RELEVANT FACTS / MATERIAL QUESTIONS OF FACT TO BE DETERMINED
56 As one would expect in an application under the ADJR Act, most of the relevant facts are not contentious. SAPN's pricing proposal dated 21 May 2015 proposed four new tariffs in the pre-existing tariff class "Low Voltage Residential", making six single rate tariffs in this tariff class:
(a) Low Voltage Residential - Single Rate
(b) Low Voltage Residential - Single Rate with Controlled Load
(c) Low Voltage Residential Solar - Single Rate
(d) Low Voltage Residential Solar - Single Rate with Controlled Load
(e) Low Voltage Residential Social Tariff - Single Rate
(f) Low Voltage Residential Social Tariff - Single Rate with Controlled Load.
57 It is helpful at this point to refer in some detail to SAPN's Pricing Proposal, including in particular the data explaining SAPN's reasons for the proposed introduction of the solar tariff and the social tariff.
58 In the executive summary, SAPN began by setting out the components of an electricity customer's retail bill, noting that the first four are the subject of the Pricing Proposal:
(a) distribution network charges (also called DUOS charges);
(b) transmission network charges (also called TUOS charges);
(c) jurisdictional scheme amount charges required to fund the South Australian Government's Feed-in-Tariff credit to the owners of qualifying solar photo-voltaic electricity generators (also called PV FiT tariffs); and
(d) metering charges.
59 The executive summary went on to refer to the proposed introduction of the solar tariff and the social tariff in the following terms:
New residential tariffs are proposed to apply to those customers with solar PV (residential Solar tariff), and for customers on hardship programs with retailers (residential Social tariff).
Analysis of residential customer load profiles has shown that customers with solar have a different load profile to other residential customers, with an average load factor over 20% less favourable than customers without solar. This is primarily caused by lower usage during the day when the sun is shining. The new residential Solar tariff ensures that these customers pay a fair price for the capacity they require whilst recognising the benefit to the network of less energy throughout on extreme summer days when the sun is shining. The new tariff gives residential Solar customers a saving of $88 in their annual retail bill. This is in addition to any other savings they have made through the use of solar PV to date.
Over the last year of the 2015-20 Price Reset consultation, the concept of a Social tariff has been discussed with some key stakeholders and was included in our Regulatory Proposal. Following feedback from stakeholders (positive and negative) we have revised the concept. This resulted in identifying residential hardship programs as a possible beneficiary of a residential Social tariff, to be financed from other residential tariff reform (eg the residential Solar tariff). There are 10,000 customers currently on such programs with retailers. The Social tariff excludes any charge for PV FiT recovery and the distribution supply charge. It was considered inequitable that hardship customers should be required to contribute to PV FiT cost-recoveries. It was also felt that requiring such customers to contribute at average levels to costs in excess of long run marginal cost may have economic repercussions for such vulnerable customers, resulting in the deletion of the supply charge for these customers. For a typical 5 MWh residential customer, the residential Social tariff will provide a further -$198 saving.
A comparison of the three residential tariffs for 2015/16 is set out below.
…
Note that where a typical residential customer previously using 5 MWh pa has installed solar PV, their annual usage will have dropped by about 20% to 4 MWh although they will still require the same network capacity for their peak summer evening demand. A 4 MWh solar PV customer's annual network bill is $729, which is 10% lower than the typical 5 MWh residential customer's bill that uses the same capacity.
60 Hence, it is said by SAPN the key drivers of the decision to introduce the solar tariff were threefold. First, analysis of available data demonstrated that PV customers exert a similar peak demand on the distribution network, and so their capacity requirements are equivalent to those of non-PV customers. Second, about 50% of network costs are attributable to the level of capacity that the network is required to bear at times of peak demand (on hot summer afternoons and evenings), even though such times are rare. Third, the capacity needs of the network to address times of peak demand drives up prices per unit of energy. This is exacerbated by the circumstance that retail customers' metering is not yet sufficient to be able to charge the majority of customers' differential rates depending on whether and to what extent they draw electricity from the network at peak times. The vast majority of customers' meters are type 6 meters, which merely show electricity consumed from the network between manual meter reads.
