessarily, the context is one where necessary disbursements for the making, producing or supply of the goods or services will not be incurred until after the price determination has been made. Operating costs, for example, including the costs of acquiring electricity through the NEM for the sale of electricity to premises, will not be incurred until after the price determination under s 90(1) is made and the tariff year commences. Further, as previously mentioned, "actual costs" are but one of the factors to which the pricing entity must have regard under s 90(5).

[72] As also previously mentioned, the methodology of making the price determination by applying the BRCI index to the preceding tariff year's tariffs was replaced by the new s 90(5). BRCI had been based on a cost of energy, which was required to reflect a view of costs to purchase energy, which view was itself required to be based on an estimate of LRMC, estimated by complying with a "theoretical framework".

[73] In this overall context, while costs in the form of actual disbursements already made or actual liabilities already incurred are plainly capable of constituting actual costs of the identified activities under s 90(5)(a)(i), there seems no reason to disregard necessary disbursements for the purpose of making producing or supplying the relevant goods or services. The use of "actual" does not direct the pricing entity away from the costs which will be incurred in "making, producing or supplying" the relevant goods or services over the future period of or within the relevant tariff year. Instead, it seems to me, it is directing the pricing entity away from a theoretical framework of estimating costs which are not and will not be the costs of making, producing or supplying the relevant goods or services.

[74] So, in context, "actual costs" are at least in part a prospective estimate of the anticipated real costs.

[75] LRMC has already been mentioned. It is convenient to deal with it a little more extensively at this point, having regard to the discussion above about actual costs.

[76] The repealed price determination provisions employed LRMC in the methodology for determining the cost of energy as part of the calculation of the adjustment of the BRCI index which was to be applied to the last year's tariff rates to determine an upcoming year's tariff rates.

[77] However, it was not argued by the applicants that LRMC constitutes actual costs of supplying customer retail services. It was common ground that LRMC is a theoretical measure. And it was common ground that it was a measure of the costs of generation of electricity, not the costs of retail supply of the sale of electricity as such.

[78] The only ways in which the applicants sought to positively deploy LRMC were to suggest that it was an available measure of the costs of generation of electricity which, as subsequently discussed, they contended were costs of producing or supplying electricity, and that LRMC can be used as a proxy measure of the costs of PPAs. However, the applicants did not grapple with the difficulty that LRMC was a theoretical measure, not an actual measure, whether of generation costs or the costs of PPAs.

[79] It is a commonplace for a statute to direct that a discretionary power conferred upon a decision-maker is to be exercised in a way that has regard to identified matters. A direction that a decision-maker "must have regard to" an identified matter will ordinarily identify a condition of the valid exercise of the power.

[80] In one sense, when the matter is identified by the decision-maker, in making the decision, the decision-maker has had regard to the matter. But what is the consequence where the matter is not an element of or does not substantially affect the result arrived at?

[81] This question was explored by the parties in their written submissions as a matter of generality. However, in my view, it does not assist much in the resolution of what is required by a "must have regard to" direction under a particular statutory provision to analyse the meaning of that expression divorced from the particular context.

[82] The applicants contended that when "must have regard to" is directed to "substantive" matters it requires both that the decision-maker must "give them weight as a fundamental element in making the decision" and that the decision-maker must "give proper, genuine and realistic consideration to those matters". As well, they contended that: "the requirement to have regard to something is not satisfied by the decision-maker merely adverting to or considering the matter, but then rejecting it or jettisoning it as irrelevant to the decision making process".

[83] The language of "weight as a fundamental element" stems for the reasons of Mason J in R v Hunt; ex parte Sean Investments Pty Ltd,[22] where the section in question directed that a decision-maker "shall, in determining the scale of fees in relation to a nursing home... have regard to costs necessarily incurred in providing nursing home care in the nursing home". Mason J said:

"When sub-s (7) directs the Permanent Head to 'have regard to' the costs, it requires him to take those costs into account and to give weight to them as a fundamental element in making his determination. There are two reasons for saying that the costs are a fundamental element in the making of the determination. First, they are the only matter explicitly mentioned as a matter to be taken into account. Secondly, the scheme of the provisions is that, once the premises of the proprietor are approved as a nursing home, he is bound by the conditions of approval not to exceed the scale of fees fixed by the Permanent Head in relation to the nursing home. In many cases it is to be expected that the scale of fees will be fixed by ascertaining the costs necessarily incurred and adding to them a profit factor. In the very nature of things, the costs necessarily incurred by the proprietor in providing nursing home care in the nursing home are a fundamental matter for consideration."

[84] It will be observed that in Sean Investments the relevant costs were those for a particular nursing home, not nursing homes, in general. And that they were the costs "necessarily" incurred for that nursing home, which thus included the rent payable, not an adjusted reasonable rental.

[85] The expression "weight as a fundamental element" has only been used once again in the High Court of Australia.[23] Three observations may be made about using it as a criterion of application of "must have regard to", either in general, or in this case.

[86] First, many provisions direct decision-makers and courts to have regard to specified matters in all sorts of different statutory contexts - but it is not usually said that in respect of all such matters "weight as a fundamental factor" must be given to them. The applicants sought to confine their suggested application of the requirement to "substantive" matters, but that distinction draws no general support from the cases, in my view, although Finn J referred to it in Tobacco Institute of Australia v National Health and Medical Research Council.[24] A secondary process of inquiring whether the matter is "substantive" is, in my view, no substitute for interpreting the requirements of and meaning of the particular provision in question, without a starting gloss of the kind submitted.

[87] Secondly, in the context of judicial review of an administrative power, it is trite that "in the absence of any statutory indication of the weight to be given to various considerations, it is generally for the decision-maker and not the court to determine the appropriate weight to be given to the matters which are required to be taken into account": Minister for Aboriginal Affairs v Peko-Wallsend Ltd.[25] This approach is informed by the fundamental principle reaffirmed in Minister for Immigration and Multicultural Affairs v Yusuf:[26]

"The considerations that are, or are not, relevant to the Tribunal's task are to be identified primarily, perhaps even entirely, by reference to the Act rather than the particular facts of the case that the Tribunal is called on to consider. In that regard it is important to recall, as Brennan J said in Attorney-General (NSW) v Quin:

'The duty and the jurisdiction of the courts are expressed in the memorable words of Marshall CJ in Marbury v Madison: 'It is, emphatically, the province and duty of the judicial department to say what the law is.' The duty and jurisdiction of the court to review administrative action do not go beyond the declaration and enforcing of the law which determines the limits and governs the exercise of the repository's power. If, in so doing, the court avoids administrative injustice or error, so be it; but the court has no jurisdiction simply to cure administrative injustice or error. The merits of administrative action, to the extent that they can be distinguished from legality, are for the repository of the relevant power and, subject to political control, for the repository alone.'" (footnotes omitted)

[88] Thirdly, other cases arising in cognate contexts show that "weight as a fundamental factor" is not a criterion of application of "must have regard to", in general. Thus, in Rathborne v Abel[27] the High Court of Australia construed a provision for the determination of fair rent for prescribed premises, which required that the decision-maker "shall have regard to" ten matters listed in lettered paragraphs One paragraph referred to the capital value of the premises at the prescribed date. Another referred to the justice and merits of the case and the circumstances and conduct of the parties. Barwick CJ considered the requirement to have regard to relevant matters thus:[28]

"Whilst, of course, it may not be universally true that a direction 'to have regard to' certain facts or circumstances does no more than require the tribunal to which the direction is given to consider whether it should give any and, if so, what weight to the particular circumstance when performing the duty or exercising the right which is given to it, it can, I think be said that in general a direction in such terms does not do more than that."

