Guest Feature: What Killed Australia's Electricity Market?
by James Fleay
“We can ignore reality, but we cannot ignore the consequences of ignoring reality”
- Ayn Rand
Australia once had the cheapest energy prices in the world. Earlier this month, its electricity market collapsed under its own weight. How did we get here?
Things Fall Apart
On the 15th of June, the Australian Energy Market Operator (AEMO) suspended all market operations supplying electricity to more than 9 million industrial, commercial and residential premises. The National Electricity Market (NEM) is the electricity grid stretching from the northern tip of Queensland to the sparsely populated western reaches of South Australia and via undersea cables to the remote south of Tasmania.
This is a first in the 24-year history of NEM and it is not clear what conditions will allow the resumption of normal market-driven electricity dispatch. For the time being, AEMO is manually directing generators to provide electricity to the grid based on their availability. Generators will be compensated via a formula that allows cost recovery with defined margins.
This action followed an earlier decision on the 12th of June to implement a $300/MWhr price cap in Queensland following a sustained period of escalating prices (see graph below). Predictably, this price cap spread to all the other states within 48 hours. The result: 5 GW of capacity not bidding into the market (≈12% of total dispatchable generation). Whether or not these generating companies were “gaming” the market or were refusing to sell electricity at a loss is uncertain, but it is likely that both factors contributed to the unprecedented suspension of the market.
Once the price caps were implemented, and generating capacity was withheld, AEMO forecasters realised the game had changed and began issuing blackout warnings for the evening peak in electricity demand (known as LOR3 or lack-of-reserve 3). These evening blackout warnings started in Queensland but soon spread to other states and several warnings remain in place at the time of writing.
To be sure, there are several short-term factors which have contributed to this situation.
An increase in global LNG demand (and price) resulted from intensive bolstering of European gas reserves during the northern Autumn of 2021. Locally, a combination of short-sighted policy and general energy ignorance resulted in the price of domestic gas in eastern Australia becoming highly correlated to global LNG prices - US policy makers take heed!
But, like in America, NEM has a high penetration of intermittent solar and wind resources. As a result, the NEM relies on gas to fill in the gaps at night and when the wind isn’t blowing. So, the price of domestic gas in eastern Australia has become highly correlated to global LNG prices. Thermal coal, also relied on, started a concurrent run up in price. It remains an inconvenient fact that gas and coal still account for ≈76% of Australia’s electricity and our reliance on their dispatchability inevitably forced up the cost of the electricity they produce.
Then Russia invaded Ukraine. The price of coal and gas really took off as many nations began reducing Russian energy imports. Australia is a significant supplier of both LNG and coal to the world and because we have neglected to ring-fence much of it for our own use. Thus, international demand pushed up Australian prices. Electricity retailers sounded the alarm in April and May (with several smaller retailers closing their doors) whilst others advised their customers to go elsewhere to “avoid” a doubling of their retail electricity rate – as if cheaper operations were available.
To cap it all off, about 25% of the eastern Australian coal generator fleet is out of service for maintenance, repair or can’t source fuel due to mine flooding, whilst a cold front has increased demand and sapped the performance of both wind and solar.
In a country which is amongst the wealthiest in the world and endowed with enviable energy resources, Energy Ministers for federal and state governments have been reduced to pleading with rate payers to switch off any appliances “where it is safe to do so”.
But this crisis has been many years in the making and there is ample blame to share amongst policy makers, industry, regulators, lobby groups, and the media.
Our electricity market is dead. Who killed it?
Substantial culpability lies with economists - specifically, competition economists. Most of Australia’s electricity assets – the transmission and distribution networks and coal-fired power stations - were built and operated by State governments before the 1990’s. Market liberalization and privatization throughout the 90’s and the creation of the NEM in 1998 fundamentally changed electricity system planning and capital allocation decisions.
It’s difficult to say if this pivot towards a “free market” and competition was net-positive but electricity prices did not initially rise and declined slightly in some areas. State governments were happily relieved from doing the hard work of power system planning and renewing capital stocks of critical infrastructure – or so they thought.
