As a wholesale spot market for the price of electricity, the National Electricity Market (NEM) plays an essential role in interconnecting five regional market jurisdictions: Queensland, New South Wales (including the Australian Capital Territory), Victoria, South Australia, and Tasmania. This Chapter will explore issues that the Committee has identified as requiring further consideration when reviewing the operation of the NEM. The issues include:
market based mechanisms for improving flexibility and dispatchability;
market based demand management;
bidding in the wholesale market;
five minute settlement in the market;
market concentration in the generation sector; and
mechanisms to lower barriers to entry for new generators.
During the course of this inquiry, the Committee heard from many stakeholders—including Canegrowers, the NSW Farmers Association and the National Irrigators Council—about the real impact that increased electricity prices were having on industry. A particular concern to industry was that electricity market generators were getting a far greater return than is economically justified. The Committee’s attention was drawn to the way that the grid assets are priced.
This section of the Chapter will review capacity mechanisms, and outline five types of capacity mechanisms:
It will then review capacity markets that are currently in operation.
A capacity mechanism is an administrative measure to ensure the desired level of security of supply by remunerating generators for the availability of resources. A capacity mechanism may be needed to ensure that the provision of reserve generation capacity is economically viable.
Capacity mechanisms are considered problematic because they risk distorting the electricity market. However, where they exist on a grid such as the NEM the cost-effective concerns are reduced as the mechanism provides for inter-regional and intra-regional participation.
The introduction of variable renewable energy sources—such as wind and solar—in the electricity mix has led to a growing need for backup generation capacity. In a market with increasing amounts of renewable energy, the profitability of conventional power plants is decreasing for two reasons:
shorter run times, as for energy only purposes they are not required to the same extent as they were before the advent of renewables; and
variable renewables have a low or zero marginal cost, resulting in lower wholesale electricity prices in a competitive market.
This results in concerns that there is not enough dispatchable generation capacity available to ensure the security of electricity supply. Besides generation resources, energy storage, more interconnectors and demand response can also contribute to reliability of electricity supply.
There is debate as to whether capacity mechanisms are necessary or whether an energy-only market with time-variant scarcity pricing (based only on supply and demand for electricity) can provide sufficient incentives for the provision of spare capacity.
Figure 5.1: Capacity mechanisms
Source: Agency for the Cooperation of Energy Regulators, “Capacity Remuneration Mechanisms and the Internal Market for Electricity”, 30 July 2013, p. 5
The flow chart in figure 5.1 outlines five types of capacity mechanisms, including price based, and volume based mechanisms. These mechanisms are outlined below.
It is important to note that in volume-based mechanisms, policy-makers decide on the required volume of capacity and let the market set the price. In price-based mechanisms, policy-makers set the price and let investors decide how much they are willing to invest for a given price. Targeted mechanisms reward only specific plants or technologies, whereas market-wide mechanisms reward all capacity providers.
In a strategic reserve scheme, some generation capacity is set aside to ensure security of supply in exceptional circumstances, which can be signalled by prices in the day-ahead, intra-day or balancing markets moving beyond a specified threshold. A market operator, such as AEMO, determines the amount of capacity required to be set aside and dispatches it when the circumstances require it. This capacity is typically procured and paid for through a tender process with the costs borne by the network users.
A capacity obligation scheme is a decentralised scheme where obligations are imposed on large consumers and suppliers to contract a certain level of capacity linked to their self-assessed future consumption or supply obligations. The capacity to be contracted is typically higher than the level of expected future consumption and supply. The obligated parties can fulfil their obligation through ownership of plants, contracting with generators/consumers and/or buying tradable capacity certificates. Contracted generators/consumers are required to make the contracted capacity available to the market in periods of shortages. A market for capacity certificates may be established to promote the efficient exchange of these certificates.
A capacity auction scheme is a centralised scheme in which the total required capacity is set several years in advance and procured through an auction by an independent body. The price is set by the forward auction and paid to all participants who are successful in the auction. The costs are charged to the suppliers who charge end consumers.
