Electricity network tariff reform: the household solar paradox
CKI Utilities Development Pty Ltd v Australian Energy Regulator [2016] FCA 17
To encourage take-up of household solar, or to remove an inefficient cross-subsidy? The Federal Court dismisses SA Power’s challenge to the AER’s refusal of a higher network tariff for solar household customers
South Australia is providing a foretaste of just some of the challenges that the rise of renewables pose for the incumbent generation and network infrastructure of the national electricity market. Its harsh summer heatwaves have driven extensive take-up of residential air conditioning, and households have also keenly taken up the opportunity provided by household solar generation to reduce and offset their electricity bills. At least at first blush, this is to be encouraged: not only for environmental reasons, but because in-house generation reduces load on the network, and so ought to lessen the need for network capex, which consumers ultimately pay for.
But the equation is not so simple. In basic terms, networks need to be built to handle peak aggregate demand. And although solar households consume less energy from the grid overall, their peak demand requirements are similar to the peak demand of non-solar households. This is because their peak demand tends to occur just when the sun has stopped shining: around 9 pm on the 3rd or 4th day of a heatwave and the temperature outside is still stuck in the low 40s. Both the solar and non-solar households’ air conditioners are running at full tilt, and the solar panels are no longer reducing overall load on the network.
Here lies the paradox: because network tariffs are mostly charged on a simple per-kWh energy usage basis, solar households pay less to the distributors than their non-solar neighbours do. And so the solar households pay less towards the construction and replacement of network assets, even though they contribute as much to the need for those assets as the non-solar households do. This creates a cross-subsidy from non-solar households in favour of solar households (on top of the cross-subsidy involved in State government feed-in tariffs), thereby introducing an element of allocative inefficiency.
In 2015, SA Power Networks responded to that cross-subsidy by seeking to introduce a new tariff for solar household customers: the solar customers would pay a higher per-kWh usage charge than non-solar households, but would still pay lower amounts overall, reflecting their lower overall energy offtake from the grid.
In June 2014, the Australian Energy Regulator refused SA Power’s attempt to introduce the new solar tariff. SA Power applied to the Federal Court for judicial review, and the Total Environment Centre intervened in support of the regulator’s refusal to allow the solar tariff. Not unsurprisingly, the TEC was critical of the higher solar tariff, as it would act as a disincentive against continued expansion of household solar generation.
This appears to have been the first time that the AER had refused a network’s request to introduce a new category of tariff, and was the first time that the introduction of a new tariff had been litigated. Each distributor must submit its network tariffs for approval to the AER each year. Ordinarily, this involves an arithmetical check to ensure that the expect tariff revenue for that year aligns with the regulated revenue allowance for that network.
This highlights another significant aspect of SA Power’s proposal: the introduction of the solar tariff would be revenue neutral for SA Power overall, in part because SA Power offset the revenue gain from the higher solar tariff by also creating a new “social tariff” for certain customers experiencing economic hardship. Rather, the controversy was whether SA Power was able to impose network costs differentially on solar and non-solar households.
The critical provision of the National Electricity Rules provides that, in assigning a solar customer to a “tariff class”, the solar customer must be treated no less favourably that a non-solar customer with a similar load profile. Under that rule, SA Power’s challenge decision hinged on two key questions.
First, in establishing the new solar tariff, was SA Power was creating a new “tariff class”, or merely a new “tariff” within its existing low voltage household tariff class? The distinction between “tariff classes” and “tariffs” was not clearly defined in the rules, and had not previously been considered by a court. If the solar tariff was merely a new “tariff”, then the solar no-less-favourable treatment rule would not apply. Mansfield J addressed the question as a matter of text and purpose of the Rules, and accepted the argument that to characterise the solar tariff as a mere “tariff” would enable the solar no-less-favourable treatment rule to be circumvented by an “artifice of nomenclature”.
The second question was whether the AER’s decision that solar and non-solar households have a “similar” load profile was void for unreasonableness. SA Power argued that the AER’s interpretation of the data it had provided was statistically flawed, and both SA Power and the AER led expert statistical evidence on the question. Mansfield J concluded that the AER’s assessment of the data provided was not unreasonable against the (demanding) test of unreasonableness that remains after MIAC v Li (2013) 249 CLR 332. Here, his Honour cautioned against the unreasonableness ground providing a vehicle for de facto merits review.
In dismissing SA Power’s judicial review application, Mansfield J noted that SA Power had articulated an economically efficient rationale for the introduction of the solar tariff, but concluded that the AER had not applied an incorrect rule or arrived at an unreasonable decision.
The direct impact of Mansfield J’s decision may be limited, as the rules under which the case was decided will be amended from 2017, including by requiring distributors to submit 5-yearly “tariff structure statements” for approval by the AER. But the competing policy concerns that have been exposed by this case will likely remain contentious over coming years, in particular as smart meters are more widely rolled out, more elaborate time-of-use tariff structures are introduced and as households begin to combine solar generation with battery storage, further reshaping patterns and peaks of demand for network services.