There are few areas of human economic activity that can’t benefit from some competition and electricity markets are probably no exception. However, there must be an underlying rationale for market design that is robust – this is not the case at present. The nonsensical nature of the competition due to the absence of engineering principles and good practice, not to mention plain common sense, needs to be examined. Those practitioners who have a deep technical understanding of power networks and generation technologies need to drive fundamental redesign of this market.
Different power generating technologies have wildly different characteristics which make them more suited to fill a particular role in the grid whilst being ill-suited to others. Striking the right balance between technologies and the resultant network design was the raison d’être of traditional power system planning engineers.
To take two extremes, let’s compare two zero carbon technologies: nuclear energy and solar/wind.
Nuclear power stations can operate for 60-80 years and are characterized by high capital costs, variable development times, high-capacity factor and reliability, very low marginal operating costs and zero emissions. This makes it an ideal technology for providing low cost, baseload electricity when allowed to continuously run close to full power. However, for economic reasons relating to utilisation, nuclear power is not suited for filling the role of the flexible, swing producer on the grid.
In contrast, solar and wind have lower capital costs and nearly zero operating costs which makes the energy they produce uniquely ideal for storage. However, their energy is intermittent, they require significant grid investment past a certain (sensible) point of grid penetration and need to be replaced every 25 years. Traditional power system planners might find that solar and wind, combined with reasonably priced energy storage, could play an important role in meeting daily demand peaks in favourable latitudes. However, they wouldn’t try and aggregate these technologies to serve as baseload power sources.
Coal, gas, and hydropower each have a different set of strengths and weaknesses. The point is this: each of these technologies has different characteristics which makes them suited to fill different roles within the overall power system depending on the natural resource endowments of that area. Power system planning engineers historically designed and built our legacy infrastructure accordingly. Introducing the parameter of CO2 emissions requires an adjustment to the math but the underlying philosophy is robust, prudent, and ignored at our peril.
However, the competition economists who directed the transformation of Australia’s state-owned power system into a market did not seem to understand this. They took pride in being agnostic about the source of electrons. No attempt was made to design the market framework to allow the various technologies to play to their relative strengths or to combine them in ways that were complementary to a given location. The engineers were locked in the basement.
The market framework that was implemented neither differentiated between which technologies nor “parts” made up the generation system, nor did it comprehend that the performance of the power system would, in time, be significantly dependent on the interactions between these technologies.
These defects of analysis and design have, over time, led to an accumulation of increased risks and amplified performance failures.
The ailing reliability of once reliable coal-fired power stations is the natural consequence of mismanaging the interaction between coal and solar/wind.
Tackling Carbon Emissions
Everything went well for a few years until concerns about the impacts of CO2 emissions on the climate became widespread. Having recently introduced a ban on nuclear power (1998), Australian governments at both federal and state levels had limited options for reducing the emissions in the electricity sector. The de-facto policy was that solar and wind were worth a shot. Generous out-of-market subsidies were introduced and, 21 years later, Australia has one of the highest penetrations of solar power in the world and is no laggard when it comes to wind. Unfortunately, a serious discussion on the challenges of intermittency was studiously avoided. The results were predictable.
Like many western nations, Australia has home-grown academics and non-technical analysts who can be relied on to trot out models that “prove” that 100% renewable is not only possible but is sensible. Energy analysts with well-founded technical reservations about grid management are not afforded a platform or a hearing. The challenges (not to mention the costs) of overcoming the intermittent nature of solar and wind are not submitted to independent scrutiny or public debate. A mainly compliant media has allowed governments, oppositions, and fringe parties to avoid any difficult questions about how the penetration of wind and solar can continue to rise.
But like sharks to blood, powerful financial institutions were attracted to the generous production credits available for solar and wind projects. These out-of-market credits allowed them to underwrite massive investment in these technologies despite reasonably low power prices – demand-driven investment disappeared completely and was replaced by subsidy-driven investment. The project developers, investment banks and superannuation funds did not have time to stop and consider how the grid and the market would absorb all this intermittent energy: they were too busy making money.