"Finally, to require that regard be had to a particular matter in making a discretionary judgement is not to require that that matter shall be allowed an actual influence upon the ultimate result."

[90] Notwithstanding the number of occasions on which the "weight as a fundamental factor" language has been mentioned as applicable in subsequent cases, in my opinion, there is no general principle that a matter required to be taken into account must be given "weight as a fundamental factor", if that proposition is intended to convey more than that the matter must be regarded. Beyond that, the question of what is required is a matter which depends on the particular provision in its context. Any other general approach trends judicial review into the merits of administrative action, in my view, and is better avoided. The question is what does the section, properly construed, require? Once that question is answered, there is no further "weight as a fundamental factor" principle to be applied.

[91] In the particular context of this case, to identify s 90(5)(a)(i) as carrying some particular weight as a fundamental factor would raise other questions: what is the weight as a factor to be given to the effect of the price determination on competition? Or the matters QCA is required by delegation to consider? Or the matters that the pricing entity considers relevant? Are they to be gauged one against another? If so, how? In my view, these are not questions which, objectively speaking, it is likely that the legislature intended would have to be answered, as a matter of law, in order for the pricing entity to be able to validly determine prices under s 90(1) of the Act.

[92] In like vein, the language of "proper, genuine and realistic consideration" stems from reasons of Gummow J, as a Judge of the Federal Court of Australia, in Kahn v Minister for Immigation and Ethnic Affairs.[30] That case concerned the exercise of power in accordance with a rule or policy without regards to the merits of the particular case. It was in that context that it was found that in considering all relevant material, the decision-maker was required to "give proper, genuine and realistic consideration to the merits of the case". Two observations may be made about using it as a criterion of application of "must have regard to", in general, or in this case.

[93] First, Gummow J was not purporting to lay down a new principle of administrative law in Khan. He sourced the requirement of proper, genuine and realistic consideration in earlier judgments, including Kioa v West,[31] and decided the case by reference to s 5(2)(f) of the Administrative Decisions (Judicial Review) Act 1977 (Cth), namely that the power was exercised in accordance with a rule or policy and without regards to the merit of the case. It is in later cases that it has been suggested that the proper, genuine and realistic consideration requirement is a test of what is involved in taking something into consideration.[32]

[94] Secondly, in the context of acting in accordance with a rule or policy, in Neat Domestic Trading Pty Limited v AWB Pty Limited,[33]Gleeson CJ said:[34]

"There is nothing inherently wrong in an administrative decision-maker pursuing a policy, provided the policy is consistent with the statute under which the relevant power is conferred, and provided also that the policy is not, either in its nature or in its application, such as to preclude the decision-maker from taking into account relevant considerations, or such as to involve the decision-maker in taking into account irrelevant considerations. The policy, and its application, must be measured against those requirements, having regard to the matter presented for decision, and the information and arguments, if any, advanced for or against a particular outcome."

[95] Thirdly, in Minister for Immigration and Citizenship v SZJSS,[35] in a joint judgment, the High Court of Australia said:[36]

"In Swift v SAS Trustee Corporation, Basten JA (with whom Allsop P agreed) noted Khan's case and said of the language of "proper, genuine and realistic consideration" (at [45]):

[45] That which had to be properly considered was "the merits of the case". Taken out of context and without understanding their original provenance, these epithets are apt to encourage a slide into impermissible merit review." (footnotes omitted)

[96] In the present case, the applicants seek to deploy the "proper genuine and realistic consideration" language. But they did not contend that QCA pursued a rule or policy which evidenced no genuine consideration being given to the actual costs of making producing or supplying customer retail services. Instead they sought to identify subject matters which they submitted were not addressed, or not addressed in the way that they submitted the statute required, and to contend that those matters were matters required to be taken into account, and vice versa, as making out the Relevancy grounds.

[97] In dealing with the Relevancy grounds, "the level of particularity with which a matter is identified for the purpose of applying this principle may be significant".[37] But, in my view, the answer to the question whether a particular matter must or must not be taken into account is not advanced by invoking the proposition that "proper, genuine and realistic consideration" must be given to the subject matter articulated at a higher level of generality.

[98] Lastly, the contention that "the requirement to have regard to something is not satisfied by the decision-maker merely adverting to or considering the matter, but then rejecting it or jettisoning it as irrelevant to the decision making process" stems from the plurality reasons for judgment in East Australian Pipeline Pty Limited v Australian Competition and Consumer Commission.[38]

[99] In that case, the section prescribed a method of establishing an amount for the initial capital base value of a pipeline. The initial capital base was required to be established by considering factors identified in paragraphs (a)-(k) of the section, including three specific valuation methodologies in paragraphs (a)-(c). The decision-maker chose not to consider one of the required specific valuation methodologies at all, and instead rejected it, for reasons which it gave. Gleeson CJ, Heydon and Crennan JJ said:[39]

"There is nothing in the overall structure of the section which indicates that factors listed in paras (e)-(j) or indeed para (k) would allow the person considering all of the s 8.10 factors to jettison or ignore the factors covered by paras (a)-(d) or to give them cursory consideration only in order to put them to one side. Those first four factors are fundamental to the practical exercise which is being undertaken."

[100] Again, in my view, this passage is not accurately characterised as some general principle about the way in which a decision-maker must or must not treat every factor to which regard must be had, as a proposition of law. On the contrary, it was a statement about how the particular provision under consideration operated and a conclusion that, construing the section, the decision-maker was not permitted to substitute its own preferred valuation methodology and process for one which the statute specifically required had to be taken into account.

Actual costs of making producing or supplying customer retail services

[101] The three grounds of review encompassed by the Relevancy grounds focussed on the text of s 90(5)(a)(i) that the pricing entity must have regard to "the actual costs of making, producing or supplying the goods or services".

[102] One step in the applicants' oral argument was to identify the relevant goods or services in s 90(5)(a)(i) as being goods, namely electricity. This was a contentious point which it may not be necessary to resolve. However, some of the associated arguments are relevant. A price determination made under s 90(1) is a decision as to the prices (or the methodology for fixing the prices) for "customer retail services" or "charges or fees relating to customer services" or "other goods and services prescribed under a regulation".[40]

[103] As defined, "customer retail services" are "for premises ...the sale of electricity to the premises". So, the relevant prices to be decided by QCA in the Determination were those for the "sale of electricity" to relevant premises. The relevant costs, under s 90(5)(a)(i), were the costs of making, producing or supplying the sale of electricity as services, unless electricity itself is treated as "the goods" within the meaning of s 90(5)(a)(i).

