Robusto Investments Pty Ltd v Essential Services Commission of South Australia - [2025] SASC 87

SUPREME COURT OF SOUTH AUSTRALIA
(Appeal to a Single Judge)
ROBUSTO INVESTMENTS PTY LTD v ESSENTIAL SERVICES COMMISSION OF SOUTH AUSTRALIA
[2025] SASC 87
Judgment of the Honourable Chief Justice Kourakis
30 May 2025
ADMINISTRATIVE LAW - ADMINISTRATIVE TRIBUNALS - SOUTH AUSTRALIAN CIVIL AND ADMINISTRATIVE TRIBUNAL - APPEAL, REVIEW OR REHEARING
ENERGY AND RESOURCES - WATER - WATER MANAGEMENT - WATER SUPPLY - SOUTH AUSTRALIA
ENERGY AND RESOURCES - WATER - WATER MANAGEMENT - WATER USAGE RIGHTS - WATER ALLOCATION - WATER SUPPLY SCHEMES
STATUTES - ACTS OF PARLIAMENT - INTERPRETATION - INTERPRETATION ACTS AND PROVISIONS - STATUTORY DEFINITION PROVISIONS GENERALLY
This is an appeal by the applicant, Robusto Investments Pty Ltd (‘Robusto’), against an order of the South Australian Civil and Administrative Tribunal (‘the Tribunal’), dated 17 November 2023, setting aside a decision of the respondent, the Essential Services Commission of South Australia (‘ESCOSA’), made on 26 August 2021 (the ‘Price Determination’).
Robusto is a supplier of reticulated water to residents of Mount Compass and had agreements in place that tied water rates to the rates charged by SA Water. The rates were secured by way of an encumbrance on the customer’s land. Those agreements were superseded by an ESCOSA approved ‘standard water customer contract’, gazetted in September 2018, but the linkage to the rates charged by SA Water was not changed.
On 25 May 2021, the initial price determination issued by ESCOSA on 25 May 2021 (the ‘initial determination’) capped the prices at which Robusto could supply drinking water to its customers in Mount Compass, pursuant to a licence to provide portable water retail services. Robusto applied for an internal review of the initial determination on 23 June 2021. ESCOSA, on 26 August 2021, made the price determination.
The price determination capped the charges Robusto could levy for the services it provided. The pricing controls were calculated to allow Robusto to recover its reasonable operating costs, together with reasonable amounts by way of a return on its assets and for the depreciation of those assets. The annual depreciation was calculated to allow the recovery of the value of Robusto’s assets over their remaining useful lifetime. Robusto lodged an application to the Tribunal challenging the price determination, pursuant to s 34 of the South Australian Civil and Administrative Tribunal Act 2013 (SA) (the ‘SACAT Act’).
The Tribunal referred the decision back to ESCOSA for reconsideration and, pursuant to s 37(1)(c)(ii) of the SACAT Act, the Tribunal directed the ESCOSA to reconsider the allowance it had made in its price determination for operating costs.
The appeal raises a question of law of general importance as to the scope and nature of the directions which can be given pursuant to s 37(1) of the SACAT Act in the context of an appeal against a decision of ESCOSA by which the Tribunal is precluded by s 32(6) of the ESC Act from varying or substituting ESCOSA’s decision with its preferred determination.
The appeal also raises for consideration principles of general importance as to:
1.the identification of the appropriate date on which to value the regulated asset base of a licensed water provider for the purposes of making an allowance for a return of, and on, its capital investment;

2. the valuation of the regulated asset base; and
3.the assessment of the efficient operating costs of a business licensed to provide services in a regulated industry.

Held, granting permission to appeal and allowing the appeal:
1.Section s 37(1) of the SACAT Act does not empower the Tribunal to remit only some aspects of the price determination. If the Tribunal was satisfied the price determination was neither correct nor preferable by reason of an error in respect of one or more of its components, the Tribunal was bound to set it aside and send the matter of the fixing of a price determination to ESCOSA for reconsideration.
2.The Tribunal was correct to consider each line item of the price determination, even though a decision on whether the price determination should be confirmed or set aside could only be made after netting out the ‘unders and overs’.
3.The objective in making a price determination is to ensure the licence holder operates as efficiently as reasonably practicable in the circumstances pertaining to its business and industry and that the benefits of those efficiencies are passed on to the consumer after allowing the licence holder a reasonable return which assures its long-term financial sustainability as a provider of reliable, good quality services, but no more, so that monopoly rents are not recovered. The assessment of a regulated entity’s efficient operating costs, pursuant to the Essential Services Commission Act 2002 (SA), must have regard to the business environment in which the regulated entity operates, including its water resources, supply infrastructure, and the size, nature and geographical spread of its customer base. This assessment will generally require an analysis of the regulated entity’s actual costs for any inefficient, overstated or exceptional expenditure and taking into account the absence of a close comparator against which an assessment can be made.
4.The approach of ESCOSA to the determination of the appropriate return on capital was inconsistent with the National Water Initiative Pricing Principles. ESCOSA must assess the efficient financing costs of an entity operating the water retailer business which is calculated to ensure the sustainable delivery of reliable good quality water services to its customers.
5.The regulated asset base used to calculate an allowance for a return of, and on, capital includes all infrastructure actually in use at the commencement of the regulatory period. The remaining functional life of such infrastructure is a question of fact and may differ from its generally excepted useful life. In this context, the Tribunal erred in confirming ESCOSA’s determination that assets which had reached the end of their expected useful life but remained in use should be valued at nil. Instead, those assets should be assigned a value as at the commencement of the first regulatory period, with historical losses disregarded.
6.The return on capital should be calculated to allow the licence holder to provide sustainable, reliable good quality services without allowing the extraction of monopoly rents. The assessment of the appropriate WACC must be based on a stand-alone benchmark efficient entity and not the actual financial arrangements entered into by the water retailer, but must be made by reference to an entity operating the licence.
7.The terms ‘correct and preferable’ in s 34(4) of the SACAT Act must have a single meaning in their application to every Tribunal jurisdiction. The contention that the standard on a review from a decision of ESCOSA is whether the determination is economically reasonable is not with the ordinary meaning of the words ‘correct’ or the word ‘preferable’ or the combined meaning of those words. To impose the standard for which ESCOSA contends is beyond the judicial power of statutory construction. It can only be effected legislatively. A price determination may be one of a number of economically reasonable decisions but be neither the correct nor the preferable decisions. Nor can it be accepted that all the considerations to ESCOSA must have regard can properly be described as economic ones.

Essential Services Commission Act 2002 (SA); Public Sector (Honesty and Accountability) Act 1995 (SA); Public Sector Management Act 1995 (SA) s 64(4); South Australian Civil and Administrative Tribunal Act 2013 (SA) ss 31, 34, 34(4), 37, 37(1), 37(1)(b), 37(1)(c), 37(c)(i), 37(1)(c)(ii), 71; Water Industry Act 2012 (SA), referred to.
House v The King (1936) 55 CLR 499; Minister for Immigration and Citizenship v Dhanoa [2009] FCAFC 153; Vardon v Promotions & Grievance Appeals Tribunal (No 2) [2007] SASC 137; Warren v Coombes (1979) 142 CLR 531, considered.

ROBUSTO INVESTMENTS PTY LTD v ESSENTIAL SERVICES COMMISSION OF SOUTH AUSTRALIA
[2025] SASC 87
Civil: Appeal

KOURAKIS CJ: This is an appeal by the applicant, Robusto Investments Pty Ltd (‘Robusto’), against an order of the South Australian Civil and Administrative Tribunal (‘the Tribunal’), dated 17 November 2023, setting aside a decision of the respondent, the Essential Services Commission of South Australia (‘ESCOSA’), made on 26 August 2021 (the ‘Price Determination’). Robusto is a supplier of reticulated water to residents of Mount Compass (the ‘Mount Compass Water Business’). The Price Determination capped the charges Robusto could levy for the services it provided. The Tribunal referred the decision back to ESCOSA for reconsideration, pursuant to s 37(1)(c)(ii) of the South Australian Civil and Administrative Tribunal Act 2013 (SA) (the ‘SACAT Act’).
The Tribunal directed ESCOSA to reconsider the Price Determination having regard to:
(a)the Tribunal’s reasons for decision published on 27 October 2023;
(b)the actual operating costs incurred by Robusto during the relevant period; and
(c)any further relevant information submitted by Robusto or otherwise available to ESCOSA.
The Price Determination was made pursuant to Essential Services Commission Act 2002 (SA) (the ‘ESC Act’). A uniform national approach to the supply of, and charges for, water was agreed by the 1994 COAG Water Reform Framework and the 2004 Intergovernmental Agreement on a National Water Initiative. The National Water Initiative (‘NWI’) and the National Water Initiative Pricing Principles (‘NWIPP’) were agreed under that intergovernmental framework. ESCOSA must apply the NWIPP in making pricing determinations for water services under the ESC Act.
The Price Determination was made after an internal review of the initial price determination issued by ESCOSA on 25 May 2021 (the ‘Initial Determination’). The Initial Determination capped the prices at which Robusto could supply drinking water to its customers in Mount Compass pursuant to a licence, granted by ESCOSA in August 2016, to provide portable water retail services.
The Initial Determination imposed the following price controls on Robusto for the regulatory period 1 April 2021 to 30 June 2022 (the first regulatory period):
(1)Total revenue is limited to $248,395.00 (in nominal terms);
(2)Maximum nominal charges on its residential customers is limited to:
(a)    a quarterly supply charge of $92.26
(b)    Tier one charge of $3.36 per kl
(c)    Tier two charge of $4.81 per kl
(d)    Tier three charge of $5.20 per kl
(3)Adoption of the formula for fixing prices set out in a Price Schedule to the Initial Determination.
(4)Compliance with specified NWI pricing principles and any applicable industry, rule or guideline set by ESCOSA.
The Price Determination was calculated to allow Robusto to recover its reasonable operating costs, together with reasonable amounts by way of a return on its assets and for the depreciation of those assets. The assessment of the appropriate return on its assets and for regulatory depreciation was made by valuing Robusto’s regulated assets (no additional capital expenditure was claimed by Robusto) before applying a proper return on that capital amount for a business operating in a regulated industry. The annual depreciation was calculated to allow the recovery of the value of Robusto’s assets over their remaining useful lifetime.
ESCOSA reviewed Robusto’s claimed operating costs of $294,033 against the reasonable operating costs of an efficient provider of the service. To that end, ESCOSA considered “comparator entities”. ESCOSA allowed $179,537 for operating costs based on a four-year average. I set out in the table below the downward adjustments on Robusto’s claim made by ESCOSA in the Initial Determination:

Robusto Initial Determination

Bank Fees $1,033 $500

Accounting Fees $5,163 $4,100

Financial Service Fees $1,033 NIL

Water Testing $12,391 $5,300

Legal Fees $10,326 $5,850

Staff $61,956 $30,000

PR $24,782 $5,000

Debt recovery $20,652 $12,000

Bad debt $18,582 $5,500

UV System $2,065 NIL

The downward adjustments were made on the grounds that the expenses were one off or non-recurring costs, were the product of inefficiencies or included costs more properly attributed to another of Robusto’s businesses.
ESCOSA also reduced Robusto’s claim for return of capital of $110,375 to $31,000 and Robusto’s claim for a return on capital of $168,750 to $53,647. ESCOSA also declined to make any allowance for Robusto’s past losses.
Robusto sought a review of the Initial Determination on the grounds that it did not make proper allowance for Robusto’s operating costs and did not provide a sufficient margin for a return of, and on, its capital investment.
Robusto also contended that the pricing control should allow it to recover losses it had sustained prior to the commencement of the first regulatory period.
The internal review affirmed the Initial Determination save that it made a further downward adjustment to the bad debt allowance, fixing it at $625.00.
Robusto applied to the Tribunal for a review of the Price Determination.
The Tribunal examined each line of operating costs allowed by ESCOSA. It found that ECOSA’s annual emergency maintenance allowance of $20,652 was erroneously generous to Robusto. However, after reviewing each line item of Robusto’s operating costs, the Tribunal concluded that inadequate allowances for other items had resulted in an inadequate total allowance for Robusto’s operating costs:[1]
[127] As explained above, we are not satisfied that the allowances arrived at by ESCOSA for accounting fees, legal fees, staff costs, public relations and communications or debt recovery, mediation and dispute resolution were appropriate on the evidence before us, which of course is different from the evidence available to ESCOSA. In particular, we have been able to have regard to Robusto’s actual costs, which we consider relevant to a number of these items. We have also had the benefit of Mr Harris’ opinions as well as those of Mr Houston.
[128] We have also concluded that the adjustments required for these items are likely to exceed the amount by which ESCOSA’s allowance for maintenance exceeded the amount actually spent by Robusto. As we do not consider it practicable or appropriate for ESCOSA to reconsider discrete items in isolation, we have therefore decided to remit the whole question of the appropriate allowance for Robusto’s operating costs to ESCOSA for further consideration having regard to our observations in these Reasons.
[1] Robusto Investments Pty Ltd v Essential Services Commission of South Australia [2023] SACAT 2021/SA002866 (Member Bean and Assessor Rungie) (‘Reasons’).
It will be observed that the last sentence of [128] purports to ‘remit the whole question of the appropriate allowance for Robusto’s operating costs to ESCOSA’. As we shall see the Tribunal was not empowered to order a limited remittal of the Price Determination such that only some, but not all, of its elements were reconsidered.
The Tribunal confirmed the downward adjustments made by ESCOSA in respect of Robusto’s claims for a return of, and on, capital.
Robusto contended before the Tribunal that because of ESCOSA’s delay in making the Initial Price Determination it had lost earnings which it quantified as the difference between the income in fact generated by its historical contracts and the additional income to which it became entitled pursuant to the Initial Price Determination. It sought an annual amount of $121,321 as the amortisation of its total losses incurred in the years preceding the first regulatory period. The Tribunal confirmed ESCOSA’s position that no allowance should be made on Robusto’s historical losses claim.
Robusto does not complain on this appeal that the Price Determination ought not have been set aside. Robusto’s complaint is that different directions should have been given to ESCOSA on how to approach its reconsideration. Robusto contends that the Tribunal should have given the following additional directions to ESCOSA:
1.   ESCOSA is to have regard to the following matters:
(a).    that the Robusto’s initial asset base should be assessed as at the date that it was granted a license namely 10 August 2016 on the basis that from that date 31 March 2021 the Appellant was subject to a Price Determination namely the Economic Regulation of Minor and Intermediate Retailers of Water and Sewerage Services – June 2013 and its successors.
(b).    In the alternative to (a), that the Robusto’s initial asset base should be assessed as at a date after 10 Augst 2016 but prior to 31 March 2021 in light of the correspondence between the parties up to and including 17 March 2017.
(c).    In the alternative to (a) and (b), that Robusto’s initial asset base should be assessed as at a date after 10 August 2016 but prior to 31 March 2021 in light of the date when the value of the asset base was agreed by the parties namely 18 December 2018.
(d).    The appropriate WACC for Robusto during the period from 10 August 2016 to 30 June 2022 in light of all the evidence before the Tribunal.
(e).    The income received by the Robusto and its prudent and efficient operating expenses of the Appellant in the period from 10 August 2016 to 30 June 2022.
(f).    Whether, Robusto recovered any of its capital on the basis of the income received by the Appellant and its prudent and efficient operating expenses in the period from 10 August 2016 to 30 June 2022.
Robusto requires permission to bring its appeal. I would grant permission. The appeal raises a question of law of general importance as to the scope and nature of the power to remit and directions which may accompany that remittal pursuant to s 37(1) of the SACAT Act in the context of an appeal against a decision of ESCOSA in which, as we shall see, the Tribunal is precluded by s 32(6) of the ESC Act from varying or substituting ESCOSA’s decision with its preferred determination.
The appeal also raises for consideration principles of general importance as to:
·the identification of the appropriate date on which to value the regulated asset base of a licensed water provider for the purposes of making an allowance for a return of, and on, its capital investment;
·the valuation of the regulated asset base;
·the assessment of the efficient operating costs of a business licensed to provide services in a regulated industry.
On my construction of s 37(1) of the SACAT Act, the Tribunal was not empowered to remit only some aspects of the Price Determination. If satisfied the Price Determination was neither correct nor preferable by reason of an error in respect of one or more of its components, the Tribunal was bound to set it aside and send the matter of the fixing of a price determination, as a whole, to ESCOSA for reconsideration.
The Tribunal was correct to consider each line item of the Price Determination, even though a decision on whether the Price Determination should be confirmed or set aside could only be made after netting out the ‘unders and overs’. That is the essential structure of the Tribunal’s approach. Unfortunately, on some occasions the Tribunal spoke in terms of remitting the matter ‘for further consideration’ of the allowance for particular operating costs’, and spoke in terms of it being necessary ‘to remit this aspect of the determination to ESCOSA for further consideration’.[2] Indeed, as we have seen by [128] of its decision, the Tribunal purported to remit only ‘the question of the appropriate allowance for Robusto’s operating costs to ESCOSA’.
[2] Reasons (n1) at [42], [84] and [100].