Reliability options are instruments similar to call options, where contracted capacity providers (usually generators) are required to pay the difference between the wholesale market price (the spot price) and a pre-set reference price (the strike price). Whenever this difference is positive, the option is exercised. In exchange, they receive a fixed fee, thereby benefiting from a more stable and predictable income stream. Under a reliability options scheme, the incentive for the contracted generator to be available at times of scarcity arises from the high market price and from the fact that, if not available and therefore dispatched, it will have to meet the payments under the reliability option without receiving any revenue from the market.
The holders of ROs effectively cap their electricity purchase price at the level of the strike price, since whenever the market price increases above this level, the excess will be ‘reimbursed’ through the payment made under the reliability option. A scheme based on reliability options usually rests on an obligation imposed on large consumers and on suppliers to acquire a certain amount of reliability options, linked to their self-assessed future consumption or supply obligations.
Capacity payments represent a fixed price paid to generators/consumers for available capacity. The amount is determined by an independent body. The quantity supplied is then independently determined by the actions of market participants.
Capacity markets in other jurisdictions
AEMO gave evidence that markets such as New York have a short-term capacity market:
Every year the New York independent system operator—again the AEMO equivalent—will go out and identify what they need for capacity into the market and procure that separately.
AEMO noted that in Ireland and Texas the market operator operates on a day-ahead basis to identify whether or not there are enough dispatchable resources available and if not the market operator will pay for these resources to come on—paying for operating reserve at a different price to contemporaneous spot price.
AEMO also gave evidence that PJM, a market operator in the United States, has a capacity market, where they run annual auctions and procure capacity for the next three to five years.
The Committee heard that the Western Australian market is a capacity market, which is designed to ensure continuous supply. However, the Committee heard that:
… therein lie some of the challenges, I think, that we face in WA around the future and extent to which we can more efficiently utilise that capacity to put downward pressure on prices.
One of the challenges with a capacity market is that the pricing involves a large, fixed cost all year long whether that capacity is required or not. The large fixed cost removes some of the volatility out of the marginal operating costs. It becomes quite arbitrary whether the capacity is ever utilised. It also has the effect of removing an incentive to provide capacity when there is a real opportunity to make substantial amount of money – during period of high demand. As such, there may not be sufficient incentive to provide capacity on days of high demand.
Five minute settlement resulting in an increased need for a capacity market
The Committee heard that in circumstances where a five minute settlement rule was to be introduced, which would put pressure on those forms of energy that are slower to get moving, an ‘obvious compromise’ would be to introduce a capacity market for reserve peaking capability, or introduce a day ahead market.
AEMO noted that it could operate a reserve obligation market by designing a capacity market, bringing it through the Energy Security Board and then on to COAG Energy Council for approval. In circumstances where auctions were required, AEMO would seek to run them in 2018.
The Committee heard that certain environments favour the introduction of measures to ensure capacity, in particular an influx of renewables and a move to five minute settlement. Notwithstanding concerns that capacity markets can result in consumers paying for idle power stations in order to safeguard against blackouts, given the current growth of the renewable sector, and the approved rule introducing a five minute settlement period, the Committee considers it timely that various capacity mechanisms be considered for the NEM.
The Committee particularly notes the proposal of the Energy Security Board to ensure dispatchability and AEMO’s evidence to the Committee as to how it would introduce a reserve obligation market.
The Committee recommends that the Australian Energy Market Operator review the mechanisms available to achieve appropriate generation capacity, including capacity markets, greater generation and enhanced planning.
Prioritise the inclusion of flexible dispatchable resources
Dispatchable generation refers to sources of electricity that can be dispatched at the request of power grid operators or plant owners according to market needs. Notable forms of non-dispatchable energy are tidal power, wave power, solar power, and wind power.
Dispatchable resources that are flexible are resources that have the capability to rapidly increase or decrease their output to balance system load, within 15 minutes according to AEMO.
The Australian Renewable Energy Agency (ARENA) gave evidence that the NEM does not currently price inertia. This is because historically it was provided for free by big spinning turbines, powered by coal fire, gas and hydro. ARENA noted that renewables and batteries are able to provide inertia, but it does not come for free when the system is installed. ARENA argued that renewables and batteries could be encouraged to do so either through regulation or an incentive mechanism.