[104] In context, there are several reasons which tend against the conclusion that electricity constitutes goods. First, in s 90(1), and, therefore, in s 90(5)(a)(i), the sale of electricity to the premises is services, namely "customer retail services", so there is no immediate contextual reason to identify the cost of producing electricity as a cost in relation to electricity as goods. Secondly, by way of contrast, the cost of making or producing goods has work to do in s 90(5) where there are "other goods" prescribed under a regulation, for which a price must be decided under s 90(1). Thirdly, "goods" is a word used only in ss 55, 90(5) and 91(5) of the Act and does not in any of those sections necessarily or most likely mean electricity. Fourthly, the definition of "electricity" in s 5 of the Act does not identify electricity as goods. Fifthly, at common law, and in the context of contracts for the sale of goods, electricity was not clearly identified as being "goods".[41] Hence, where a statute intends that "goods" should include electricity it is appropriate to do so by express inclusion, as in s 4 of the Competition and Consumer Act 2010 (Cth) and s 2 of the Australian Consumer Law. In my view, there is no reason in the context of s 90(5)(a)(i) to construe "goods" as meaning the electricity supplied as the subject of the sale of electricity to premises which constitutes customer retail services.

Retail entity must acquire electricity through the NEM

[105] There is another point to be made here about electricity as goods or services which are made or produced by an "electricity retailer". At times, the applicants' submissions strayed into language which assumed that a retail entity might supply electricity which it had produced (and not acquired from another person) as a sale of electricity to premises constituting customer retail services.

[106] As previously discussed, under the NEM and the Act, a generation entity which is the holder of a generation authority may connect generating plant to the transmission grid or a supply network as stated in its authority and may sell electricity through the spot market in accordance with the National Electricity Rules. The generation entity sells electricity through the NEM to AEMO as the counterparty, not to a retail entity or customers who receive customer retail services as defined under the Act. A distribution entity under a distribution authority is authorised to supply electricity using a supply network. And a customer who owns or occupies premises is connected to a supply network. However, a distribution entity is prohibited from holding a retail authority and was so prohibited before the amending Act was enacted.[42] A retail entity is authorised by its retail authority to supply customer retail services, comprising the sale of electricity to premises.

[107] For practical purposes, all electricity to be sold to a customer connected to a supply network who receives customer retail services is acquired by the relevant retail entity by purchase through the NEM. The retail entity does not sell electricity produced by it as a generator directly to such a customer. Whether or not the retail entity might itself (or via a related corporation) be (or have an interest in) a generation entity which generates electricity does not affect the fact that it acquires the electricity which it provides as customer retail services through the NEM by purchase. This industry structure has operated from a time prior to the introduction of s 90(5) in its current form on 13 September 2011. It separates the activities of a generation entity from the activities of a retail entity, under the Act, whether or not a retail entity may "invest" in generation assets.

Costs of producing electricity are distinguished from costs of supplying electricity under s 90(5)(a)(i)

[108] Another step in the applicants' argument was that the relevant actual costs under s 90(5)(a)(i) include the costs of producing electricity, meaning the generation of electricity. In this contention they distinguished costs of producing electricity from costs of getting the electricity produced to the consumer. They characterised the costs of purchase of electricity as a cost of supplying whatever is being supplied "where this, rather than the cost of making or producing it is the relevant cost of obtaining it". In this way, they seek to construe s 90(5)(a)(i) as distinguishing between the costs of generation of electricity, which are costs of "making or producing" and the costs of acquisition of electricity by purchase on the other hand, as costs of "supplying" the sale of electricity to premises.

[109] This submission, it seems to me, was also based on the assumption that a retail entity, under the structure provided for in the Act, might provide the sale of electricity that it produces, that is generates, by way of customer retail services. As discussed above, it does not seem to me that the Act is structured in that way. Accordingly, in my view it would not be appropriate to construe s 90(5)(a)(i) on the assumption that it does.

[110] Returning to the text of s 90(5)(a)(i) in its context in s 90(5) and s 90(1), it must be kept steadily in mind that the costs to which regard must be had are those of the customer retail services. It is only a retail entity which supplies those services. And it is only for the purpose of fixing the price of those services for regulated tariffs that the costs are being examined. All that supports the conclusion that it is the costs of a retail entity in providing the customer retail services which are the relevant costs, where s 90(5)(a)(i) is engaged in circumstances like those in this case.

[111] Further, because the tariff which is to be fixed is for all retail entities in that class of tariff, the costs which are referred to in the sub-paragraph are not specific to an individual retailer.

Costs of producing electricity are a necessary precursor to supplying

[112] Thirdly, the applicants alternatively submitted that the costs of supplying the sale of electricity to premises include the costs of producing electricity, meaning generating electricity, for a vertically integrated retail entity. In oral argument, the applicants went a step further, and submitted that the costs of generating electricity are costs under s 90(5)(a)(i) irrespective of whether they are costs of a vertically integrated retail entity providing customer retail services. The argument draws on the proposition that the costs of producing electricity are a necessary precursor to the supply of electricity and, therefore, s 90(5)(a)(i) is to be seen as specifically directed to the costs of producing electricity.

[113] Again, the difficulty with this argument is that the costs to which regard must be had are being assessed for the purpose of a price determination of customer retail services, that is, the sale of electricity to premises, which electricity has been acquired by the retail entity through the NEM. If it be accepted that the cost of the acquisition of the electricity which is provided as customer retail services is an actual cost of supplying the sale of electricity, an obvious question becomes what is the purpose of requiring that regard must be had to the costs of producing electricity as well as the more direct costs of acquisition of electricity?

[114] In my view, it must be accepted that the cost of the acquisition of electricity to be provided as customer retail services through the NEM is an actual cost of supplying those services. All retail entities incur those costs, whether or not they are involved in generation activities. For the moment, it may be assumed that generation costs can be described as costs of producing electricity under the Act. The applicants' contention is that regard must be had to both the costs of producing electricity which will become the subject of the sale of electricity to premises and the costs of supplying the sale of electricity by a retail entity acquiring electricity through the NEM. That contention re-casts "making, producing or supplying" in s 90(5)(a)(i) as though it provided "making, producing and supplying".

[115] Both QCA and the Minister submitted that there was no warrant to read s 90(5)(a)(i) as though it provided "making, producing and supplying". They submitted that "making" or "producing" were apt words to apply to "goods" whereas "supplying" applied to "services". The applicants responded by submitting that whether the subject was goods or services, a sale constituted "supplying", which supported reading the sub-paragraph as extending to the cost of "underlying" goods or services. The applicants submitted that QCA and the Minister's arguments should not be accepted because "in order to undertake the sale of electricity, the retailer must first have electricity to sell. It achieved this by either producing the electricity itself, or by purchasing it. Thus the cost of selling electricity includes the cost associated with either generating or purchasing electricity".