The formal order of the Tribunal was not so strictly confined. It ordered and directed as follows:
Set aside and refer decision for reconsideration
Made under the South Australian Civil & Administrative Tribunal Act 2013 section 37(1)(c)(ii).
Orders of the Tribunal
1.   For the reasons set out in the Tribunal’s Reasons for Decision published on 27 October 2023, the Tribunal sets aside the decision under review and refers the decision back to the Respondent for reconsideration.
2.   The Tribunal directs that the Respondent is to reconsider the allowance in the Determination for operating costs, having regard to the following:
(a).    the Tribunal’s Reasons for Decision:
(b).    the actual operating costs incurred by the Applicant during the relevant period; and
(c).    any further relevant information submitted by the Applicant or otherwise available to the Respondent.
In directing ESCOSA to have regard to the Tribunal’s reasons, Order 2(a) requires ESCOSA to consider the Tribunal’s analysis of each line item and its decision either confirming it or identifying a preferable allowance. The direction purports to confine ESCOSA’s attention to the Tribunal’s detailed consideration of the operating costs in respect of which the Tribunal reached a different conclusion. In that respect the order is overly prescriptive and unduly confines ESCOSA’s reconsideration of the Price Determination. On the other hand, the directions do not give any guidance to ESCOSA on the underlying principle the Tribunal applied in its analysis of the operating costs. A direction identifying that principle should be given.
I am also persuaded that ESCOSA’s approach to the determination of the appropriate return on capital was inconsistent with the ESC Act. A direction should be given that ESCOSA assess the efficient financing costs of an entity operating the Mount Compass Water Business which is calculated to ensure the sustainable delivery of reliable good quality water services to the residents of Mount Compass.
On the controversy over the return on and of capital, I am persuaded that the Tribunal erred in confirming ESCOSA’s determination that Robusto’s assets which had reached the end of their expected useful life, but were still in use, should be valued at nil. A direction should be given that all assets still in use should be included in the regulated asset base albeit at a heavily depreciated value. On the other hand, a direction should be given that those assets should be valued as at the commencement of the first regulatory period.
A direction to disregard the historical losses should also be given.
I would therefore set aside Paragraph 2 of the orders made by the Tribunal on 17 November 2023 and in their place direct that the respondent is to reconsider the decision under review in accordance with the following directions:
1.The objective in making a price determination is to ensure the licence holder operates as efficiently as reasonably practicable in the circumstances pertaining to its business and industry and that the benefits of those efficiencies are passed on to the consumer after allowing the licence holder a reasonable return which assures its long-term financial sustainability as a provider of reliable, good quality services, but no more, so that monopoly rents are not recovered.
2.The assessment of a regulated entity's efficient operating costs, pursuant to the Essential Services Commission Act 2002 (SA), must have regard to the business environment in which the regulated entity operates, including its water resources, supply infrastructure, and the size, nature and geographical spread of its customer base. This assessment will generally require an analysis of the regulated entity's actual costs for any inefficient, overstated or exceptional expenditure and taking into account the absence of a close comparator against which an assessment can be made.
3.The regulated asset base used to calculate an allowance for a return of, and on, capital includes all of the infrastructure actually used at the commencement of the regulatory period. The remaining functional life of that infrastructure is a question of fact which may differ from the generally expected useful life of that infrastructure.
4.The return on capital should be calculated to secure Robusto' s continuing investment, and attract future investors, in the Mount Compass Water Business in order to secure the sustainable and reliable provision of good quality services without allowing the extraction of monopoly rents. The assessment of the appropriate WACC must be based on a stand-alone benchmark efficient entity and not the actual financial arrangements entered into by Robusto but must be made by reference to an entity operating the Mount Compass Water Business. That assessment should have regard to, and place more weight on, the particular matters mentioned in the EY report dated 9 September 2019 than on the return on capital allowed for the substantially larger enterprises from which an adjusted return for Robusto was derived in the KPMG report.
5.No allowance should be made for income foregone in the period before the price determination was made.
6.The respondent shall consider any further relevant information submitted by the appellant or otherwise available to the respondent.
My reasons follow.
Background
Robusto purchased the business of supplying reticulated water to residents of Mount Compass from Hillrise Investments Pty Ltd (‘Hillrise’) in June 2016 and was licensed by ESCOSA as a water retailer in August 2016. At the time of its purchase of the business, Hillrise had agreements referred to as either the Hillrise or Bizana Pty Ltd agreements) which had been in place for some decades with its customers. Those agreements tied water rates to the rates charged by SA Water. The rates were secured by way of an encumbrance on the customer’s land. Those agreements were superseded by an ESCOSA approved ‘standard water customer contract’, gazetted in September 2018, but the linkage to the rates charged by SA Water was not changed.
At a meeting on 28 November 2016, a senior manager of ESCOSA requested Robusto’s principal, Mr Connor, to provide ESCOSA with a proposed pricing regime which was consistent with the general principles set out in a pricing determination referred to as the 2013-2017 Regulatory Determination (‘Regulatory Determination’). The Regulatory Determination applied to minor and intermediate retailers of water. Robusto was also asked to provide a draft standard contract for review by ESCOSA.
On 17 March 2017, Robusto applied to ESCOSA to fix prices for the water it supplied, in accordance with Regulatory Determinations and the NWIPP. Robusto forwarded its price submission to ESCOSA for review and, in that submission, proposed an annual revenue for its business over the following five years of $648,100 per annum.
There followed a protracted series of exchanges between ESCOSA and Robusto, during which Robusto submitted three further pricing proposals, the last of which was submitted in September 2019.
On 4 December 2019, ESCOSA informed Robusto that it had resolved to make a price determination apply to Robusto pursuant to Part 3 of the ESC Act and Part 4 of the Water Industry Act 2012 (SA) (‘WI Act’). On 7 January 2020, ESCOSA published its decision to make a price determination on its website.
Robusto was informed by ESCOSA on 17 January 2020 that it would proceed to make a determination expeditiously, but Robusto did not hear from ESCOSA until 24 July 2020 when ESCOSA asserted that it was no longer required to adopt the NWIPP and suggested a different process; a process which would not just consider Robusto’s current costs, but also expected future costs (including prudent capital upgrades).
On 5 August 2020, Robusto expressed its concerns to ESCOSA that the decision to adopt a new process appeared to have been made by ESCOSA staff and not the Commission itself. Information subsequently obtained by Robusto on a Freedom of Information application revealed that very little had been done by August 2020 to draft a price determination.
On 25 August 2020, ESCOSA informed Robusto that it intended to publish a new price determination for public consultation after its meeting scheduled for 22 September 2020.
On 29 September 2020, ESCOSA issued a draft price determination to Robusto for fact checking, instead of issuing a draft determination for public consultation.
On 25 May 2021, ESCOSA published the Initial Determination.
Robusto applied for an internal review of the Initial Determination on 23 June 2021. ESCOSA engaged external consultant firm HoustonKemp to prepare a report as part of its internal review process. After receiving the HoustonKemp report, ESCOSA, on 26 August 2021, made the Price Determination and adopted HoustonKemp’s report as its reasons for the purposes of s 31 of the SACAT Act.
Robusto lodged an application to the Tribunal challenging the Price Determination, pursuant to s 34 of the SACAT Act.
The Price Determination was fixed to operate for the period 1 April 2021 to 30 June 2022 (the first regulated period). While the Price Determination’s legal effect is now spent, future determinations by ESCOSA of the rates, charges and caps for the supply of water by Robusto will be affected by the properly determined base rate for that period.
The Tribunal’s power to give directions
By Ground 1, Robusto complains that the Tribunal erred in holding that it was not empowered by s 37(1)(c)(ii) of the SACAT Act to give prescriptive or detailed directions to ESCOSA governing its reconsideration of the Price Determination.
Before considering that ground further, it is appropriate to set out the statutory framework under which water pricing determinations are made.
Section 17 of the WI Act declares the water industry to be a regulated industry for the purposes of the ESC Act. A water industry relevantly means any operations associated with the provision of water services,[3] and water services relevantly includes the reticulation or supply of water.[4] Robusto’s business was therefore a regulated industry pursuant to s 17 of the WI Act.
[3] s 4 WI Act (definition of ‘water industry’).
[4] Ibid (definition of ‘water service’).
Section 35 of the WI Act provides that ESCOSA may regulate prices for retail services, subject to such pricing orders as may be issued by the Treasurer setting out policies or parameters in respect of pricing. The pricing orders of the Treasurer include an order that ESCOSA apply the NWIPP.
Section 35(5) of the WI Act provides that a pricing order takes effect on the date specified in the order. Section 35(5) of the WI Act must be read together with s 26(6) of the ESC Act, which provides that a price determination made pursuant to a provision like s 35(5) of the WI Act takes effect on the day on which notice of its making is published in the Government Gazette or on a later date of commencement specified in the determination. Section 26(6) of the ESC Act therefore contemplates that the commencement date specified in a determination will only be prospective. The preferable construction, on reading the Acts together, is that ESCOSA is not empowered to make retrospective pricing orders.
Part 3 of the ESC Act empowers the Commission to make price determinations if so authorised by other legislation which regulates a particular industry. Section 25 of the ESC Act provides that a price determination may regulate prices in a number of ways and mandates the matters to which the Commission must have regard:
25—Price regulation
(1)The Commission may make determinations regulating prices, conditions relating to prices and price-fixing factors for goods and services in a regulated industry.
(2)The Commission may only make a price determination if authorised to do so by a relevant industry regulation Act or by regulation under this Act.
(3)A Price Determination may regulate prices, conditions relating to prices or price‑fixing factors in a regulated industry in any manner the Commission considers appropriate, including-
(a)     fixing a price or the rate of increase or decrease in a price;
(b)     fixing a maximum price or maximum rate of increase or minimum rate of decrease in a maximum price;
(c)     fixing an average price for specified goods or services or an average rate of increase or decrease in an average price;
(d)     specifying pricing policies or principles;
(e)     specifying an amount determined by reference to a general price index, the cost of production, a rate of return on assets employed or any other specified factor;
(f)     specifying an amount determined by reference to quantity, location, period or other specified factor relevant to the supply of goods or services;
(g)     fixing a maximum average revenue, or maximum rate of increase or minimum rate of decrease in maximum average revenue, in relation to specified goods or services;
(h)     monitoring the price levels of specified goods and services.
(4)In making a price determination, the Commission must (in addition to having regard to the general factors specified in Part 2) have regard to—
(a)     the particular circumstances of the regulated industry and the goods and services for which the determination is being made;
(b)     the costs of making, producing or supplying the goods or services;
(c)     the costs of complying with laws or regulatory requirements;
(d)     the return on assets in the regulated industry;
(e)     any relevant interstate and international benchmarks for prices, costs and return on assets in comparable industries;
(f)     the financial implications of the determination;
(g)     any factors specified by a relevant industry regulation Act or by regulation under this Act;
(h)     any other factors that the Commission considers relevant.
Part 2, s 6 of the ESC Act sets out the following general factors which are picked up as mandatory considerations by s 25(4):
In performing the Commission's functions, the Commission must –
(a) have as its primary objective protection of the long term interests of South Australian consumers with respect to the price, quality and reliability of essential services; and
(b) at the same time, have regard to the need to—
(i)promote competitive and fair market conduct; and
(ii)prevent misuse of monopoly or market power; and
(iii)facilitate entry into relevant markets; and
(iv)promote economic efficiency; and
(v)ensure consumers benefit from competition and efficiency; and
(vi)facilitate maintenance of the financial viability of regulated industries and the incentive for long term investment; and
(vii)promote consistency in regulation with other jurisdictions.
The considerations prescribed by paragraphs (a), (b) and (c) of s25(4) of the ESC Act mandate a focus on the particular circumstances and costs of the operators which will be bound by the price determination. Subparagraphs (d) and (e) require a broader industry focus. Section 6 of the ESC Act binds ESCOSA to pursue the long-term interests of consumers in [affordable] prices, and in good quality and reliable service provision.
The considerations in subparagraph (b) promote efficiency through fair competing and the passing on of benefits of that competition to consumers instead of the capture of those benefits by the licensed operators through a misuse of monopoly or market power.
In summary, the objective in making a price determination is to ensure the licence holder operates as efficiently as reasonably practicable in the circumstances pertaining to its business and industry, and that the benefits of those efficiencies are passed on to the consumers after allowing the licence holder a reasonable return which assures its long-term financial sustainability as a provider of quality services.
Section 26 of the ESC Act mandates certain procedures for the making of determinations.
Section 27 of the ESC Act provides that it is an offence for a regulated entity to contravene a price determination, or part of a price determination, which applies to it.
In that statutory context, the exercise of the power to make a price determination requires ESCOSA to:
·make findings of fact as to the costs of providing services;
·make evaluative judgments on an appropriate return of, and on, capital and appropriate benchmarks; and
·weigh the many competing considerations in the exercise of discretion to fix the price point which best meets the objectives set out in s 6 of the ESC Act.
Section 34(4) of the SACAT Act provides:
On a rehearing, the Tribunal must reach the correct or preferable decision but in doing so must have regard to, and give appropriate weight to, the decision of the original decision-maker.
A ‘preferable decision’ more aptly refers to a decision on a question of fact or to the exercise of a discretion. On a review of findings of facts and discretionary decisions, some deference must be given to the decision of the primary decision maker. On the other hand, decisions as to the applicable law must be reviewed against the correctness standard. Unlike findings of fact and discretionary decisions, decision makers cannot choose from a menu of applicable legal rules. There is but one law which must be correctly identified by every court or tribunal which is called upon to apply it. It follows that the weight to be given to the original decision will vary according to the nature of the decision. In that way, s 34(4) of the SACAT Act provides for a true merits review. I acknowledge that the application of the words ‘correct or preferable’ may overlap and that the expression has the nature of a hendiadys. Nonetheless, it remains useful to bear in mind that one part or the other of that duality may be more apt on some reviews than in others.
On the other hand, the appeal allowed by s 71 of the SACAT Act to this Court is an appeal by way of rehearing. Findings of fact are reviewable in accordance with the decisions of Warren v Coombes.[5] Reviews of discretionary decisions are limited to legal error in accordance with the principles in House v The King.[6]
[5] (1979) 142 CLR 531.
[6] (1936) 55 CLR 499.
Section 37 of the SACAT Act sets out the powers of the Tribunal:
37—Decision on review
(1)The Tribunal may, on a review under this Division—
(a)     affirm the decision that is being reviewed; or
(b)     vary the decision that is being reviewed; or
(c)     set aside the decision being reviewed and—
(i)substitute its own decision; or
(ii)send the matter back to the decision‑maker for reconsideration in accordance with any directions or recommendations that the Tribunal considers appropriate,
and, in any case, may make any order the Tribunal considers appropriate (including any interim order pending the reconsideration and determination of the matter by the decision‑maker, or any ancillary or consequential order, that the Tribunal considers appropriate).
(2)The fact that a decision is made on reconsideration under subsection (1)(c)(ii) does not prevent the decision from being open to review by the Tribunal.
(3)     The decision‑maker's decision as affirmed or varied by the Tribunal or a decision that the Tribunal substitutes for the decision‑maker's decision—
(a)     is to be regarded as, and given effect as, a decision of the decision‑maker; and
(b)     unless the relevant Act states otherwise or the Tribunal orders otherwise, is to be regarded as having effect, from the time when the decision reviewed would have, or would have had, effect.
(4)Without limiting subsection (3)(a), the decision‑maker has power to do anything necessary to implement the Tribunal's decision.
(5)Despite subsection (3)(a), the decision as affirmed, varied or substituted is not again open to review before the Tribunal as a decision of the decision‑maker (but may be subject to appeal under this Act).