Unlike solar and wind projects, it is nearly impossible to finance new gas or coal power or even justify sustaining capital for existing assets within a spot market. The inability to justify financing new (non-subsidised) reliable generation projects as state-built legacy assets reach retirement is a persistent feature of electricity spot markets around the world.
Most of the legacy assets that we still rely on required high capital investment, underwritten with long-term contracts. This means that, after trying to underwrite dispatchable investment from the private sector, Australia’s federal government is now back in the business of developing and owning generation assets including an enormously expensive pumped hydro scheme and two large gas peaking units.
Unable to compete with heavily subsidised solar and wind during daylight hours or windy conditions, the utilisation of coal generation assets has declined, and they have begun to exit the market. Maintenance budgets for remaining units have been slashed – why maintain assets that policymakers have explicitly told you they plan to eliminate? The owners of Australia’s remaining coal-fired power stations are now in a race to retire their assets as soon as possible and retirement dates are being brought forward, leaving the Government and AEMO scrambling for alternative firm supply.
The thinking at the time was that flexible open-cycle gas-fired power would be able to fill the gaps.
Around the same time, Queensland discovered that it could produce natural gas from underground coal seams and a large LNG exporting industry was built in that state. This occurred whilst eastern Australia’s legacy gas fields in Moomba and the Bass Strait were winding down. Political forces and local activism resulted in the state governments of Victoria and New South Wales deciding that a moratorium on gas development would be electorally popular for climate reasons.
Just as our electricity system was becoming more reliant on it, Australia found a way to reduce the supply of domestic gas and have its price move in lockstep with international LNG prices.
A System Under Strain
AEMO, the market operator, fully appreciated the rising costs and difficulties of maintaining adequate generation capacity and grid stability: it was showing up in the non-energy corner of the market. Market mechanisms designed to keep the system stable (FCAS) and secure emergency reserves (RERT) were being used more frequently as the penetration of wind and solar increased, resulting in steadily increasing costs (see chart below).
AEMO glossed over these growing challenges as mere inconveniences on the otherwise clear pathway to a 100% renewable grid.
Note: The vertical axis on the RERT chart below will need a logarithmic scale to capture the 2021-2022 costs.
Quo Vadis Coal?
Despite the claims of ESG-focused financial analysts, Australian coal has not become a stranded asset and shows no signs of becoming so. The value of Australian coal exports broke records last year and will do so again this year. Because 65% of our power still comes from coal, the high international prices are being felt by Australian electricity consumers as most Australian power stations buy some or all of their coal on similar terms to international customers. Electricity consumers from the world’s largest coal exporting country (by value) are paying top dollar for coal-fired electricity.
Where to from here?
Australia has the largest reserves of uranium in the world and has a substantial export industry, but we refuse to even talk about availing ourselves of this proven, zero-emissions, reliable technology to replace our legacy coal-fired fleet.
The impacts on Australian industrial power users, their supply chains, and employees will be felt for many years.
And yet, with no sense of irony, the Government is desperate to assure us that the long-term solution is to get as much solar, wind and storage into the system as possible and to speed up the exit of coal and then gas.
Climate change and our rapidly emerging energy crisis mean our government, unburdened by alternative and contrasting viewpoints, has no time to pause and reconsider how we plan, deploy, finance, and operate one of our nation’s most important industries.
As we say in Australia; this doesn’t pass the “pub test.” Slogans are not policy and the laws of physics, and the limits of engineering are reasserting themselves. Prudent politicians are signaling a return to integrated planning. However, the stated plans of AEMO, rooted in economics (and little else), are deeply flawed. Do our politicians know this? If they don’t, they should. Credible institutional capacity is difficult to find, and people are growing tired of hearing politicians talk about energy.
James Fleay is the Chief Executive Officer of Down Under Nuclear Energy.