[116] I reject the last part of the applicants' submissions, for the reasons previously outlined. They seek to construe s 90(5)(a)(i) by reference to a false assumption, namely that a retail entity which provides customer retail services has not acquired the electricity the subject of the sale through the NEM by purchasing it. Whether or not that entity has itself or by a related corporation simultaneously engaged in generation, the costs associated with purchasing electricity as a retail entity will have been incurred.

[117] The question is returned to whether there is any reason in the text or context of s 90(5)(a)(i) to construe it as being directed to the costs of generation of electricity, as a separate subject mater from the costs of supplying the sale of electricity.

[118] As appears from the preceding discussion, the applicants' arguments identify "producing" goods or services comprising customer retail services with generating electricity. However, the Act does not usually refer to generating electricity as producing electricity. The only place where it clearly does so is s 55DB which refers to small customers described as "qualifying generators" with sufficient capacity to "produce" a total of 8 megawatts or more of solar power. Logically, if s 90(5)(a)(i) had been intended to refer to the costs of generation of electricity, it would have been easy enough to say so.[43]

[119] In order to assess whether there is any identifiable purpose of the Act which would be achieved by the interpretation contended for by the applicants, I sought to identify with the applicants' counsel what was required to comply with a requirement that regard be had to the costs of generation of electricity. They submitted that the pricing entity was required under s 90(5)(a)(i) to make a general inquiry as to the costs of generation. They submitted that in order to fulfil that requirement the pricing entity had a power to obtain required information under s 90A of the Act. It was further submitted that the methodology of LRMC could be used.

[120] In different ways, the applicants also submitted that to have regard to costs without taking into account the costs of generation could lead to a result where the price determination fixes tariffs that are lower than the actual costs of the production of electricity. In my view, it is not necessary for me to consider whether or not that is so, or what the cause of that might be, unless that consideration is discernible in the language of, the context of, or the purposes of s 90(5)(a)(i). It does not seem to me that it is necessary, or that it would be correct, because of that possible outcome, to erect a requirement that there be a general inquiry into the costs of production of electricity under s 90(5)(a)(i) on any occasion that a price determination is made, as a matter of construction of the sub-paragraph.

[121] As to s 90A, I note that a pricing entity's power to ask for "relevant information" is confined to asking a "retail entity". Section 90A does not support the suggestion that there is power to obtain information from generating entities, in general, as to costs of producing electricity. If anything, s 90A tends against the applicants' argument for an obligation to conduct a general inquiry as to the costs of generating electricity. It seems unlikely that the Act would impose such an obligation by such indirect language as "must have regard to...the actual costs of making, producing... the goods or services" and without a power to obtain information as to the costs. It is appropriate to note that the pricing entity would have access to the NEM information as to prices obtained by generating entities when selling electricity through the NEM. But that is the same information that is available as the cost of acquisition of electricity by retail entities.

All costs actually incurred by a retail entity are not actual costs

[122] Fourthly, the applicants submitted that actual costs under s 90(5)(a)(i) included all the costs actually incurred by a retail entity. Such actual costs were said to include a retail entity's costs of any power purchase agreements ("PPAs") and the costs of generating electricity where a retail entity was vertically integrated as a generation entity.

[123] It seems to me that neither the Act, in general, nor s 90(5)(a)(i), in particular, is directed to a retail entity which is vertically integrated. For the reasons already given, in my view, there is no general direction in s 90(5)(a)(i) to have regard to the cost of generating electricity. The answer to that question, in my view, should not be different because there are some integrated retail entities. The question is one of the meaning of the sub-paragraph of the section.

[124] In the preceding sections of these reasons, I have considered specific arguments that the applicants made that QCA was required to have regard to the costs of producing electricity, meaning generation of electricity, under s 90(5)(a)(i).

[125] The conclusions I have reached thus far may be summarised. In my view, QCA was not bound to take into account the costs of generation of electricity as the actual costs of making, producing or supplying customer retail services, because the costs of generation of electricity, per se, are not those actual costs.

[126] QCA took hedging costs into account in its assessment of costs for the relevant tariff year and in its price determination under the Determination. The applicants submitted that the Determination "acknowledges that PPAs and investing in generation assets are two steps that retailers can take to reduce their exposure to volatile NEM spot prices" and that these "are measures taken by retailers in supplying electricity. Therefore the QCA did not have proper or genuine regard to those matters". They also contended that "...QCA...failed to take into account relevant considerations, by rejecting PPAs and LRMC as components of the cost of supplying electricity".

[127] QCA responded that it in fact considered both the matters of PPAs and investing in generating assets. The applicants responded that the consideration was not to the extent required by law, and invoked the language of "fundamental weight"[44] or "proper, genuine or realistic consideration", and in particular contended that QCA had impermissibly "discarded" these matters as identified aspects of actual costs and went on to submit that QCA "set the relevant prices by reference to hypothetical costs".

[128] The Minister took a similar approach to QCA, but with a difference. He contended that the actual costs of supplying retail customer services did not include hedging costs and, therefore, QCA were not required to have regard to those costs under s 90(5)(a)(i).

[129] It is convenient to deal with the last point first. It must be said that it seems counter-intuitive in this context that hedging costs would not be treated as costs of the retail supply by way of sale of electricity to premises, having regard to the factual context in which s 90(5)(a)(i) in its current form was introduced into the Act.

[130] When the mischief which s 90(5)(a)(i) was intended to remedy is considered,[45] the section was amended to introduce a more "market reflective" approach to the determination of regulated prices covered by the tariffs as they then existed. The relevant market was the one introduced under the arrangements of the NEM and the Act, including the acquisition of electricity through the NEM by a retail entity for the purpose of providing customer retail services.

[131] As appears from the description of the NEM set out previously, at the time when s 90(5)(a)(i) was introduced it was a notorious feature of the market that spot prices were volatile and that electricity generators and electricity retailers, inter alia, managed the risk of market volatility by financial instruments of various kinds, collectively described as hedging. In this case, the applicants seek to include managing the risk of volatility by carrying on business as both generator and retailer in the same market as a "natural hedge", but otherwise hedging is carried out by financial contracts or instruments of one kind or another.

[132] In each of the 2009 Review and the 2011 Issues Paper, QCA propounded the N + R methodology, where R included a component for energy costs and QCA treated the costs of hedging as costs of the sale of electricity by a retailer. This was the recommended methodology to determine energy costs as part of the build up of what would become the regulated prices the subject of the tariffs.

[133] The purpose of s 90(5)(a)(i) is to require the pricing entity to have regard to relevant costs in deciding the prices or methodology for fixing the prices for the regulated tariffs. It is the text of the sub-paragraph in context which identifies the scope of the costs to which regard must be had for that purpose. Even if hedging costs were not within the s 90(5)(a)(i) costs, it does not mean that it would be impermissible for the pricing entity to have regard to them. It may do so as a matter it considers relevant. Or it may be required to do so in having regard to the effect of the price determination on competition or, in QCA's case, as a matter it is required by the delegation to consider.