In relation to synchronous condensors, which could provide inertia previously provided by thermal and conventional generation, the Committee heard evidence that General Electric was in discussions to provide such a condenser to the NEM. In particular, the Committee heard evidence that as AEMO is responsible for ensuring the system is run in a secure manner, AEMO would be an appropriate body to be liaising with General Electric in relation to such a purchase.
The Committee heard evidence that dispatchability— a resource previously taken for granted— needed to be priced. The Australian Energy Council noted the AEMO evidence that in Ireland dispatchable generators can now sometimes earn more from providing system security services than from energy.
The Committee also heard about the importance of implementing a strategic reserve to facilitate dispatchable demand response in the wholesale energy market.
The Committee considers it essential that dispatchability is maintained in the NEM. Whilst the Committee notes that the COAG Energy Council’s Energy Security Board has proposed that retailers guarantee a level of dispatchability, the Committee considers that market mechanisms— such as reverse obligation auctions— could also be considered.
As outlined in Chapter 4, demand response, sometimes referred to as demand management, involves having the capacity to remove demand from the grid by agreement. This leads to grid reliability, as the market can have confidence that at times of peak demand there are mechanisms to manage some of that demand rather than relying on supply tools alone.
ARENA gave evidence that there was inadequate incentive for demand response. Noting that it was working with AEMO to try and inform the rule maker as to what might be needed in order to optimally incentivise demand response, ARENA noted that when compared with international jurisdictions Australia was not as innovative.
ARENA also provided evidence that demand response incurs some costs, including:
Sales cost – attracting the interest of a company or individuals with respect to demand management;
Contract cost – signing an agreement with that company or those individuals to implement demand management;
Metering system cost – so that the system can be monitored and managed; and
Training cost – to run through the process to ensure that when the demand management process was to be implemented, it was properly executed.
ARENA also highlighted current inefficiencies with the demand response framework, whereby AEMO was not empowered to sign demand management agreements that exceeded nine months.
The Committee heard evidence that the most economical approach to addressing the ageing coal-fired fleet was a combination of demand management in conjunction with renewables —wind and solar —as well as storage, being batteries and pumped hydro.
The Committee heard concerns that the Australian Energy Market Commission (AEMC) had removed a demand response mechanism in 2016 without proposing a replacement.
The Committee heard evidence as to the importance of the AER’s Demand Management Incentive Scheme being expedited and becoming operational as quickly as possible.
The Energy Network Association has outlined that there is strong consensus from industry, government and customer advocates that the current pricing frameworks are outdated and do not reflect the changing ways consumers use the electricity network. Notwithstanding the classes of customers who have access to peak and off peak usage charges, the Association noted that some customers are charged for how much energy they use, regardless of when they use it. As well, the real driver of network costs is the maximum electricity demand at a point in time. As customers are using the electricity network in increasingly different ways, the tariff structures are now outdated and result in ‘big, unintended subsidies from some customers to others’.
The Australian Energy Regulator (AER) gave evidence to the Committee that it has been working to implement reforms in tariffs to move to more cost reflective pricing so that consumers are better able to make informed decisions about how they use energy and technology.
The AER gave evidence that consumers do not just pay the spot price, but have overlaid the contract market and the hedges the retailers enter into with the generators to hedge against high spot prices.
ARENA noted that if there is a solar project and a battery storage is right next door, a company cannot use the public transmission network without paying the full costs of the transmission and distribution network in that entire state. ARENA argued that having transmission and distribution costs that are more reflective of the actual costs of using the system might be a useful reform.
In Western Australia there is a uniform tariff model for residential households. It combines a fixed and usage charge. The Committee heard:
Up until 2016 that [fixed] supply charge ran at around 48.6 cents per day. For July 2017, the supply charge had increased by around 95 per cent to 94.9 cents per day. The cost per unit per kilowatt hour for WA is 26.47 cents.
The Committee notes these prices were not appreciably lower than the NEM prices. It acknowledges that it is quite difficult to compare WA with other jurisdictions, precisely because of the WA uniform tariff. There is also a $500 million per annum government subsidy operating within the WA scheme.