The Tribunal does not enjoy the powers conferred by s 37(1)(b) and (c)(i) of the SACAT Act on a review from a price determination made by ESCOSA by reason of s 32(6) of the ESC Act which provides that those subparagraphs do not apply to such a review.
The price determination did not comprise a collection of separate and independent decisions on each of the line items of the operating costs and the allowances for return of, and on, capital. It follows that the only power available to the Tribunal was to set aside entirely the price determination and to send back the matter, that is, the making of a price determination, back to ESCOSA pursuant to s 37(1)(c)(ii) of the SACAT Act, with such directions or recommendations as the Tribunal considers appropriate. Accordingly, the order of the Tribunal to send the entire matter back was required by law and not because the Tribunal did not consider it to be practicable or appropriate to order that ESCOSA only reconsider certain line items.[7]
[7] Reasons (n1) at [128].
ESCOSA contended before the Tribunal that on a review of ESCOSA’s decisions, the power of the Tribunal to make directions or recommendations must be read down in order to preclude the Tribunal from giving directions which have the practical effect of varying or substituting its own decision for ECOSA’s decision. Further, ESCOSA contended that giving prescriptive directions left it with very little discretion but to make the determination which the Tribunal would have made, indirectly achieving that which is denied by s 32(6) of the ESC Act.
ESCOSA relied on the decision of this Court in Vardon v Promotion & Grievance Appeals Tribunal (No 2).[8] In Vardon, the Court considered the scope of the power of the Promotion Grievances and Appeals Tribunal (‘PGAT’) to give directions to the Chief Executive of a Public Service Department pursuant to s 64(4) of the Public Sector Management Act 1995 (SA) (‘Public Sector Act’) that ‘are, in the opinion of the Tribunal, necessary or desirable to redress the grievance’.[9] Justice Gray concluded that the Parliament intended the Disciplinary Appeals Tribunal, but not the PGAT, to have the power to set aside a decision, stating:[10]
…To construe the Tribunal’s power to give directions as extending to the making of a direction that a particular decision should be substituted for the decision under review would frustrate the apparent intention of Parliament that the Tribunal should not have the power to set aside and substitute.
[8] [2007] SASC 137 (‘Vardon’).
[9] Ibid [18].
[10] Ibid [23].
I observe here that the power conferred by s 37(1)(c)(ii) of the SACAT Act is to give any directions or recommendations that the Tribunal considers appropriate, in respect of the reconsideration of the substantive decision. On the other hand, s 64(4) of the Public Sector Act empowered PGAT to only give those directions which were necessary or desirable to redress the grievance. Justice Layton founded her decision that PGAT’s powers to give directions on setting aside a decision could not be prescriptive on those particular words, construing them to relate to a grievance about the decision making process.[11]
[11] Ibid [68].
The Tribunal accepted ESCOSA’s submissions that the Full Court in Vardon is applicable by analogy to s 37 of the SACAT Act:[12]
26.We accept the correctness of this submission. So far as we are aware there are no extrinsic materials which shed light on the reasons for the modification of the Tribunal’s powers in reviews of this kind. However, it is clear the Parliament has made a deliberate decision to limit the Tribunal’s powers when reviewing a price determination, such that it is not empowered to substitute its own decision. When regard is had to the implications of a price determination for consumers and others not a party to the review, the consultation requirements attaching to a determination, the obligations placed on the respondent with respect to a price determination once made and the unique and complex nature of the task, it is not difficult to understand why the Tribunal is not empowered to make its own determination, or vary an existing determination without further input from the respondent.
27.We also accept that for us to make prescriptive directions effectively requiring the respondent to make a further price determination in very specific terms would thwart the clear intention of the Parliament that the respondent retains the ultimate power to determine the specific terms of a price determination. Therefore, in the event we conclude we should set aside the existing determination, we accept we do not have power to tie the respondent’s hands by making detailed and prescriptive directions to the effect that the respondent’s further determination must be in particular terms.
[12] Reasons (n 1) at [26]-[27].
The first ground of Robusto’s notice of appeal challenges the Tribunal’s conclusion that it was not empowered to make prescriptive directions which would dictate the result of ESCOSA’s reconsideration. Robusto’s contention must be rejected for the following reasons.
Despite the difference in wording between the provisions, the reasoning in Vardon is applicable here. The power to give directions cannot go so far as to overcome the express removal of the powers to vary and substitute. Robusto’s reliance on the decision of Moore J in Minister for Immigration and Citizenshipv Dhanoa[13] is misplaced. His Honour’s observations about the distinction between a recommendation and a direction were made in a very different statutory context.
[13] [2009] FCAFC 153.
However, there is a good reason to construe s 37(1)(c)(ii) widely enough to include the power to give directions on the proper construction and application of governing legislation, and the pricing principles which should guide a price determination. They are aspects of ESCOSA’s decision making to which the correctness standard applies. The former obviously so, the latter because the pricing principles are a product of the proper construction of the NWIPP which ESCOSA is bound by the Treasurer’s pricing order to apply, and of the relevant considerations prescribed by s 25 of the ESC Act.
The Tribunal correctly observed that its conclusions on the correct or preferable decision will inform the directions it gives:
35. …Consistently with that conclusion, we consider that we are required to “stand in the shoes” of the respondent for the purposes of considering the issues and materials and reaching our own conclusions on what we consider to be the right questions. We consider our conclusions are then intended to form the basis of any directions or recommendations we make, in the event we do not consider the existing decision to be the correct or preferable decision on the material before us.
The Tribunal was also correct to reject ESCOSA’s contention that by reason of the limitation on its remedies, the standard of review prescribed by s 34(4) was impliedly modified. The Tribunal explained:
30. In our view, these submissions run the risk of conflating two separate, albeit related, issues. The first issue relates to what remedies we have power to grant in the event we are not minded to affirm the decision under review. The second relates to how we should approach our task of reviewing the determination before us, what questions we are required to ask and what “touchstones” we should have regard to in undertaking that task.
31.As we have indicated, we accept there are limitations on the nature of any directions we can make in this matter. However, we are not persuaded it would be appropriate or lawful for us to review the price determination before us by reference to whether we consider it, or elements of it, to have been “reasonable”. In our view, if we were to take that approach, we would no longer be undertaking merits review. We would instead be embarking on what could be regarded as a ‘hybrid’ species of review, somewhere between judicial and merits review.
…
33.As canvassed above, the Tribunal is empowered to review a price determination under s 34 of the SACAT Act. The review is by rehearing and the Tribunal is required to reach the “correct or preferable” decision. We pause to observe that this formulation is a classic hallmark of merits review. It is well accepted that “correct” in this context refers to a circumstance in which there is only one correct outcome, whereas “preferable” is apt to describe a situation in which a discretion must be exercised and there are a range of possible outcomes.
34.We acknowledge that the requirement to arrive at the “correct or preferable” decision sits somewhat awkwardly with the limitations on our powers in this matter, given we are not empowered to arrive at a different decision. There is in effect nothing in the SACAT Act or the ESCOSA Act which clarifies how the Tribunal should approach a review of this kind. However, in our view, it is clear that the review is intended to be on the merits. We consider s 34 requires that we undertake a fresh consideration of the issues before the respondent, having regard to all of the evidence before us, and reach our own conclusions on those issues. We note s 32 of the ESCOSA Act also requires that in reviewing a price determination the Tribunal must be constituted by at least one person with “knowledge of, or experience in, a regulated industry or the fields of commerce or economics”. In our view, this is a further indication that the Tribunal is required to fully engage with and reach its own conclusions on the issues relevant to the price determination, rather than simply considering whether the determination under review is reasonable.
The Tribunal concluded:
43. It follows that in making the decision, ESCOSA was obliged to make the “preferable” decision from the range of possible outcomes. It also follows that in reviewing the determination, we are in the same position. We are also obliged, within the limitations discussed above, to arrive at what we consider to be the preferable decision on the material before us. Given the limitations on our powers, in practical terms we consider this requires us to arrive at conclusions on each of the relevant issues and, if those conclusions are different from the conclusions reached by ESCOSA, to make any directions or recommendations we consider necessary and appropriate to guide ESCOSA’s reconsideration of the Determination.
44. It will be apparent from what we have said that we do not accept the respondent’s submission that our role is to determine whether the Determination under review was “reasonable”. We see little support for that proposition in the statutory framework or the applicable case law. As we have already observed, were we to adopt that approach, we consider we would be departing from the well accepted principles of merits review, which require the Tribunal to conduct “its own, independent, assessment and determination of the matters necessary to be addressed (citation omitted)”
That reasoning is both orthodox and sound. It was correct to acknowledge the tension between the standard of review and the absence of the power to substitute the correct and preferable decision. However, that tension is much greater in respect of factual and discretionary decisions than in respect of decisions on questions of law. Moreover, even in respect of the former, it is easy to understand why the legislative scheme has provided for the Tribunal to give guidance on the approach to resolving disputed facts and on the relative weight to give to competing considerations whilst, at the same time, leaving it to ESCOSA to apply those directions.
The respondent filed a Notice of Alternative Contentions, Ground 2 of which, maintained the contention it put before the Tribunal on the modified standard of review. On the appeal, ESCOSA put the modified merit review for which it contends as follows:
In reaching the “correct or preferable decision” for the purposes of section 34(4) of the SACAT Act, the Tribunal was required to consider whether the Internal Review Decision was economically reasonable in the context of the Essential Services Commission Act 2002 (SA) and the Water Industry Act 2012 (SA).
ESCOSA’s contention must be rejected and Ground 2 of the Notice of Alternative Contention dismissed. The mandate of s 34 of the SACAT Act is expressed in strong and clear terms. It is difficult to attribute to the legislature an intention to fix a different standard of review, when having expressly denied the Tribunal the power to vary or substitute decisions, it did not also expressly modify the standard of review set by s 34(4) of the SACAT Act.
The terms ‘correct and preferable’ must have the same meaning in their application to every one of the Tribunal’s jurisdictions. That meaning has been explained above. The contention that the standard on a review from a decision of ESCOSA is whether the determination is economically reasonable is not with the ordinary meaning of the words ‘correct’ or the word ‘preferable’ or the combined meaning of those words. To impose the standard for which ESCOSA contends is beyond the judicial power of statutory construction. It can only be affected legislatively. A price determination may be one of a number of economically reasonable decisions but be neither the correct nor the preferable decision. Nor can it be accepted that all of the considerations to which ESCOSA must have regard can properly be described as economic ones.
The maintenance of the ‘correct and preferable’ standard of review can be reconciled with the removal of the power of the Tribunal to vary the determination or substitute with its own. A determination which is set aside because it is affected by an error of fact, or principle, or which does not give the preferable weight to the competing considerations, might nonetheless be best reconsidered by the specialist regulator in accordance with the directions of the Tribunal.
I acknowledge that the acceptance that ESCOSA might yet make the same decision, leaves open the concerning prospect that a price determination might be bounced up and down between ESCOSA to the Tribunal. That is unlikely for several reasons.
First, upon remittal, ESCOSA must reconsider its decision conformably with the directions given by the Tribunal, and it is therefore likely that a different determination will be made.
Secondly, further evidence may be called on the remittal of the reconsideration of the entire Price Determination and ECOSA is bound to receive any further relevant evidence tendered in respect of any of its components. No direction given pursuant to s 37(1)(c) of the SACAT Act can deny or limit the statutory scope of ECOSA’s reconsideration. That, too, increases the likelihood of a different determination.
Thirdly, it does not follow that the Tribunal will set aside a decision merely because it reaches a different view as to one or more lines of expenditure, or the appropriate depreciation and return on capital or investment. The differences, positive and negative, may balance out, or the difference may be so minor as to leave ESCOSA’s decision intact as the preferable one.
I observe in passing that, strictly, Ground 2 of ESCOSA’s Notice of Alternative Contentions, ought to have been brought by way of a cross-appeal. If the contention had been accepted, its effect would have been to vitiate the Tribunal’s order setting aside the Price Determination. However, having rejected the contention nothing turns on the ESCOSA’s choice of procedure.
Grounds of Appeal
Given Robusto’s acceptance of the correctness of the order to remit the matter to ESCOSA for reconsideration, its grounds must be understood to support its contentions that different and additional directions ought to have accompanied the remittal.
Ground 2 – Reliance on the HoustonKemp Report
By Ground 2 Robusto complains:
The Tribunal erred in law by receiving into evidence relying upon evidence of an expert when then Tribunal correctly found that reasons [50] that the expert evidence was directed to the wrong question, namely whether the price determination was “reasonable” rather than whether it was preferable.
The Tribunal should have excluded from its consideration opinion evidence directed to the question of reasonableness and determine the correct or preferable decision on the other evidence before it.
Ground 2 must be understood to contend that the HoustonKemp report ought to have been disregarded in determining the directions which should accompany the remittal.
On the other hand, ESCOSA by Ground 1 of its Notice of Contention contends that the Tribunal erred in affording little weight to the decision of ESCOSA because it did not ask the correct question on the internal review.
For the reasons I have given, the Tribunal correctly rejected ESCOSA’s contention that the question on the Tribunal’s review was whether the Price Determination was economically reasonable. The question before the Tribunal was not whether the determination was reasonable or reasonably likely to meet those objectives. The correct question was, of the set of reasonable determinations which may be made, is the Price Determination the preferable one.
The HoustonKemp report generally responded to the former. In para 7.2 of its report, the question HoustonKemp posed for itself was whether the commission’s final determination was reasonable.
The HoustonKemp report explained that the reasonableness of the Price Determination had been assessed by reference to the objectives of the WI Act, the ESC Act and the NWIPP, each of which may be synthesised as being to promote the long-term interest of customers with respect to the price, quality and reliability of water services. Accordingly, HoustonKemp evaluated whether the Price Determination is reasonably likely to:
·protect consumers from the misuse of power;
·offer the service provider an opportunity to recover its efficient costs; and
·have a reasonable rate of return so that it would continue to provide the services.
However, the HoustonKemp report contained much more than its bare conclusions. The analysis of the data and the application of the regulatory and commercial benchmarks to that data was relevant to the Tribunal’s review of the Price Determination. The HoustonKemp report contained expert opinions and analysis of data which could properly inform the Tribunal’s conclusion as to the preferable determination. The report’s consideration of the degree to which the Price Determination met the statutory objectives was also relevant to the question whether or not the preferable determination had been made.
In that respect, the Tribunal correctly held: [14]
We note that not all of the expert evidence before us was directed to what we regard as the right question. Much of this evidence was directed to the question of whether the Determination was “reasonable” rather than whether it was preferable, in the sense of being the best decision that could be made on the evidence. However, we accept the submission of Mr Garnaut for the respondent that the detailed evidence of the experts is relevant to many of the issues and it would not be appropriate for us to disregard all of this simply because much of it was directed to the wrong question. We accept that, providing allowances are made for errors in the experts’ approach, their opinions on key issues are nevertheless relevant and helpful to us in resolving many of the questions before us and must be taken into account accordingly.
(citations omitted)
[14] Reasons (n1) at [50].
I would dismiss Ground 2 of the Notice of Appeal.
The reasoning of the Tribunal in the paragraph just cited, also correctly identifies why the decision of ESCOSA on the internal review could only be afforded little weight. The decision itself applied the wrong test. However, the analysis and exposition of the regulatory principles it applied, like the HoustonKemp report and the evidence of Mr Houston, were relevant and helpful. I dismiss Ground 1 of the Notice of Alternative Contentions.
Ground 3 – Operating Costs