[134] But in the context of the NEM, and the division of the functions of a generating entity and a retail entity under the Act, it seems to me to be more logical to treat the costs incurred as a necessity of doing business as a retail entity as costs of making, producing or supplying the customer retail services provided by such an entity to non-market customers, recognising that non-market customers are a segment of the customers of a retail entity's business. Given the malleable meaning of "costs" in differing contexts, as previously discussed, in my view it would best achieve the Act's purpose in s 90(5)(a)(i)[46] to include necessary hedging costs in the costs to which the pricing entity must have regard.

[135] The Minister's contrary argument focussed on the contention that the costs of hedging are "not costs of the sale of electricity" and that because the costs of hedging are not "expenses incurred by the vendor in achieving receivability of the consideration in respect of the sale" they are not within the scope of the sub-paragraph. But on that footing many other direct and indirect costs of a retail entity in providing customer retail services would not be included. It is not helpful to speculate by examples, but in my view there is no reason to read s 90(5)(a)(i) as constrained to the costs of the sale. The retail obligation of a retail entity to provide customer retail services is not necessarily as confined as that. And it is pursuant to that obligation that a retail entity may be required to provide customer retail services to numerous customers.

[136] For those reasons, in my view, s 90(5)(a)(i) is to be construed so that the actual costs of supplying customer retail services include the necessary costs of hedging.

[137] It is next convenient to deal with the applicants' arguments that s 90(5)(a)(i) required QCA to have regard to PPAs in a way that wasn't done. It assists to give a factual context to their existence and potential relevance.

[138] As previously stated, in its 2011 Issues Paper, QCA sought stakeholders' views on its proposed hedging based model, the appropriate mix of hedging contracts, how (if at all) bilateral agreements should be taken into account and whether there were any other factors QCA should consider in relation to this issue.

[139] Origin responded by a submission dated 5 August 2011, stating inter alia that:

"Bilateral contracts between retailers and generators are usually long term contracts. Bilateral contracts are confidential arrangements that effectively underwrite future energy security. Long Term Power Purchase Agreements (PPAs) are necessary to finance large upfront capital investments such as new power stations and are best taken into account by the LRMC as the floor in the pricing determination process".

[140] That statement suggests that PPAs are an underpinning of a generation entity's business. The question in this case, however, is whether the pricing entity must have regard to PPAs as actual costs of a retail entity under s 90(5)(a)(i).

[141] After the first iteration of the Delegation was made by the then Minister on 22 September 2011, QCA published a document entitled "Draft Methodology Paper - Regulated Retail Electricity" dated November 2011. In response to that paper Origin made a submission dated December 2011 which again took up the subject of PPAs, in the following fashion: "Origin believes that LRMC provides several advantages including... it is a forward looking approach that better approximates the actual costs of retailers' purchases through power purchase agreements".

[142] In March 2012, QCA published a document entitled "Draft Determination Regulated Electricity Prices 2012-2013" ("Draft Determiation"). In response to the Draft Determination, Origin made three submissions which referred to PPAs:

(b) a supplementary submission dated 11 May 2012; and

(c) a confidential supplementary submission dated 11 May 2012.

[143] The supplementary submissions were thus made barely four weeks before the due date for the final determination to be made, in a final process which had been begun more than a year before.[47]

[144] The April 2012 submission encouraged QCA to reconsider the inclusion of LRMC in the method for estimating energy costs, contending that: "Large retailers are required to hedge outside of the contract market... and have accomplished this through generation and power purchasing agreements (PPAs)" and advocating that "if LRMC is ignored then there is potential for the QCA to apply a suitable risk premium or appropriate escalation of contract market prices."

[145] In the supplementary submission, Origin submitted that: "Retailers with substantial retail loads seek to manage (ie hedge) this liquidity and price risk through... in the case of mass market retail load - longer term PPAs and physical generation assets. These sources better match the longer term duration of the retail mass market and provide retailers that have significant retail demand with price and volume certainty". Origin referred to "using energy from its own generation assets - such as Darling Downs and Mt Stuart and long term PPAs like Braemar 1 and Braemar 2". And it submitted: "The QCA should expand its market-based definition so that it takes into account retailers' actual costs, which include the costs of longer term PPAs and physical generation assets"

[146] In the confidential supplementary submission, Origin summarised the base contract and cap prices for "Origin's Queensland physical generation assets and PPAs" and submitted that "[t]hese prices are appropriate for inclusion in the QCA's market-based hedging model". The information for Darling Downs and Mt Stuart did not identify the entity that was the owner or operator of the assets but identified the fixed and variable costs and "price" on a $/MWh basis, incorporating a percentage required rate of return in the total. It described the "cap" cover of the Mt Stuart and Braemar PPAs as "effectively an insurance product against high spot prices".

[147] The applicants' written submissions as to PPAs, in this proceeding, referred to most of the matters identified above. As well, they identified all other references to long term agreements of that kind in submissions made by other stakeholders in response to the Draft Determination. It would be unhelpful to dilate on them further. Four additional brief points may be made about those submissions: first, it was stated that the details of PPAs were highly confidential; secondly, no other submission identified any amounts said to identify actual costs of PPAs; thirdly, nearly all proponents who referred to PPAs urged the adoption of LRMC as a proxy measure of the cost of PPAs; and fourthly, the only other methodology proposed to assess the impact of PPAs was an approach which involved a theoretical calculation.[48]

[148] In the Determination, QCA stated: "...there are a range of measures that a retailer can take in order to reduce its exposure to volatile prices in the spot market including... entering longer term power purchase agreements with generators or investing in generation assets...". It noted the responses to the Draft Determination included the contention that: "...the low level of trading on the SFE is evidence the market participants are using longer term contracts such as power purchasing agreements (PPA) to source a considerable proportion of their electricity". QCA continued: "While there is no transparent data source for PPA transactions, retailers contend that PPAs are based on LRMC calculations."

[149] QCA stated its position on LRMC in the Determination, concluding that "LRMC is an estimate of the long term generation costs rather than the costs to a retailer of purchasing wholesale electricity" and that it "ignores the existence of the NEM and the major impact it has had on the wholesale price of electricity". In considering submissions that LRMC be included as a "floor" in regulated retail prices QCA noted that it might "provide additional security for investment in generation" but questioned whether that was necessary and continued: "Furthermore, while it is possible that a regulated price based on energy purchase costs could be less that the actual costs faced by retailers (including the costs of PPAs) it would be expected that if this were the expected long term situation, there would be little or no discounts to the regulated price available in the market place and competition would not be vigorous. But this is not the case in Queensland now and the new prices are in excess of current levels."

[150] As to its hedging-based approach, QCA discussed a number of matters which informed its decision in the Determination to calculate wholesale energy costs for all 2012-2013 tariffs based on the hedging-based approach and concluded: "A more detailed discussion of the hedging approach can be found in ACIL's Final Report which is available from the Authority's website".