Furthermore, the Committee heard evidence that in Western Australia, the cost to a typical household based on the current tariff is around $1,800 per year. It was highlighted that this is a much higher impost on low-income households. The impact of this is likely to be compounded following further uptake of renewables, with households that cannot afford to install solar PV left stranded on a grid that they would have an increased burden in funding.
Equity issues on the grid
One option for reform was proposed by Professor Baldwin, from the Australian National University’s Energy Change Institute. Professor Baldwin proposed a system similar to the HECS university loan payment scheme, which would see loans provided to assist households to install rooftop solar PV, batteries and computerised demand management systems in their households, with repayments linked to income levels.
Furthermore, the Committee heard of existing cross-subsidies that already exist within the NEM. Examples included:
Households without air conditioning units subsidising those with units, with a value of approximately $700 per annum;
Households in Queensland with solar PV installed on their rooftops subsidising those without solar PV, with a value of about $200 per annum.
With air conditioning units, the cross-subsidisation occurs as a result of a lack of cost reflective pricing in the market – owners of air conditioning units are not paying for the fact that they are using those units at times when electricity is trading at the highest prices on the NEM.
Providing demand management options to households can lead to savings of up to $600 per annum on energy bills, which the Committee heard could go some way to addressing inequity in the NEM.
Box 5.1: Smart meters
58 per cent of respondents reported have a smart meter installed at their premises. Of those respondents that did not have a smart meter, 76 per cent indicated that they were planning to install one in the next three years.
The Committee notes that the current pricing model does not differentiate between consumers and their usage in a nuanced manner. Given the advent of smart meters, it is incongruous to the Committee that those consumers who participate in peak power periods are not financially incentivised not to do so.
The Committee was heartened to hear that a number of stakeholders have been turning their minds to the issues of inequity on the power grid, and was particularly drawn to concessional loan style repayment approaches, such as those discussed by Professor Baldwin.
Whilst the Committee acknowledges that a move towards cost reflective pricing involves layers of complexity relating to data security and the roll out of technology such as smart meters, it is of the opinion that it is timely to review cost reflective more closely with a view to providing consumers with better value for the electricity they use—or do not use.
The Committee recommends that the Australian Energy Market Operator consider the appropriateness of a roll out of smart meters to all National Electricity Market users, ensuring that the costs of installation are borne by the beneficiaries with issues of equity accounted for.
AER to examine bidding in wholesale market
The Committee heard evidence of the importance of looking at the issues of where the grid is open to market manipulation, which engineer Graham Davies identified as one of the key reasons behind spikes in pricing. A rule change in relation to this was considered critical.
The Committee heard evidence from the Australian Energy Regulator (AER) that, following a whole series of high-priced events over the 2016/2017 summer period, it produced a series of reports that explain some of the circumstances surrounding those high-priced events. AER’s view is that the apparent rebidding by generators in Queensland did not breach any of the rules.
The AER highlighted that if there were concerns about anti-competitive behaviour, it was the Australian Competition and Consumer Commission rather than the AER that had jurisdiction. However, the AER noted that in December of 2016 it was given a new function of looking at the effectiveness of competition in the wholesale electricity market. This new responsibility will require the AER to report every two years on a broad range of issues, including barriers to entry and effectiveness of the wholesale market. The first report is due in December 2018. Following questioning from the Committee, the AER agreed that it was still a constituent part of the Australian Competition and Consumer Competition and will still be a constituent part of the ACCC at the time of reporting in December 2018.
The Committee notes that some of the transactions occurring on the grid appear to take advantage of the 5 minute bidding being settled in 30 minute increments. Given the draft rule that will see settlement occur every five minutes, the Committee makes no further comment in relation to this issue.
Of particular interested to the Committee is whether the market is designed with enough certainty to encourage the types of investment required. Whilst noting concerns that the $14,000 cap could be the subject of market manipulation, the Committee also notes the cap’s purpose in attracting new investment onto the grid.
Nevertheless, the Committee is of the opinion that accusations of ‘gaming the grid’ compromise the reputation of the NEM and should be thoroughly investigated.
The Committee recommends that the Australian Energy Regulator review concerns about rebidding practices in New South Wales during 2017.