61 It is not the role of the Court on an application under the ADJR Act to decide the correctness or otherwise of those key drivers. Nor is it the role of the Court critically to assess their significance or whether there are then other significant matters which should have informed the SAPN proposal. The Court has noted above the broad structure under which such decisions are made (subject to their approval by the AER).
62 The peak demand/capacity considerations outlined above, and their pricing implications, were introduced in the Pricing Proposal as follows:
3.4 Climatic conditions
Adelaide and much of South Australia has a dry climate featuring greater extremes of summer temperature than most other Australian capitals. Extended periods of heatwave conditions can occur in summer (March 2008, January-February 2009, November 2009 and January 2014 are recent examples of extended heatwaves).
During these heatwave periods, summer daytime temperatures can exceed 40oC for several days in a row and overnight minimums can remain above 30oC for some of those days.
3.5 SA Power Networks' customer and demand profile
The South Australian climate has led to an extraordinary demand for air conditioning. Over 90% of homes are air conditioned with the airconditioned floor space of these homes increasing each year. The consequent high peak network demand occurs for only a small part of the year. At other times in summer, milder weather often occurs which requires no air conditioning in most homes.
Extremely "peaky" conditions such as these heatwaves require network assets and capacity that is under-utilised during much of the year, driving distribution costs higher on a per unit of energy served basis than comparable interstate networks. These conditions also provide the impetus for SA Power Networks' network tariff strategies and innovative tariff developments described in Sections 5.4 to 5.8.
A more recent development has been the customer uptake of PV systems by small customers. Over 23% of residential customers now have PV systems operating, reducing their use of energy when the sun is shining. The incentives of the solar PV feed-in tariff schemes have been popular, and customers have responded to the incentives provided. We are seeing quite different load profiles from such customers today which warrant some consideration. This is discussed later in the proposal.
63 Chapter 4 of the Pricing Proposal sets out the tariff classes and tariff structure adopted by SAPN, and the cost components to be recovered under that structure. As explained by SAPN in section 4.2:
The grouping of customers into standard control service tariffs has historically distinguished between customers on the basis of the following factors:
• The nature and extent of usage of different types of customer;
• For business customers, nature of connection to the network, including the capacity and location or voltage of connection;
• Whether the customer also receives a controlled load service; and
• The type of meter installed at the premises, with a distinction between Types 1-4 metering and Types 5-6 metering.
Two new factors have been created for this pricing proposal:
• Whether a residential customer has a solar PV system installed, as this results in a different load profile being assumed; and
• Whether the retailer has advised that the customer is participating in a hardship program, as this may result in the assignment of the residential Social tariff.
It should be noted that SA Power Networks uses net metering for all customer billing. We distinguish in the network tariff between residential customers with PV micro-generation and those without only to the extent that they reflect demonstrably different load profiles in the customer usage, not in respect of any energy exported.
Clearly, in establishing tariff classes that are to be used for the purpose of monitoring pricing compliance, it is desirable and appropriate that similar individual tariffs should be grouped together.
…
SA Power Networks' network tariffs and tariff classes for 2015-16 are shown in Figure 3 below. The tariff classes have been constituted with regard to the provisions of clause 6.18.3(d) of the Rules concerning economic efficiency and transaction costs. As is apparent from the diagram, the suite of tariffs provides:
• A range of tariffs which are dependent upon a customer's size, consumption characteristics and voltage of connection (these factors are generally related); and
• A level of sophistication in the pricing and billing. In some cases facilitated by the metering arrangements.
[Figure 3 omitted]
64 Section 4.3 addressed the Low Voltage Residential tariff class. It explained that the tariff class included single rate and controlled load tariffs, controlled load tariffs being available for certain permanently installed storage water heaters, storage space heaters, and the like, accompanied by separate metering.
65 As to the single rate tariff, it was explained that there are three variations of the single rate tariff for low voltage customers in the pricing proposal:
residential;
residential with solar PV connected; and
residential social tariff (following nomination by a retailer offering hardship programs).