(a) in section 2.2, stated that "the inclusion of generation production costs in the calculation of WEPC [wholesale energy purchase costs] potentially imposes the consequences of inefficient investment decisions on end-users despite the NEM being designed to pass on the benefits of the competitive market to those end-users...";

(b) in section 3 stated that "[i]n general retailers favoured incorporating an LRMC approach in the calculation of the EPC [energy purchase costs] to reflect the cost of purchasing energy. ACIL Tasman has considered these proposals and has concluded that even if a sizeable portion of a retailers energy purchases are through Power Purchase Agreements (PPA) and reliable pricing information was available (which it is not) that would not be appropriate to incorporate these PPAs in the calculation of the EPC.

PPAs are long term instruments usually running across several years or over the expected life of a generating asset with the PPA price designed to provide stable long term return to the asset owner. In this sense the PPA price would generally be expected to reflect costs to a retailer no higher than purchasing through a combination of the electricity pool and electricity hedges over the term of the PPA. Therefore, the market price over the term of the PPA would be expected to provide a ceiling to a well priced PPA.

This does not mean that a well priced PPA would be expected to be lower than the market price every year but would be expected to be lower in some years and on average no higher than the market price over the term of the PPA. This of course assumes that PPA prices are efficient. Separately in section 2.2 we considered the issue of inefficient generation investments. PPA prices linked to inefficient generation investments may always be higher than the expected market price over the life of the PPA, but there is no merit in attempting to pass these inefficiencies through to end users.

The 2012/2013 year in Queensland is in ACIL Tasman's opinion, supported by modelling, characterised by an oversupply of generation. This implies that it might be expected that the expected market price will be lower than even efficiently priced PPAs."

[152] The subject matter of present concern is not whether these views on the part of QCA, including the hedging approach it adopted as discussed in its consultant's report, are correct. That is a matter for the administrative decision-maker. This proceeding does not challenge the Determination because it lacked rationality or was so devoid of merit as to attract a challenge based on unreasonableness.

[153] The question is whether QCA illegally did not have regard to PPAs as actual costs of making, producing or supplying retail customer services. The applicants submitted in chief that "the costs associated with PPAs were also entirely excluded" by QCA.

[154] QCA submitted that "it cannot be said the [QCA] did not have regard to the fact that some retailers entered into PPAs or their significance. The Minister submitted that QCA "had regard to PPAs but thought another method produced a more accurate estimate".

[155] The applicants submitted in reply that the matters set out above showed that QCA "did not incorporate either PPAs or the costs of generation (whether by LRMC or otherwise) into its assessment of actual costs, instead preferring to discard them". This was in addition to the contention that: "It is not for the applicants to say how the QCA should have had regard to the different practices that retailers adopt to obtain electricity and hedge their exposure...". As well, it was submitted that "QCA had the power to require retailers to provide information to it (including about PPAs) for the purposes of making its price determination".

[156] I confess that the applicants' approach is not immediately attractive. So far as the evidence showed, the only information as to actual PPAs which was provided to QCA was in the confidential supplementary submission made by Origin on 11 May 2011. The stratagem adopted by those who made submissions about PPAs, in general, was to urge the adoption of LRMC as a proxy costing and pricing measure, even though it was not a calculation or estimation of the actual costs of actual PPAs.

[157] Stripped down, the applicants' contention must be that because PPAs are part of the costs of business of a significant proportion of the retail entities which supply customer retail services, QCA was obliged to pursue an investigation into the costs of actual PPAs.

[158] It is at this point that one of the questions identified previously, namely that "the level of particularity with which a matter is identified for the purpose of applying this principle may be significant" is engaged. In identifying relevant costs for the purpose of estimating costs in arriving at its price determination, there is no dispute that QCA identified hedging costs of different kinds. Nor is there a question that it identified that some retailers incurred costs via PPAs. In seeking to identify relevant costs, it rejected the proposition that it should go to LRMC. That was no error of relevancy, in my view. But if it proposed to do that, was QCA obliged to inquire further into actual PPA costs?

[159] In my view, there are a number of matters which assist in coming to an answer to that question. First, as mentioned above, a PPA price is designed to provide stable long term return to the asset owner. The focus in that statement is on the generator's side.

[160] Secondly, the generic description PPA, as an abbreviation for "power purchase agreement", masks the nature of such a contract. In the example of a PPA given to QCA by Origin in the confidential supplementary submission, the contract is not for the purchase of power, meaning electricity, at all, which is no surprise given that all supplies by a generation entity by way of sale are to be made through the NEM. The contract is not even confined to the subject of the sale of electricity. However, due to its confidential nature, I will not identify the other subject matter. It is essentially a contract in which exchanges of promises of payments are made, depending on numerous factors, and by reference to the NEM pool prices, for amounts of electricity to be supplied into the NEM by the generator. The "retailer" is to pay a weekly amount by reference to a nominated volume or volumes and the generator is to pay amounts to the retailer where the pool price exceeds the nominated strike price or prices. None of the provisions of the contract is directed to the costs of any particular part of the retailer's business of supply of sale of electricity to customers.

[161] There are many forms of financial instrument by which a retail entity may manage the risk of volatility of the cost of electricity acquired through the NEM. I have concluded that hedging costs are so fundamental to the operating costs of a retail entity that they comprise actual costs of making, producing or supplying customer retail services. However, it does not follow from the requirement to have regard to the category of hedging costs that QCA is bound to ascertain or value the amounts of the costs of all the different financial instruments entered into by all retail entities. The requirement is to have regard to the actual costs which will be incurred for the purpose of fixing the tariffs under the price determination, which is a single price in the relevant category for all retail entities. The different actual costs of doing business by each of the retail entities, incurred in ways that vary individually, are not the actual costs of making, producing or supplying customer retail services of a retail entity for the purposes of s 90(5)(a)(i). QCA is not bound to ascertain and to take all of those individual variations of costs into account under s 90(5)(a)(i) in order to make a valid price determination.

[162] Thirdly, in my view, there is a difference between a hedging strategy for the costs of electricity acquired by a retail entity over a relevant tariff year for the supply by way of sale of electricity to premises as customer retail services, on the one hand, and the risk management of a PPA of a kind which extends over many years and which is not specifically directed to the relevant part of the retail entity's business, on the other hand. The former can be seen to represent actual costs of the retail entity in providing customer retail services for the relevant year. The latter, in my view, has a less direct connection to the costs of customer retail services for the year. That does not mean that QCA is not permitted to take PPAs into account. But that is a different thing from being required to ascertain their amounts in order to have regard to the actual costs of making producing or supplying customer retail services under s 90(5)(a)(i).

[163] In my view, QCA was not required, by s 90(5)(a)(1), to conduct an inquiry into the actual costs of PPAs by compulsorily obtaining information as to those costs from all retail entities under s 90A of the Act. Once that point is reached, it seems to me that the applicants' arguments based on the regard which was had to PPAs fall away. As well, and at the risk of repetition, I do not consider that s 90(5)(a)(i) required QCA to adopt LRMC as a measure of PPAs as a cost or to take LRMC into account as actual costs of making, producing or supplying retail customer services..

[164] The conclusion I have reached is that QCA was not bound to take into account the costs of PPAs as a relevant consideration in making the price determination because the costs of PPAs are not, per se, the actual costs of making, producing or supplying customer retail services.