Ground 3 of the Notice of Appeal complains that the Tribunal erred in not determining every line item of Robusto’s operating costs by reference to its actual operating costs.
On the question of the adequacy of the allowances for operating costs, both before the Tribunal and on appeal, Robusto relied heavily on the opinion of its expert witness, Mr Harris, that substantial weight should be given to the actual costs incurred by it as a supplier in a regulated industry because the profit motive would, in itself, drive efficiency and reduced costs. Robusto’s complaint in Ground 3 is that the Tribunal should have given that opinion greater weight and should have proceeded on that premise in evaluating the allowance on every line item of the operating costs.
At a level of generality, it can be accepted that the profit motive will drive efficiency, but many businesses fail to operate efficiently, despite the profit motive, for a wide range of reasons including poor management and decision making. Moreover, proprietors who operate more than one business, like Robusto, may base expenditure decisions on taxation planning and cost shifting strategies. ESCOSA validly contends that Robusto’s monopoly position, accompanied by an expectation that its actual costs will be recovered in a price determination, might also weaken the incentive. Robusto’s contention that the profit motive ensures that it always operates as efficiently as possible and that its costs are necessarily the most efficient costs, must be rejected.
ESCOSA correctly contended that a regulated water business should aim to recover their efficient costs but no more, so that monopoly rents are not recovered.[15] The objective is a succinct statement of the objectives of the ESC Act discussed in [49] above. It is a synthesis of the competing consumer-focused consideration in placita (i)-(v) and need to maintain the financial viability of regulated industries in placitum (vi). Only a balanced resolution of those considerations can ensure the primary objective stated in subparagraph (a) which is to protect the long-term interests of consumers.
[15] Report of HoustonKemp dated August 2021 at para [3.2.2].
It follows that it is the efficient operating costs of a water supply business with the customer base, water sources and infrastructure of the licence holder which must be determined. It is important to keep in mind that there is a relationship between efficient operating costs and the replacement of infrastructure. Any reduction in the operating costs by reason of aged infrastructure will generally need to be counterbalanced by increasing the assumed asset base of the business. That may be problematic if the business does not propose to renew that infrastructure.
Robusto’s actual annual costs for maintenance in the 2020-2021 and 2021‑2022 financial years were $11,656 and $10,395 respectively. However, ESCOSA had allowed $20,652 for each year on that line item. The Tribunal accepted that the error gave a windfall benefit of approximately $15,000 for the first regulatory period. The Tribunal considered each of the other line items with a view to ascertaining whether that windfall was counterbalanced by inadequate allowances on other items.
The bank fees and charges claimed by Robusto included significant amounts for charges which were not shown to be connected to the Mount Compass water business. The Tribunal was therefore satisfied that ESCOSA was correct to allow only $500. The Tribunal’s finding is not attended by error.
Robusto claimed an increase in the allowance for accounting fees from $4,000 to $5,000. Its actual costs in the financial years 2021 and 2022 were $8,515 and $7,210 respectively. The primary contention put by ESCOSA, based on the evidence of Mr Houston, was that Robusto’s accounting was not transparent as to the attribution of accounting fees between the Mount Compass water business and other businesses it operated. Mr Harris’ opinion was that the claim was reasonable. The Tribunal appears to have accepted that the preferable allowance was $5,000.
Robusto sought an increase in the Tribunal to the amount allowed for financial services fees from nil to $5,000. Those charges were primarily debit processing charges. ESCOSA contended, and Mr Harris accepted, that those charges had already been claimed in respect to bank fees and charges generally. Accordingly, the Tribunal accepted that there should be no allowance under this item.
In the Tribunal review, Robusto sought an increase in the allowance for its legal fees from $5,850 to $10,000. It relied on a historical expenditure in the financial years ending 21 and 22 which was substantially greater.
The Tribunal adopted a general approach of removing one off events, such as the commencement of the proceeding in the Tribunal from Robusto’s legal fees but otherwise looked at a historical average. That average for the financial years, June 2018 to June 2021 was $11,541, a little more than the amount of $10,000 claimed by Robusto. Again, the Tribunal effectively parked the reduction of the allowance by $4,150 until determining whether or not the emergency maintenance windfall was offset.
Robusto sought an increase in the allowance for staff costs (which included bookkeeping, management and vehicle) from $30,000 to $60,000 per annum. Robusto’s accounts had shown what the Tribunal described as a ‘remarkably uniform’ expenditure across the period of June 2018 ‑June 2022. The amount said to have been spent was $60,000 for the first three of those years and $80,000 for the final year.
Robusto deployed its staff across all of its businesses. The basis on which those costs were apportionment between the Mount Compass water business and Robusto’s evidence was not satisfactorily explained. There was some general evidence of how much time some staff, and in particular, the manager Mr Connor, was spent on the different businesses.
Both Mr Harris and Mr Houston agreed that the cost of bookkeeping was $10,000 per annum. Mr Houston allowed $15,000 for staff costs based on one day per week at an annual salary of $60,000.
Mr Harris was of the view that $24,000 per annum reflected the efficient cost of managing the regulated business based on one day per week at an annual salary of $120,000. Mr Harris opined that an additional $10,000 should be allowed for the making of regulatory submissions to ESCOSA but the Tribunal found that that one-off expenditure should not be allowed. The allowance for remaining management costs contended for by Robusto and ESCOSA was therefore $24,000, $15,000 respectively. The Tribunal accepted that the management fee sought by Robusto was appropriate on the basis that it was one fifth of an annual salary of $120,000.
ESCOSA estimated vehicle costs at $5,000 per annum. Mr Harris criticised ESCOSA’s inscrutable estimate.
Mr Harris’ estimate of vehicle costs, based on the likely kilometres travelled, was $14,000 per annum. Mr Harris apportioned the costs pro-rata between businesses allowing 50 per cent of the annual costs to Robusto being $7,000. Mr Houston questioned why an SUV vehicle was required for the water business and challenged the number of kilometres assumed by Mr Harris. The Tribunal split the difference on the vehicle costs and allowed $6,000.
The Tribunal’s preferred allowances for management fees and vehicle costs increased the overall operating allowance by $10,000. The Tribunal effectively parked that amount, it being offset by Robusto’s actual costs for lesser and actual cost for emergency maintenance.
Robusto sought an increase in the allowance made by ESCOSA for public relations and communications from $5,000 to $24,000. It relied on its history of expenditure on this item which ranged between $10,000 and $20,000 in the period commencing with the 2019 financial year and ending with the 2022 financial year. The Tribunal concluded that the correct allowance was $10,000, an increase of $5,000 on ESCOSA’s determination.
Robusto sought an allowance of $20,000 for debt recovery, mediation and dispute resolution. In its initial determination, ESCOSA allowed $12,000 but following the internal review, the Price Determination allowed only $500.
Robusto’s historical expenditure had included amounts paid to the Ombudsman for its investigations of complaints. Mr Houston proposed the lesser allowance of $500 based on a combination of the annual fee for being part of the Energy and Water Ombudsman SA (‘EWOSA’) scheme, plus a small allowance for the time spent by EWOSA on Robusto cases derived from benchmarking an industry average for the number of disputes with customers.
Robusto’s EWOSA complaints ranged from 230 complaints per 10,000 customers to 1,264 per 10,000 customers, whereas SA Waters complaint rate was six customers per 100,000 at its highest. Understandably, Mr Harris took issue with the comparison to SA Water because of the ‘markedly different operating and customer characteristics given the economies of scale enjoyed by SA Water’. Mr Harris adhered to his opinion that $12,000 was appropriate. Mr Harris also opined that Robusto’s level of complaints was related to its legacy asset base.
ESCOSA’s initial assessment was based on information it had received from EWOSA about the number of complaints against Robusto which it had investigated. In its initial determination, ESCOSA had made an adjustment for efficiency from Robusto’s actual costs $20,652 to $12,000. The Tribunal concluded that $500 was a significant underestimate of Robusto’s efficient costs. The Tribunal expected the adjustment to be significant and result in an amount closer to ESCOSA’s original allowance of $12,000 for that line item. It proposed to ‘recommend that [ESCOSA] reassess this item’ having regard to the information put before it, including Robusto’s actual costs.
It will be observed that the Tribunal’s approach to each line item of the operating costs was to investigate Robusto’s actual costs for any inefficient, overstated or exceptional expenditure. It used benchmarking with other operators as an analytical tool and not as a pattern to which Robusto was required to conform. That approach is consistent with the principles and objectives of the ESC Act.