66 Chapter 5 of the proposal sets out SAPN's network tariff strategy, including its tariff reform objectives. It relevantly explained the nexus between the proposed new solar tariff and the advancement of economic efficiency in the provision of services via the distribution network, by the closer alignment of tariffs for services to customer segments with the long run marginal costs of supply of those services, avoiding cross-subsidisation by others. The alignment of costs with prices in that way is said to be a fundamental tenet of economic efficiency, because of the accurate price-signalling effect of doing so. That is, consumers will tend to make decisions as to whether and to what extent to consume a good or service that is more efficient in terms of costs to the economy as a whole if they are incentivised by a price that more accurately reflects the long run marginal cost of supply of the good or service: see Re Telstra Corporation Ltd [2006] ACompT 4 at [94] and [142]. The Pricing Proposal sought to establish the foundational facts for the introduction of the solar tariff. That economic justification for the proposal was not directly relevant to the present application. Whether it is correct in the present circumstances is therefore not a matter which the Court has to decide.
67 The tariff objectives are sensible: revenue sufficiency; minimising revenue volatility; pricing efficiency (as noted above); customer equity - tempering prices to some customers to a "reasonable allocated share of costs … to limit their impact on some customers; pricing stability; and pricing simplicity and transparency."
68 The need for tariff reform is said to flow from high summer peak demands during heat wave conditions, "when elements of the system have least capacity and the power factor of loads is poor". That was further explained in s 5.3 of the Pricing Proposal as follows:
Over the last five years, the new development has been solar PV systems, especially within the residential networks. We now have over 23% penetration of these systems and there can be in excess of 500 MW of generation occurring near midday for over half of the year when the sun is shining. The network challenges are different today, with problems of low load during days with mild but sunny weather. In summer, the peaks that used to occur between 2pm and 5pm have moved to become slightly lower peaks between 5pm and 8pm. There are still business network peaks during the day but residential systems and co-incident systems are now peaking later.
The next development coming is battery storage, and perhaps electric vehicles. The battery storage has the potential to soak up a lot of the excess energy being generated during sunshine and shift that to later in the day when the network peaks. How this will develop in summer is unclear, but network tariffs that give good signals for economic peak-lopping will assist in the development of a more efficient and economic network. If electric vehicles do take off, we will again need clear signals to customers that enable them to recharge their car with convenience but without creating new peaks on the network. Technologies over the next five years will create quite strong pathways and opportunities for the future development of the network.
As a consequence, the management of summer demand has a high priority in SA Power Networks' tariff reform strategies. This leads to an emphasis on providing network price signals that will encourage both residential and business customers to manage their demand by the following means:
• The price levels of existing tariff structures;
• The development of more cost reflective tariff structures; and
• The development of innovative new tariff structures.
69 Apart from securing "revenue sufficiency", the intent of SAPN by its Pricing Proposal included signalling the long run marginal cost of supply through the network tariffs.
70 Section 5.7 of the pricing proposal also explained the rationale for SAPN's proposed introduction of the social tariff.
71 It is for vulnerable customers involved in a retailer hardship program. This tariff specifically excludes the PV FiT recovery charges (as such cost-recoveries can impact severely on customers spending a more significant proportion of their disposable income on energy). It also excludes the supply charge. The tariff was to be limited to 12,000 applicants, or to 1.4% of any retailer's residential customers, based on retailer allocation. SAPN said the cost of the discounts is about $2M, and is to be financed within the Residential tariff class by the "Residential Solar Initiative". Business customers were not being asked to contribute to this initiative.
72 SAPN also discussed new meter-based tariffs for PV customers. It said:
After allowing for the diversity between these customers, we found that the average load factor (ie the average level of demand to peak demand) was 20% lower for the customers with PV than it was for those without. We are not permitted to charge differently for customers with micro-generation compared to those without except for the difference in usage patterns and load profile. As we allocate our network costs equally across customers on the basis of 50% for peak demand and 50% for energy, then the residential PV customers should be paying a higher price - the same 50% for the same capacity and 40% for the use of energy (PV customers typically use 80% of the energy from the network relative to that of non-PV customers). This implies that PV customers should pay 90% of the annual network bill paid by typical residential customers, whereas they have been paying 80% to date. PV customers should continue to pay less than other residential customers for the same capacity, just not as much less as they have been paying.