[165] In a number of the particularised grounds of review the applicants contend that QCA took into account what it considered to be the costs of an efficient representative retailer, or the efficient costs of supplying electricity, or the costs of an independent retailer and that, in so doing, the QCA took into account an irrelevant consideration to the exercise of the power to decide the price determination under s 90(1).

[166] It will be necessary to identify precisely in what respects the applicants contended that these matters were illegally taken into account. However, the articulation of these particularised grounds involved a necessary intermediate step, namely that the costs of an efficient representative retailer, or the efficient costs of supplying electricity, or the costs of an independent , were treated by QCA as elements of or considerations relevant to the actual costs of supplying electricity within the meaning of s 90(5)(a)(i).

[167] In my view, care must be taken with that step. A determination of prices must be made having regard to other matters, not just the actual costs of making, producing or supplying the sale of electricity to premises. The matters identified in sub-paragraphs (a)(ii) and (iii) of s 90(5) are other things to which QCA must have regard. But, as well, QCA is permitted to have regard to any other matter it considers relevant under s 90(5)(b). It scarcely needs to be repeated that the first of the objects of the Act is to set a framework that promotes "efficient" electricity supply and that the other objects include "to ensure that the interests of customers are protected" and "to take into account National competition policy requirements".

[168] The question becomes where does the Act require that the costs of an efficient representative retailer, or the efficient costs of supplying electricity, or the costs of an independent retailer must be excluded from the matters to which QCA may have regard in deciding the price determination? At its root, the argument must be that those matters are impliedly excluded from the matters to which regard may be had by the express requirement that QCA must have regard to the actual costs of making producing or supplying the sale of electricity to premises.

[169] Such an argument invokes the canon of statutory interpretation once labelled by the latin maxim expressio unius est exclusio alterius, which was characterised by the High Court in Houssein v Seretary of Industrial Relations and Technology (NSW)[49] thus:

"That maxim must always be applied with care, for it is not of universal application and applies only when the intention it expresses is discoverable upon the face of the instrument: Saunders v. Evans [1861] EngR 335; (1861) 8 HLC 721, at p 729 [1861] EngR 335; (11 ER 611, at p 615) . It is 'a valuable servant, but a dangerous master': Colquhoun v. Brooks (1888) 21 QBD 52, at p 65."

[170] In this context, it is more than a bridge too far to suggest that at no point in the price determination process may QCA have regard to the efficient costs of supplying electricity or the costs of an independent retailer. As well, in considering what are actual costs under s 90(5)(a)(i), I have previously concluded that necessary disbursements for the purpose of making, producing or supplying the relevant goods or services are included, in the sense that the sub-paragraph is looking forward to the costs which will be incurred for the period of the relevant tariff year, not merely costs which have been incurred in the past.

[171] The applicants written submissions in chief tied this point in respect of the Relevancy grounds to an inconsistency between "the decision-maker disregarding the actual costs and having regard, instead, to other hypothetical costs" and continued: "[Section 90(5)(a)(i)] excluded considerations such as the costs of a notional 'independent retailer' or a hypothetical 'efficient representative retailer'". However, the specific submissions focussed on the ACIL Tasman Final Report, rather than the Determination itself, on the footing that QCA adopted the ACIL Tasman Final Report.

[172] In their written submissions in reply, the applicants made reference to the Determination as follows:

(a) "the QCA specifically stated that an allocation for headroom should be provided above the 'efficient' costs of supply', indicating that the QCA determined the price by reference to its assessment of efficient costs"; and

(b) "The QCA estimated the costs that it considered would be incurred in the ensuing year by a retailer with a particular profile (but not one found in the real market-place)".

[173] The applicants' written submissions in reply continued, in some detail, to criticise QCA's estimate of costs and included a number of factual assertions that were neither established before the decision-maker nor in this proceeding as factual matters for a challenge based on the Relevancy grounds.[50] They culminated in the contention that: "setting prices at an unrealistically low level does not foster competition". The court should decline to engage in this sort of analysis on judicial review based on the Relevancy grounds, and it should not be taken as accepting or rejecting the factual assertions by not dealing with them seriatim.

[174] The relevant question is whether QCA was precluded from having regard to the particular matters in arriving at the Determination, either in having regard to the actual costs of making, producing or supplying customer retail services or otherwise. There is an aspect of this argument that seems like trying to put a finger on mercury. It is best, therefore, to proceed by reference to the Delegation and the Determination, in the first place.

[175] The Delegation made an express reference to an "efficient representative retailer" in relation to the Retail costs to be incurred by a retail entity. In determining the Retail cost component of each regulated retail tariff, the Minister directed QCA under s 90(5)(a)(iii) that QCA "must consider the retail costs that would be reasonably incurred by an efficient representative retailer". This was something that QCA was obliged to consider, unless it was inconsistent with some statutory requirement. Although the applicants contended that the Delegation was invalid, they did not submit that it was invalid because of that direction.

[176] In response to that direction, in section 4.2 of the Determination, QCA identified a "representative retailer" as "an incumbent, stand alone Queensland electricity retailer with a substantial and representative cross-section of customer types", referred to the need to "reflect the actual level of costs incurred by the representative retailer" and continued: "...estimating the actual level of costs will have a direct impact on the resulting tariffs." This discussion occurred in the context of Retail costs, which are not costs observable on an external market like the NEM. Also, they will be costs that vary across the range of retail entities, according to their particular business characteristics. It was no error, in my opinion, to normalise the assessment of the actual costs of a retail entity in that way. The concept of a "representative retailer" and QCA's employment of that concept in assessing Retail costs did not constitute taking into account an irrelevant consideration. The applicants did not contend that it was, but instead treated the discussion of the concept in the context of Retail costs as though it had occurred in the context of the discussion of energy costs, which was not accurate.

[177] As previously mentioned, the 2011 Issues Paper did employ the "representative retailer" concept in the context of discussing energy costs. QCA submitted, however, that the lack of reference to the "representative retailer" concept in part 3 of the Determination, which dealt with energy costs, is not surprising, because the costs information deployed in the discussion of those costs was derived from prices paid by all participants in the market which did not differ among the retail entities. That submission does not, however, recognise that the hedging strategy adopted requires assumptions to be made about a retail entity's approach, as will appear.

[178] The relevant section of the Determination is section 3.2 headed "Wholesale Energy Costs". Section 3.2.1 comprises a discussion of the approach taken, including whether LRMC or a "market based approach which estimates the costs a retailer would incur in purchasing electricity at prevailing market prices over a given period" should be adopted. The references to "estimates" and "costs a retailer would incur", in context, refer back to the "estimate of wholesale energy costs" by arriving at "forecast customer load profiles which must be supplied in 2012-2013", the "hedging strategy to be used in settling the forecast loads against the forecast prices", the "forecast energy spot prices to apply in 2012-2103" and "the costs of carbon". In other words, the estimation was of the forward looking costs which would be actually incurred, not a hypothetical methodology for arriving at those estimates.