The Tribunal observed that it expected the reconsideration to result in an allowance of approximately the $12,000 allowed in the initial determination. It also expected that together with other upward adjustments, the preferable determination would exceed ESCOSA’s allowance for operating costs, despite the $15,000 windfall for emergency maintenance. As we have seen, it held that for that reason, it would be necessary to remit the operating allowance component of the Price Determination to ESCOSA.
Grounds 4 and 5 - Return of Capital and Return on Capital
It is common ground that in making a Price Determination in a regulated industry allowance must be made to enable the regulated business to extract a reasonable return on its capital investment in the business and, over time, secure the return of that capital.
On Robusto’s calculation, $88,300 should be allowed for each of the financial years ending in 2021 and 2022. ESCOSA’s on the other hand, yielded an annual amount of $24,804 for those years.
Robusto sought an allowance of $88,300 annually for its return of capital in July 2018.
In respect of the return on capital, Robusto’s proposed claim was for $135,000 for the 2020-21 financial year and the 2021-22 financial yar. That claim proceeded on a regulated asset base of $900,000 to which was applied a return of 15 per cent.
It is convenient to commence by explaining the way in which ESCOSA arrived at those allowances and the Tribunal’s acceptance of the correctness of ESCOSA’s approach.
It is necessary to allow for the return of capital expenditure through a price determination so that the service supplied is sustainable. It allows prudent business operators to make provision for the replacement of their plant and equipment as it passes its useful life. Even though the formulae for calculating an annual return of capital is likely to approximate tax deductibility of the depreciation of plant and equipment, the very different objectives of both concepts must be kept in mind.
It is common ground that the equation from which the return of capital is derived is:
DRC = Depreciation Replacement Cost
The Commission then calculated the depreciation schedule for each individual asset using the following formula:
Robusto’s complaints about the calculation of its DRC are found in Grounds 4 and 5 of the Notice of Appeal.
Ground 4
·The Tribunal erred in law in finding that the depreciated replacement cost of the regulatory assert base was valued at $799,941 to be considered from 1 April 2021 for the purposes of the allowance for the return on capital.
·The Tribunal should have found that the depreciated replacement cost of the regulatory asset base was $862,000 to be considered from 18 December 2018 for the purposes of the allowance for the return on capital.
Ground 5
·The Tribunal findings at reasons [155] a wrong as a matter of law as the analysis proceeds on the assumption that there was “no [price] determination in place”. The Tribunal failed to consider its own findings at reasons [6] that in the period up to 1 April 2021 the appellant was subject to the respondent’s general determinations as a minor retailer of water and sewerage services.
·The Tribunal should have found that the legacy date was either the date that it was granted a licence, namely, 10 August 2016 or the date when the parties agreed the capital Deed Depreciated Replacement Cost of the regulatory asset base was $862,000, namely 18 December 2018.
Robusto engaged the engineering firm at Wallbridge Gilbert Aztech (‘WGA’) to estimate the replacement cost of its infrastructure. WGA estimated ‘the as-new replacement cost of Robusto’s water assets to be $1,958,399 in December 2018 dollars.’[16]
[16] Reasons (n1) at [131].
WGA’s estimate of the ‘as new’ replacement cost reflected future purchases as the life span of the assets expired but was expressed in December 2018 terms because the expenditure was not yet necessary.
In January 2019, ESCOSA engaged the accountants BRM Holdich to determine the Robusto’s DRC. Based on the WGA evaluation of remaining life and as new replacement cost, BRM Holdich calculated a DRC of $861,950 in December 2018 dollars.
Robusto disputed that valuation, but in order to expedite the making of a price determination, took the position that the asset base should be ‘indexed back’ to June 2016, when Robusto purchased the business. It is not clear to me what is meant by ‘indexing back’ and how that concept relates to the fundamental principles on which a return of capital is determined for a regulated industry. Mr Houston understood Robusto’s position to be that BRM Holdich’s valuation of the assets should be treated as a valuation as of June 2016 and that the asset value in September 2018 should be taken to be $900,410. I am not sure that Mr Houston’s characterisation is correct. Rather, it appeared to me from Robusto’s submissions that its position was that its return of capital in the regulated years should be calculated to take into account the return it had not recovered from when it commenced operating the Mount Compass water supply business in 2016. It seeks a price determination which would allow it to recover a return of capital for the period 2016 to 2018 foregone in those years. On either view, the contention must fail. If the latter, Robusto’s submissions relies on a higher regulated asset base when its assets had depreciated between 2016 and 2018, and no further capital expenditure had been expended. If the former, it was a claim for historical losses which must be rejected for the reasons given in [169]-[173] below.
A subsidiary dispute between ESCOSA and Robusto was whether or not any remaining asset life should be attributed to assets that had passed WGA’s estimated useful life as of December 2018. If an asset is treated as having no useful remaining life, it is removed from the DRC. Robusto proposed that they should be given a positive future life value of between one to five years. By giving those assets a useful life, they would add to Robusto’s DRC.
ESCOSA rolled over the regulatory asset base from one year to another by depreciating it by that sum of $24,804. Robusto claimed that the assets that were ascribed zero value had a value of $407,514. Robusto relied on its arm’s length purchase of the water retail business for $1.05 million in 2016.
Mr Houston’s position was that it was standard regulatory practice to attribute a nil value to assets which had surpassed their expected useful life, even if they continued to be employed in the business. ESCOSA’s justification for that approach was that the continued utilisation of assets in poor condition, or in need of replacement, may lead to lower service standards and that the service provider would have no incentive not to replace plant and equipment that was beyond its useful life.
Robusto relied on a letter from WGA dated 21 June 2021 which explained:
Based upon the age of the infrastructure originally installed at Mount Compass our inspections of the same, and reports and evidence provided to us by Myponga Plumbing who have serviced the network for many years, it was (and remains) clear that much of the network infrastructure is at or near the end of its design life.
This does not imply that it is expected to fail immediately, but prudent managers of essential service infrastructure will anticipate repairs and replacement as required as the infrastructure gets closer to the end of its projected life.
Items that we noted as due to immediate replacement could (and evidently have) last for some years longer than their design life.
As such it would be erroneous to ascribe a life span of zero years to them as we understand has been the interpretation – we suggest one – five years in most cases, but again this is dependent on local conditions. It would also be erroneous to treat these assets as having no value - this was not stated nor implied in our original report.
WGA approach is both practical and better reflects the actual condition of Robusto’s infrastructure. If ‘local conditions’ have resulted in a useful life which exceeds the industry expected life, the asset still has value. Mr Houston’s observation as to standard regulatory practice can be accepted if the value of the asset has been fully recovered over successive regulatory periods. To continue to attribute a value to that infrastructure would result in a recovery of more than the capital expenditure. It should not be applied to determine the asset base for the first regulatory period.
It follows from the very purpose of the making of a Price Determination and the acceptance of the Tribunal that it should allow for a return of capital, that the depreciated replacement costs must include all of the infrastructure actually used and ascribe a value to it. Remaining useful life is a question of fact which must be determined as at the commencement of the regulatory period. The remaining functional life should not be treated as a deemed fact based on industry averages or expectations. A regulated entity for a wide range of reasons may choose not to replace aged infrastructure with new. If it does so, it must also accept that the allowance for maintenance will not be increased by reason of the retention of infrastructure past its average functional life. Of course, the value attributed to an asset which has exceeded its expected useful life will be low. Given the generality of the WGA Report it was difficult to make or identify a precise amendment to the determination.
Mr Houston, in his report prepared for the purposes of the Tribunal’s review, assumed that Robusto’s water system assets which were reported to require immediate replacement in December 2018 still had some 2.5 years of remaining life. The regulatory period starting on 1 April 2021. Accordingly, their life would have come to zero by the start of the regulatory period. However, the proper approach is to have regard to the fact that the plant equipment and infrastructure was still operational in 2021 and to make a further assessment of its life at that point. Mr Houston opined only that it would be reasonable for the commission not to include the assets, but he accepted that there is not only one unique reasonable outcome for the calculation of the depreciation schedule.