73 Consequently, it was noted, if a PV customer considered that their load profile was such that the imposition of the solar tariff was disadvantageous to them, an option available to the customer would be to transfer to an interval meter and actual demand tariff.
74 The conclusion as to improvement of alignment of costs and prices reached on this data in the pricing proposal was as follows:
Notwithstanding what the profiles reveal, the after-diversity profile of residential customers with PV is 80% of that for residential customers, and so with a 50% capacity allocation and a 50% energy allocation, the PV customer should be paying 90% of what a similar residential customer uses for the same peak capacity, not the 80% currently being paid. A slightly higher price for block 1 and block 2 has been proposed in the pricing.
75 In short, it was said that the load profile of PV customers was significantly different, with a load factor (average load divided by peak load) that is 20% "poorer" (lower) than typical residential customers, so a different tariff is used to "correctly price their usage."
76 On 3 June 2015, two weeks after the submission of the Pricing Proposal, officers of SAPN had a discussion with an officer of the AER about SAPN's proposed social tariff and solar tariff. Following that discussion, SAPN sent an email further explaining these aspects of the proposal, and attaching correspondence supporting the introduction of the social tariff from Uniting Communities (formerly UnitingCare, Wesley Adelaide, and the Adelaide Central Mission) and from the Energy Project.
77 On 4 June 2015, an officer of the AER sent an email to SAPN outlining concerns about the inclusion in the Pricing Proposal of the solar tariff and the social tariff. In relation to the solar tariff, the email attached a staff paper that foreshadowed a rejection of the solar tariff on the basis of a purported application of cl 6.18.4(a)(3).
78 The paper went on to contend that the load profile of PV customers and non-PV customers should be considered to be similar but only on the basis that "load profile" should be construed as meaning "native" load profile. "Native" load profile, the paper explained, involved a customer's total electricity usage, irrespective of whether the source of the electricity was a micro-generation facility or the distribution network. The paper acknowledged that the network-observed load profiles of customers with and without micro-generation "will" and "must" be different, but contended that load profile should be considered to be "native" load profile:
While residential customers will have the similar underlying (native) load profiles, the network observed (net) load profile will be different due to the presence of the micro-generation. That is, if it was not for the micro-generation the load profiles of a residence with micro-generation would have a similar load profile to a residence without.
We consider that clause 6.18.5(a)(1)(3) [sic] should be read on the basis of: 1. The native or underlying load profile
The clause operates to avoid discrimination of customers with micro-generation facilities as the native load profile will be similar …
Practically this means a household with micro-generation and a household without micro-generation could not receive different price signals as they will have similar native (or underlying) load profiles even though their net (or observed) load profiles must be different due to the presence (or absence) of micro-generation.
79 On 5 June 2015, in a telephone conference with SAPN, the AER raised concerns about aspects of the pricing proposal and requested additional data in relation to the load profile (including maximum power demand and energy usage data) of customers who have photovoltaic micro-generation facilities (PV customers) and those who do not.
80 On 9 June 2015, SAPN forwarded additional data to the AER regarding the load profile of PV customers and non-PV customers.
81 On 19 June 2015, the AER made the Primary Decision to reject SAPN's Pricing Proposal on three grounds, only two of which are relevant for present purposes, and to require amendments pursuant to cl 6.18.8(b)(1) of the Rules.
82 Relevantly, in its decision on 19 June 2015, the AER rejected the solar tariff and the social tariff in SAPN's Pricing Proposal, and gave a statement of reasons that relied upon cl 6.18.4 of the Rules in these respects.
83 The AER's statement of reasons said:
Determination on Solar Tariff
We find that the proposed Solar Tariff filed by SA Power Networks does not comply with the National Electricity Rules clause 6.18.4(a)(3). This clause requires that customers with micro-generation facilities - such as solar Photovoltaic - should not be treated any less favourably than customers without such facilities, but with similar load profiles.