[179] In estimating customer load profiles which must be supplied in 2012-2013, in section 3.2.2, QCA had regard to sources of actual data referred to in the discussion, discussed the methodology to be employed for making the estimates and decided to determine "energy costs based on load profiles for AEMO 'settlement classes' in order to reflect the actual costs incurred by retailers" before considering a number of different parts of the data and possible adjustments to be made.

[180] In estimating the costs associated with the hedging strategy, in section 3.2.5, QCA referred to the Draft Determination, in which it had adopted a hedging strategy by making an assumption as to the contracts a retailer would purchase, using a four year time period which "would better reflect the actual hedging strategies of retailers". After discussing a number of specific matters, including the data sources of actual trading being used to estimate the assumed 2012-2013 hedging strategy, QCA stated, as previously mentioned, that "[f]urther detail on the modelling specific issues raised by stakeholders is provided in ACIL's Final Report" and that it was "satisfied that ACIL's hedging strategy provides for a reasonable level of protection for retailers, without being overly risk averse".

[181] In estimating the forecast energy spot prices to apply in 2012-2013, in section 3.2.6, QCA discussed the use of historical spot price data as against forecast spot prices data, noting that in response to the 2011 Issues Paper a number of stakeholders, including Origin, favoured the use of forecast spot prices, because historical prices are unable to take account of future market conditions such as the introduction of a carbon tax. The spot price forecasts were derived from a model that takes the 41 half hourly load profiles for the NEM together with 10 generator outage scenarios to produce 410 half-hourly spot price scenarios for 2012-2013. These scenarios reflect a range of potential load and spot price outcomes that may occur in 2012-2013, "to reflect the range and relative likelihood of costs being incurred by a retailer and uses the median price of the 410 prices as the energy purchase cost".

[182] Table 3.1 collected the wholesale energy cost allowances built up by the methodology just described for relevant tariff classes, excluding losses and carbon, which were subsequently dealt with.

[183] The most "representative retailer" like part of the methodology employed in section 3.2 of the Determination, as summarised above, was the adoption of the assumptions made for the hedging-based strategy of a retail entity. However, it is one thing to have regard to actual costs of making producing or supplying the sale of electricity to premises under retail tariffs. It is quite another to be required to adopt a methodology which would require modelling of separate hedging strategies for all the different retail entities in the market. At the end of the day, the applicants made no positive submission as to what QCA was required to do, beyond the points previously dealt with. That gap in their case is not filled by seeking to characterise QCA's decision in making the Determination as one made by impermissibly approaching the energy costs to which it had regard through a prism of the "representative retailer".

[184] It has been necessary to set out a brief description of the process, in order to return to the applicants contention that the estimate was "entirely hypothetical and theoretical" as based on the "representative retailer" concept. I reject that characterisation of the methodology. But more importantly, for present purposes, I reject the contention that QCA was precluded from employing the concept of "representative retailer" in the ways in which it did so in making the Determination and that to the extent that the methodology had regard to the costs of a representative retailer that matter was an irrelevant consideration to take into account under s 90(1) of the Act.

[185] I also reject the contention that QCA, by adopting that methodology, did not have regard to actual costs within the meaning of s 90(5)(a)(i). As appears above, at a number of points QCA specifically directed itself to actual costs and to that which was cost reflective. Necessarily, the process of estimation of 2012-2103 cost allocations, to some extent, involved using existing information and data and adjusting it for the year in prospect. Also, necessarily, the process required some normalisation of the individual retail entities positions to arrive at a single overall tariff of each relevant class. The applicants did not submit that process could or should be eschewed altogether.

[186] In the result, none of the grounds of the application have been made out. The application for a statutory order of review must be dismissed.

[2] I acknowledge the use of that expression to describe these grounds in Aronson, Dyer & Groves, Judicial Review of Administrative Action, 4 ed, Ch 5

[3] Grounds 1 - 3 of the grounds of review

[4] [1989] FCA 159; (1989) 23 FCR 536 at 539-540. See also McCormack v Deputy Commissioner of Taxation [2001] FCA 1700; (2001) 114 FCR 574 at [37]- [39]; Australian Retailers Association and ors v Reserve Bank of Australia [2005] FCA 1707; (2005) 148 FCR 446 at [442]- [473]; Chandra v Webber and anor [2010] FCA 705; (2010) 270 ALR 393 at [40]- [45]

[6] Plaintiff M47-2012 v Director General of Security [2012] HCA 46; (2012) 292 ALR 243 at [54], [174] and [434]

[7] Minister for Immigration and Citizenship v SZJSS & Ors [2010] HCA 48; (2010) 243 CLR 164 at [30].

[10] NEM is an acronym for the national electricity market

[13] Agreements of this kind were described as "power purchase agreements" or "PPAs" in the arguments and evidence in this case

[14] AGL Energy Ltd & anor v Queensland Competition Authority [2009] QSC 90 at [4] ff

[15] [1943] HCA 30; (1943) 67 CLR 434 at 445

[16] [1943] HCA 40; (1943) 68 CLR 401 at 417

[17] [1968] HCA 8; (1967) 118 CLR 429 at 436

[18] (1992) 27 NSWLR 326 at 344

[19] Oxford English Dictionary, Online Version, December 2012

[20] [1906] ArgusLawRp 35; (1906) 3 CLR 770

[22] [1979] HCA 32; (1979) 180 CLR 322 at 329

[23] R v Toohey; parte Meneling Station Pty Ltd [1982] HCA 69; (1982) 158 CLR 327 at 333

[24] [1996] FCA 1150; (1996) 71 FCR 265 at 277

[25] [1986] HCA 40; (1985-1986) 162 CLR 24 at 41

[26] [2001] HCA 30; (2001) 206 CLR 323 at [73]

[27] (1964) 38 ALJR 293

[30] [1987] FCA 457; (1987) 14 ALD 291

[31] [1985] HCA 81; (1985) 159 CLR 550 at 604

[32] For example, Weal v Bathurst City Council [2000] NSWCA 88 at [9]

[33] [2003] HCA 35; (2003) 216 CLR 277

[35] [2010] HCA 48; (2010) 243 CLR 164

[37] Foster v Minister for Customs and Justice [2000] HCA 38; (2000) 200 CLR 442 at [23]

[38] [2007] HCA 44; (2007) 233 CLR 229

[40] At the points in time of the Delegation and the Determination, there were no relevant prescribed other goods or services

[41] Benjamin's Sale of Goods¸7 ed, at [1-085]

[42] Section 124 of the Electricity Regulation 2006.

[43] Section 90(5)(a)(i) is also the only use of "making" in association with "goods" or "services" in the Act.

[44] Straying at times from "weight as a fundamental factor"

[45] Alcan (NT) Alumina Pty Ltd v Commissioner of Territory Revenue [2009] HCA 41; (2009) 239 CLR 27 at [47]

[46] Section 14A of the Acts Interpretation Act 1954 (Qld)

[47] Having regard to the 2011 Minister's direction notice and s 329 of the Act.

[49] [1982] HCA 2; (1982) 148 CLR 88 at 94

[50] See paragraphs 33, 40 and 41 of the Applicants' submissions in reply