Timely implementation of five-minute rule change
The Australian Energy Market Commission (AEMC) explained to the Committee why a rule change that was formally requested in December 2015, being the change from 30 minute settlements in the NEM to five minute settlement, was taking so long to consider:
One of the reasons for the delay was that affected stakeholders said to us, “We cannot really comment and give you feedback on this unless we know, if you are going to make this rule, how you would make it.” So we went through a process of designing a mechanism and asking them to provide feedback on that, which they are currently doing. It has taken a while, but I note that, in other markets where five-minute settlement has been introduced, the rule-making time has been, on average, 2½ years, and then there is a period for transition. I am not making excuses, but not only is it a fundamental change to the operations of the spot market but it also has, potentially, some huge impacts on the hedge market that operates alongside the spot market and provides much certainty and risk management in our market. So we really do need to understand the implications.
Whilst the change to five minute settlement would be to the advantage of newer forms of technology, the Committee heard that most existing participants in peaking and in demand response would be unable to find advantage in a change to five minute settlements.
The AEMC gave evidence that as at 26 October 2017 it had received close to 40 submissions on this draft rule, with a final draft to be released in November 2017. Some submissions agree with an implementation period of three and a half years, some requested four years and other recommended an implementation period of six years. Some others would like less than three and a half years.
The AEMC outlined why there would be a three and a half year implementation period, with reasons including that AEMO will need to make a number of changes around how the market is settled and market participants will also have to change a lot of their systems.
Professor Baldwin noted that in the context of forecasting, the five minute bidding system is:
basically an outmoded thing that was invented back in the eighties and nineties. It has been totally superseded by technology. If you want to bid into the market with a big battery storage system, you could do so in seconds. Basically, the NEM is unable to take into account advances in technology. You need forecasting for technology as well as the weather.
In the short term, the Committee supports the move to a five minute settlement and considers that this should happen as soon as is practicable, and in less than three and a half years’ time. Whilst acknowledging the importance of stakeholder consultation, the Committee is disappointed with the AEMC’s lengthy processes for considering and implementing important rule changes.
The Committee has had reference to Professor Baldwin’s comments that a five minute settlement may be a short lived process as real time settlement will inevitably be introduced.
The Committee recommends that the Australian Energy Market Commission expedite the introduction of a five minute settlement period in the National Electricity Market, so that this rule change commences sooner than 2021.
The Committee recommends that the Australian Energy Market Commission and the Australian Energy Market Operator consider the implications of real-time settlement in the National Electricity Market, and immediately commence consultation with industry, with a view to expediting any possible future rule change to this effect.
Market concentration in the generation sector and vertical integration
AGL is a good example of a company that participates in the NEM as both a generator and a retailer. Companies that perform both roles are often referred to as ‘gentailers’. Other examples of ‘gentailers’ include Origin and EnergyAustralia.
The Committee heard evidence that whilst a number of US markets operated in a similar manner to the NEM, there were some key differences. In addition to the size of those markets – PJM and Ercot together annually are about 550 terawatt hours for commercial-industrial, whereas in Australia it is about 85 terawatt hours—there is also diversity in wholesale providers and diversity in retail providers and a deep and liquid wholesale market in both electricity and gas in the US. Australian households—of which there were 9.9 million in 2016— consume approximately 123 terawatt hours annually, whereas the US Department of Energy reports that US households consume approximately 1,411 terawatt hours annually, with 125.82 million households. Those figures indicate that US households consume 11,214 kilowatts of electricity per annum, compared with Australian annual household consumption of 8,584 kilowatts.
The Committee had reference to the Finkel review final report which noted that one of the effects of the existence of ‘gentailers’ in the system is that they contribute less to the supply and trading of financial instruments, which can be an issue for other market participants. This is because non-vertically integrated retailers have limited access to risk management products.
In the October 2017 preliminary report into its inquiry into retail electricity pricing, the Australian Competition & Consumer Commission noted that as the demand-supply balance for electricity has significantly tightened, it is difficult for standalone retailers to compete with vertically integrated ‘gentailers’. There is also a concern that it may be having an impact on wholesale pricing.