SA Power Networks provided data and analysis to support its submission that customers with solar PV systems have a different load profile to other residential customers. They submitted that customers with solar PV systems have a 20 per cent less favourable load factor than customers without solar, primarily because of lower usage during the day.
We are not satisfied that SA Power Networks has demonstrated that PV and non-PV retail customers have sufficiently dissimilar load profiles. A PV specific tariff of the type proposed by SA Power Networks would therefore constitute less favourable treatment of retail customers with micro-generation facilities in contravention of clause 6.18.4(a)(3).
The data and analysis provided by SA Power Networks is not sufficient to conclude that customers with PV (micro-generation as per clause 6.18.4(a)(3)) have a dissimilar load profile to non-PV customers. The data provided did not show load profile variation over the day but rather consisted of averages and is selective. [Footnote: Half hourly data over a period from 1 July 2013 to 14 May 2015 over a population of customers comprising about 900 non-PV customers and circa 450 for PV customers).] For example, SA Power Networks focuses on the peak of the load profile only and use of a limited number of values extracted from extreme weather days. There was no consideration of the load profile over a reasonable time period (e.g. summer, winter, various times of the day and night) to show that the load profiles are dissimilar.
Furthermore, SA Power Networks has not given any consideration to the reasonable variation across the residential population's load profile.
We undertook analysis using SA Power Network's data that showed that the load profile of PV customers and non-PV customers could be considered similar. This is shown in chart 1. The red and green dashed lines represent one standard deviation of the non-PV customers load profile. The blue and purple lines, representing non-PV customers and PV customers' respectively; both sit within that single standard deviation across a very large portion of the profile.
We are therefore not satisfied that the proposed Solar Tariff complies with clause 6.18.4(a)(3) and hence this tariff should be removed from the tariff application.
84 In relation to the Social Tariffs, the AER's statement of reasons said:
Determination on Social Tariff
SA Power Networks filed a proposed residential Social tariff. The business submits that the concept of a Social tariff had been discussed with some key stakeholders and was included in its regulatory proposal. The proposed Social tariff would exclude any charge for PV FiT recovery and the distribution supply charge. SAPN submits that it is inequitable for hardship customers to contribute to the PV FiT cost-recoveries. In deleting the supply charge from hardship customers, SA Power Networks further argue that requiring customers on hardship programs to contribute at average levels to costs in excess of long run marginal cost may have economic repercussions.
We find that SA Power Networks' proposed Social Tariff does not comply with the National Electricity Rules clause 6.18.4(a)(1)(i) and (ii). This clause requires that customers should be grouped together on the basis of the nature and extent of their usage and the nature of their connection to the network.
There is no evidence to suggest that customers on hardship programs have different demand, usage or connection characteristics to other residential customers.
SA Power Networks should therefore remove the Social Tariff from its pricing proposal.
Overall, the AER considers the inclusion of a Solar Tariff and the Social Tariff to be deficiencies under clause 6.18.8(b) and, to correct those deficiencies, the AER requires the pricing proposal to be submitted without those tariffs.
85 The statement of reasons relevantly concluded with a decision by the AER (cl 6.18.8(b)(1) refers) requiring SAPN:
Actions for SA Power Networks to undertake
SA Power Networks is required to resubmit its 2015-16 pricing proposal within 10 working days of this letter by:
□ removing the Solar Tariff
□ removing the Social Tariff
□ inserting a reference to the ability of customers to choose their meter service provider, and meter type when installing a PV system.
The AER will then assess the resubmitted pricing proposal for compliance with the pricing principles and the 2015-20 preliminary distribution determination.
86 SAPN then on 25 June 2015 under protest, but to avoid the AER using its powers under cl 6.18.8(c) to itself amend the proposal, provided amended appendices to the pricing proposal, and on 26 June 2015 the Revised Pricing Proposal on a "without prejudice" basis, by which SAPN sought to preserve its ability to contend that the AER's rejection of the solar and social tariffs was invalid. As noted, there is no dispute on the part of the AER about its entitlement to do so.
87 Clause 6.18.8(d) provides that an approved pricing proposal takes effect at the commencement of the regulatory year, that is, 1 July 2015.
88 The Revised Pricing Proposal, as approved and published, is now in effect.