There is at least one qualification to the approach to the calculation of the depreciated replacement costs. Plainly enough, if an allowance has been made which has exhausted the value of the plant and equipment then no further allowance can be made for the continuing on account of depreciation if the plant and equipment or infrastructure continues to be used. To make such an allowance would provide double recovery of the costs. However, this was the first occasion on which there was a calculation of Robusto’s depreciated replacement costs. Accordingly, a valuation of the existing plant equipment and infrastructure based on its ‘as is’ expected life and not an industry average or benchmark was required. A direction should be given to ESCOSA to assign a depreciated replacement costs value to all plant equipment and infrastructure insofar as evidence about its functional life is provided by Robusto.
Associated with this question was whether the appropriate commencement point for the calculation was from the period over which the first price determination would apply, or when the assets were first regulated. Robusto’s position was that it was a regulated water supplier, albeit without a particular price determination, from the time it purchased the business. Robusto’s argument in that respect was that it operated under general policy constraints which prevented it from setting prices in its unfettered discretion so as to properly recover its operating costs and allow for a return of, and on, capital.
The Tribunal concluded that ESCOSA’s approach on the return of capital was not shown to be erroneous or require any adjustments. It accepted Mr Houston’s opinion for supporting ESCOSA’s approach.
The Tribunal found:
[146]…As Mr Houston, has pointed out, the regulatory period did not commence until April 2021, some 2.5 years after the date at which the assets were valued. As such, although ESCOSA did not have the benefit of this revised assessment, even taking this into account we accept it was reasonable for ESCOSA to treat the relevant assets as having a 0-remaining lifespan (and 0 value) for depreciation purposes as at April 2021. We also accept Mr Houston’s evidence that it is not uncommon for an asset to be in use for longer than the life ascribed to it for regulatory purposes, and this does not reflect any error on ESCOSA’s part. (citation omitted) [147] We also accept ESCOSA’s submission that there is simply no support in the expert evidence for Robusto’s approach of indexing the DRC back to 2016, or its contention that capital should be returned over a 10-year period rather than the periods assessed by WGA as reflecting the remaining lifespans of the relevant assets.
For the reasons I have explained, Robusto’s contention that all assets it continues to deploy as part of its infrastructure should be given a useful life for the purpose of calculating the regulated asset base should be accepted. That position is consistent with the objective of the ESC Act discussed in [49]-[52] above. However, for the reason given below in respect of Robusto’s claim for historical losses, the assets should be valued as at the commencement of the first regulatory period.
Return on Capital - Cost of finance
Ground 6 complains:
The Tribunal erred in law by ignoring the unchallenged evidence of Mr Connor concerning the inability of the appellant to secure lower costs of financing.
The Tribunal should have found that the appellant’s financing costs of 15 per cent were its efficient costs for the purpose of calculating a return on capital.
The NWI principles on costs recovery allow for a recovery for the cost of capital “using a Weighed Average Cost of Capital (‘WACC’).” The WACC is calculated by considering:
·The return required by debt providers, i.e., the cost of finance;
·The return required by equity holders, i.e. the cost of equity;
·The appropriate financial structure, i.e. the relative proportion of funding from debt and equity.[17]
[17] Report of HustonKemp dated 16 August 2021 at page 57.
Mr Houston’s evidence was that the allowance for a return on capital is typically derived by multiplying the value of the regulated asset base (RAB) by a rate of return on capital, calculated by reference to the weighted average cost of capital (WACC) for a benchmark firm facing a similar degree of risk.
Mr Houston also explained that the WACC is derived by ‘combining the return required by debt and equity holders for committing debt and equity capital for the purpose of providing the services’ by reference to an ‘efficiently managed, benchmark service provider … rather than an estimate derived by reference to the actual costs of the debt faced by the relevant service providers’.
In determining an allowance for a return on capital ESCOSA relied on advice given by KPMG. KPMG used SA Water as a comparator but adjusted its WACC to provide for a small firm premium and for small firm debt constraints. It assumed a size premium of 6.4 per cent based upon its research and adjusted SA Water’s efficient debt equity structure, it again applied the size premium adjustment. Finally, an appropriate WACC was calculated on the assumption that no finance was available, and the operator’s own equity had to be used.
SA Water’s proportion of debt was 60 per cent, its cost of debt 5.21 per cent, leading to a WACC (post tax, nominal, percentage of 5.1). The KPMG efficient capital structure, applying a small firm premium of 6.46, resulted in a cost of equity of 11.39 percent, a cost of debt of 5.21 per cent and a proportion of debt to equity of 60/40. The result was a WACC of 7.69 per cent. In its debt constrained model, KPMG assumed 100 per cent equity, with the cost of that equity being calculated at 9.33 per cent after applying the same 6.46 small firm premium. A WACC of 9.33 was the ultimate assessment. ESCOSA applied the efficient capital structure without debt constraint because it noted Robusto had obtained finance but fixed a cost of debt at 5.50, only marginally above that applicable to SA Water.
Applying that approach in the Price Determination, ESCOSA determined an overall efficient revenue requirement for return on capital of $53,467 (in December 2018) for the regulatory period derived by using a WACC of 5.5%, and a starting regulated asset base for 2021- 2022 of $799,941.
By Ground 6, Robusto’s contention is that its subjective financial and business circumstances precluded it from sourcing financing at lower rates than those which it is in fact paying. Robusto obtained its debt funding to purchase the Mount Compass Water Business from a related corporate entity Capitaline. It claimed that its interest costs at the time of the Price Determination were $100,000.00 annually at an interest rate of 9.52% on a $1.05 million loan.
In the alternative, Robusto contended that a return on capital of 10 to 15 per cent, as estimated by Ernest & Young, ought to have been accepted. I return to that report below.
The Commission adopted a standalone benchmark efficient entity instead of, simply factoring in the actual borrowing of Robusto. The use of a reference benchmark was said by Mr Houston to be ‘a near universal basis in regulatory pricing determination’. Accordingly, ESCOSA effectively set aside the information in relation to Robusto’s actual financial structure.
It is necessary to consider the KPMG report a little more closely. I first observe that the notional starting point for its assessment, the SA Water framework and assumptions, is surprising when the task is to calculate the efficient financing costs of a licensed entity supplying water services to the small town of Mount Compass. Moreover, the foundations for KPMG’s recommendation included a worldwide literature review of regulated industry rates of return. The ‘small firm’ premium review included firms with a market capitalisation of between of over $US300 million. Even ‘Micro-cap’ firms which KPMG considered had market capitalisation of between $US50 million and $US300 million.
I set out below the salient paragraphs from the KPMG report:
In considering whether the premium should be applied to the cost of debt or equity or both, we consider the most practical approach within Compass Spring’s regulatory context is to apply a premium to the cost of equity. We note that a business of the scale and scope of activity similar to Compass Springs is significantly to larger utilities. The debt raising issues associates with such small businesses make it impracticable to introduce a regulatory approach, such as that employed by the UK regulator Ofwat, which relies on observations from capital markets to determine an uplift in the cost of capital. Our recommendation is for a premium of 6.46% be applied to the cost of equity.
Set out below is a summary of the real vanilla WACC estimate adopted by ESCOSA as at 22 May 2020 (Base WACC), together with a low and high case for Compass Springs incorporating adjustments for size and levels of gearing. The low case scenario includes a small size premium adjustment to the cost of equity of 6.46%, holding all other inputs constant (including a gearing of 60%).
The high case scenario also considers the restricted access to debt capital markets for a business of the size and nature of Compass Springs, by setting the gearing to nil. In adjusting the level of gearing, consideration must also be given to the beta adopted, as the beta inherently reflects impact of financial leverage.
In order to compare a company geared at 60% to one without gearing, we have unlevered the regulatory equity beta of 0.65 adopting the formula outlined in section 5.2.6, to arrive at an asset (or unlevered) beta of 0.32.
The resulting WACC range is 5.11% to 6.78%. The midpoint in the range is 5.95%.[18]
[18] P447.
…
·General Principle: The rate of return should reflect the prudent and efficient financing strategy of an incumbent large water utility, which minimises expected costs in the long term, on a risk-adjusted basis;
·Supporting principle 1: The rate of return should reflect a long-term obligation on the utility to provide reliable and secure water and sewerage services to consumers. It should not solely reflect the new entrant cost of capital;
·Supporting principle 2; The rate of return should provide an incentive for SA Water to incur prudent and efficient investment in regulated assets and financing costs;
·Supporting principle 3: The rate of return should be based on consistent principles over time and should be predictable. It should change only to reflect material changes in evidence or regulatory practice; and
·Supporting principle 4: The assumed prudent financing strategy should not depend on the ownership of the regulated business (that is, the approach is indifferent to whether the entity is in Government or private ownership).[19]
…
ESCOSA has previously adopted a gearing assumption of 60% debt and 40% equity. This was considered to be representative of the Benchmark Efficient Entity, consistent with general regulatory practice in Australia and other parameters of WACC were set in that context.
We note that while the gearing assumption is consistent with regulatory precedent the majority of this precedent references large utilities with capital values that may not be directly comparable to Compass Springs.[20]
…
In the absence of gearing benchmarking study we recommend ESCOSA adopt a potential range of gearing. The top of the range being 60% commensurate with the SAW RD20 and the bottom of the range being a zero gearing in recognition of the potential difficulties of firms such as Compass Springs obtaining debt.[21]
…
Based on these parameter values, and the application of the Sharpe-Lintner CAPM consistent with ESCOSA’s regulatory guidance, we estimate a real post-tax vanilla WACC range of 5.11% to 6.78%. The midpoint in the range is 5.95%.[22]
[19] P465.
[20] P469.
[21] P470.
[22] P472.
KPMG’s conclusion and recommendation was:
The resulting WACC range is 5.11 per cent to 6.78 per cent. The mid-point in the range of 5.95 per cent.
The estimation of a regulatory WACC for relatively small firms such as Compass Springs is a challenging exercise with limited regulatory precedent. We note that ultimately ESCOSA will need to exercise its judgment as to the determination of a WACC that is both compliant with its regulatory framework and consistent with the intent of adopting an appropriate benchmark WACC that is reflective of competitive outcomes for a business such as Compass Springs.
There is much force in Robusto’s complaint that the KPMG report is more a survey of regulatory practices world‑wide of very different businesses, than an expert opinion on the relevant question. The relevant question, consistently the principles to which I refer in [49]-[52] above, is what return on capital should be allowed to the licence holder of the Mount Compass Water Business to ensure the sustainable, reliable provision of good quality services by entities willing to invest in purchasing or maintaining the business, without allowing the extraction of monopoly rents.
ESCOSA also obtained a report from Incenta. It recognised two possible approaches dealing with the appropriate WACC for smaller enterprises. Those approaches were to allow the small unit inefficiencies to be absorbed by the operator which would require Robusto’s position to be compared to that of an operator as large as SA Water with the same dealing and cost of debt assumption. Moreover, Robusto’s earnings left it with a much lower debt capacity. The alternative was to allow small unit inefficiencies to be absorbed by the customer. That would require an estimation of benchmark WACC parameters of a business with the size and other characteristics of Robusto. Incenta noted that ESCOSA had adopted the latter approach.
Incenta questioned the 60 per cent gearing assumption of ESCOSA’s calculations, noting that Robusto could not obtain a credit rating as a stand-alone entity and that its assets were used as security for its 100 per cent gearing.
Incenta’s opinion was that the gearing range was more appropriately between 30 and 40 per cent. It was also Incenta’s opinion that it is likely that the rate of return requirement was lower than ESCOSA’s reasons implied and that there was need for some reservation about KPMG’s findings and recommendations because Robusto does not share the same risk characteristics as small corporate firms.
However, a report of the firm EY, dated 9 September 2019 identified that Robusto did shoulder significant risk in non-payment of its charges. I set out the salient aspects of the EY report:
5. While Compass Spring’s licensed drinking water supply business can be defined as a monopoly, in the sense that its customers do not have the benefit of being able to choose between competing suppliers, Compass Springs is not a typical major regulated monopoly utility (e.g. like SA Water) because:
a.     It faces more uncertainty compared to SA Water due to the nature of its drinking water operations and its customer base. For example, Compass Springs does not have long-term contracts for the supply of water to Mt Compass in the same way that SA Water provides water and sewerage services to the majority of South Australia and has legislative requirements to do so under the South Australian Water Corporation Act 1994. In other words, there is less certainty that Compass Springs will have the opportunity to recover the costs of its investments over the life of the asset
b.     Is a very small business compared to SA Water with less than 170 customers, which not only brings risks as a small business, but also increases the risk of competition through alternative means (as opposed to alternative service providers). For example, the threat of competition from customers relying on rainwater tanks is not insignificant
c.     The small and local nature of Compass Springs’s customer base means that it is particularly susceptible to the non-payment of bills. For example, as ESCOSA is aware, Compass Springs has been in ongoing disputes with 15-20% of its customers about the 15% premium it charges on SA Water’s consumption rates, who have chosen to underpay or not pay their bills as a result. While Energy and Water Ombudsman SA recently determined that the correct consumption rates have been applied, Compass Springs has already incurred significant administrative costs and will continue to do so to collect outstanding debts
d.     It has greater risks to the reliability of water supply outside the control of Compass Springs due to frequent and regular power outages. The power outages result in customers losing water supply, as the storage tanks have a capacity of just 150 kilolitres. Compass Springs’s supply of drinking water has been disrupted by two catastrophic power failures over the past 3 years, each resulting in lost water supply for over 24 hours as a function of the unique location and nature of Compass Springs’s network. For example, when South Australia suffered a state-wide blackout in September 2016 for 24 hours, the disruptions to power supply were significantly longer at Mount Compass
e.     As a result of the factors described above, it faces a higher cost of debt than SA Water which reflects the greater risk that providers of debt finance would assume with respect to Compass Springs’s drinking water operations. We understand ESCOSA’s current approach to calculating SA Water’s cost of debt assumes a benchmark efficient entity with a BBB credit rating, that issues debt with a maturity of 10 years, and that 10% of its debt is refinanced every year. This approach would not be appropriate for Compass Springs’s drinking water operations which faces higher debt financing costs given the nature of its operations and customer base (e.g. Compass Springs does not have a credit rating for its Compass Springs water business and would be viewed as significantly more risky than BBB). To illustrate the additional costs and risks that Compass Springs faces in the market for financing:
i. No major Australian banks was willing to provide debt financing for Robusto’s acquisition of Compass Springs in 2016 given its size, regional location and perceived high degree of risk. This is despite Robusto being a major client of these banks and having a good credit history
ii. To be able to fund the acquisition of Compass Springs, Robusto resorted to debt financing from Turner Securities at a per annum interest rate of 14%, before obtaining finance from a Credit Union after servicing the Turner Securities loan for six months.
6. To reflect these additional risks, it would be appropriate for the return on assets for Compass Spring’s licensed drinking water supply business to be greater than that of SA Water and other major regulated monopoly utilities.
7. As a result, adopting the same return on assets for SA Water and Compass Springs would not adequately incorporate the costs incurred by Compass Springs in bearing risks that are unique to those experienced by SA Water. Further, it would not:
a.     Provide Compass Springs with the opportunity to earn sufficient revenues to recover the prudent and efficient costs of providing the services
b.     Be consistent with ESCOSA’s Price Determination and the National Water Initiative Pricing Principles.
8. Incorporating the factors above, the return on assets for Compass Spring’s drinking water business would likely be in the order of 10-12% post-tax nominal which has been estimated using a top-down assessment based on our experience in undertaking valuations for similar entities. This is an indicative estimate of the likely range of the appropriate return on asset and has not determined using a full assessment of the likely rate of return.