The Committee notes concerns from market players in relation to vertical integration in the NEM, particularly as it relates to ‘gentailers’. In particular, the Committee notes concerns relating to conflicts of interest, limited access to risk management products by other retailers and insufficient price pressure.
The Committee recommends that the Australian Energy Regulator review the issue of vertical integration in the generation sector and market concentration in the generation sector, with a view to considering ways to ensure that standalone retailers have sufficient access to risk management products and fairly priced wholesale electricity.
New generation: barriers to entry
As noted above, the AER noted that in December 2016 it was given a new function of looking at the effectiveness of competition in the wholesale electricity market. This new responsibility requires the AER to report every two years on a broad range of issues, including barriers to entry and effectiveness of the wholesale market. The first report is due in December 2018.
When asked by the Committee whether the $14,000 per megawatt hour spot cap on the NEM has driven investment, Mr Duffy from the South Australian Government observed that in 1998—99 South Australia entered the market in a fairly tight supply situation:
Pelican Point [power station, 20 km from the centre of Adelaide, burning natural gas in a combined cycle] was built and Snuggery [power station, near Tantanoola in the Limestone Coast region and is used as a peaking power plant] in the south-east was built. Subsequently, Quarantine Power Station was built and Hallett Power Station was built after we had a very tight summer in 2000—01. The gross energy market did send very strong signals to have investment within South Australia.
Compared with this clear signalling, Mr Duffy noted that the impact of the current policy environment was to dampen the clarity of signalling:
Over the last decade, we’ve had a very different environment for investment, and I don’t think it’s driven by the gross energy market. I think it’s the question of having a clear and coherent policy on climate and energy, and having mechanisms to drive investment in that sort of framework.
Furthermore, the Committee heard that the Rules operated so that a generator can be charged to connect to the network, but the rules forbid the network from applying charges to a generator for just using the network.
The Committee was pleased to hear that the AER would be reporting biennially on effectiveness of competition in the wholesale market. In a market that once managed price signalling effectively, it would appear there is insufficient certainty to provide the conditions to drive supply.
Sufficient investment in the grid
As outlined in Chapter 4, there is significant concern that investment in infrastructure relating to modernising the grid through the inclusion of greater levels of interconnection would pass the AER’s RIT-T.
Also outlined in Chapter 4, the Committee heard evidence that new connections to the grid were frustrated by high connection costs that seemed to be unreasonably imposed on new generators, with the result being that existing generators who were already connected to the grid had an advantage.
The Committee notes that whilst the RIT-T may be operating to restrain investment in modernising projects such as the installation of additional interconnectors on the grid, it is not clear whether this sort of investment should be achieved under the RIT-T or via alternative funding methods.
The Committee is concerned that the new connections rules currently operating with respect to the grid have the effect of making it difficult for new generators to enter the grid, advantaging the original generators who are already connected to the grid.
A revitalised RIT-T and RIT-D may drive innovative solutions in the NEM, such as increased transmission lines from the roaring forties where wind power could be harnessed of the north west coast of Tasmania.
The Committee recommends that the Australian Energy Market Commission conduct reviews of the Regulatory Investment Test for Transmission, the Regulatory Investment Test for Distribution and the Rules relating to
investment in more interconnectors with a view to capturing benefits of additional interconnectors to non-adjacent states; and
new connections to the grid.
Box 5.2: The future
I know enough about the industry to know that the multiple reasons behind high electricity prices and electricity security/reliability are diverse and complex. To achieve anything meaningful, we seriously need to be able to move beyond sound grabs and point scoring.
This report has sought to capture the issues that were brought to the Committee’s attention during the course of this inquiry. The Committee is mindful that the vast majority of the stakeholders noted the importance of policy certainty in encouraging investment in the NEM, discussion of which the Committee has captured in Chapter 3. As noted in Chapter 4, the Committee also heard about the importance of planning, with this Chapter capturing many of the shorter term reforms within the existing market structures proposed by stakeholders.
The Committee acknowledges that the grid is going through significant transition and is keen to see that its work in this area leads to tangible benefits for Australian households and businesses.
Mr Andrew Broad MP
5 December 2017