The Tribunal rejected the opinion expressed by EY and confirmed ESCOSA’s decision:
We find the Commission’s approach and conclusion in relation to an appropriate return on capital allowance to be reasonable, and not warranting any adjustment. In drawing this conclusion, we note that the task of determining an appropriate WACC and so allowed rate of return for small scale, benchmark water service provider is intrinsically challenging, particularly by reference to the difficulty in obtaining market-based benchmarks that reflect the circumstances faced by Robusto.
Notwithstanding, we find that the Commission approached this task in a considered manner, seeking opinions from respected third parties and then applying an appropriate degree of judgment by reference to the material at hand. As to the particular considerations before the Commission and its findings in light of them, we note that:
·It was appropriate for the Commission to place little or no weight on evidence as to the particular financing arrangements applying to Robusto itself, for the principal reason that such arrangements appear to be inconsistent with the standalone principle that needs to be applied;
·The EY material put before the Commission by Robusto was insufficiently substantive to be capable of receiving any weight; and
·The advice obtained by the Commission from both KMPG and Incenta involved different and, in some respects competing perspectives as to the matters that derive closest attention being, on one hand, the presence or otherwise of a small firm equity premium and, on the other the extent to which the usual benchmark level of debt financing should be applied.
The finding in the first dot point is sound and must be accepted. As to the second dot point, it can be accepted that the EY report did not include a desktop review of the assessment of the appropriate WACC for substantially larger firms operating very different businesses. However, its very strength is that it proffers an opinion on the appropriate WACC for a business in Robusto’s position.
The EY report more directly addresses the relevant question I have identified than the reports of KPMG and Incenta.
The assessment of the appropriate WACC for Robusto must consider a stand-alone benchmark efficient entity and not the actual financing arrangements entered into by Robusto. However, it must be made by reference to an entity operating the Mount Compass Water Business. That assessment must consider the particular matters mentioned in the EY report dated 9 September 2019.
Allowance for historical losses
Ground 7 of Robusto’s Notice of Appeal complains that the Tribunal erred in law in:
(a).Failing to make any allowances for the losses sustained by Robusto during the period when it was subject to the respondent’s general determinations as a minor retailer of water and sewerage services;
(b.)Further in characterising the appellant’s submission as indicating that it ‘[did] not consider Mr Harris’s opinions on this issue to be correct’; and
(c).And further ignoring the respondent’s inordinate and unexplained delay in issuing a price determination four years after it was sought.
Robusto’s contention is that because price determinations cannot operate retrospectively by reason of s 26(6) of the ESC Act, the charge fixed for the determination period should be increased to allow for the recovery of its losses in the preceding years. The losses which Robusto sought to have amortised and recovered in the period of the price determinations was the shortfall between the income earnt on its historical contracts and the income it would have earnt pursuant to a price determination. Robusto calculated the difference to be $1.39 million which, if amortised over 5 years, requires an increase in its capped revenue of $278,200.00 annually. Robusto put forward two alternative dates from which the losses should be calculated. The earliest was the grant of its licence in August 2016, and the other was the date of the valuation of its assets in 2018.
Robusto knew that the Mount Compass water business operated in a regulated industry. It knew that it could only levy charges at a higher rate than the SA Water charges, to which it was contractually tied if a price determination were made. As we have seen, it commenced the process for a determination in March 2017. It is in the nature of a regulated industry that market forces and the right of contract are controlled and modified. A regulatory regime necessarily entails a level of delay and bureaucracy in making determinations.
There appears to be some merit in Robusto’s complaints about the change of approach taken by ESCOSA during the course of its consideration of Robusto’s proposals and the length of time taken to make the Price Determination. However, Robusto’s claim fails for several reasons.
First, to make an allowance for the losses in the preceding years is inconsistent with the very prohibition on which Robusto founds its claim. Parliament having expressly prohibited retrospective determinations, Robusto’s claim is an attempt to achieve indirectly what the legislation expressly prohibits.
Secondly, Robusto’s contention on the merits of its claim for recovery of historical losses is that ESCOSA is to blame for the delay. Robusto’s claim is that it is ECSOCA’s misfeasance that has caused its losses. However, there is no principle of legal policy or fairness which would impose the burden of the lost income caused by ESCOSA on the consumer by charging high rates in subsequent determination periods. Robusto’s remedy, if it has any, is a claim against ECSOSA for misfeasance in public office or negligence. Both such claims face substantial hurdles having regard to the functions and objects of ECSOSA. In that respect the ECS Act would need to be considered together with the Public Sector (Honesty and Accountability) Act 1995 (SA). Whatever the position in respect of any such action, its theoretical availability militates against the proposition that Robusto should be allowed to recover them by increasing the charges it can levy against its customers.
Thirdly, Mr Houston’s opinion was that allowing for losses prior to the first determination period (legacy losses) was inconsistent with the NWI-PP. Mr Houston had never encountered a precedent for such an allowance.
Mr Harris agreed that a line precluding the recovery of legacy losses must be drawn, but he drew the line at the date Robusto’s legacy assets were valued in 2018. Mr Houston reported that over the period of 25 years of regulating or fixing prices for regulated assets, he never encountered an instance where the date of the initial asset evaluation was undertaken was said to have determined or anchored the start date of the regulatory period. The lack of consensus between the experts and the difficulty in identifying a rationale for either of the earlier dates is a further reason to reject Robusto’s contention.
The Tribunal was correct to confirm ESCOSA’s decision.
Conclusion
I allow the appeal. I set aside the directions given by the Tribunal. I order instead that the directions I have formulated in [28] above accompany the order remitting the Price Determination for reconsideration.

Robusto	Initial Determination
Bank Fees	$1,033	$500
Accounting Fees	$5,163	$4,100
Financial Service Fees	$1,033	NIL
Water Testing	$12,391	$5,300
Legal Fees	$10,326	$5,850
Staff	$61,956	$30,000
PR	$24,782	$5,000
Debt recovery	$20,652	$12,000
Bad debt	$18,582	$5,500
UV System	$2,065	NIL