A recent and popular claim in discussion of the current energy crisis in Australia is that the National Electricity Market (NEM) is “broken”. Can you break an electricity market? What is the NEM, where did it come from and, if it is broken as is repeatedly claimed, how do you fix it?
The creation of the NEM in the 1990s was the culmination of more than a century of electrification across Australia’s populated regions. The National electricity Market is not a national electricity market: it doesn’t include Western Australia, the Northern Territory or parts of remote Queensland, South Australia and NSW.
Electrification in Australia began in the 1880s and was rolled out variously by private companies and local municipalities across towns and cities. Gradually through the first half of the 20th century these were brought together into state owned and run utilities, enabling greater scale and scope of connection which in turn contributed to the economic development of each state.
But each state remained an electricity island. And supply remained a state issue. In the1950s a pause on new investment during WW2 and rapid economic growth resulted in supply shortages. This triggered a massive expansion in the size and reach of each state grid. In 1955 there was around 4000MW of generation capacity. This had increased to around 32000MW by 1985[i]. Much of the underlying capacity in today’s grid was built during this period.
For most of this post-war boom period the cost of capital was low and the expansion of electricity infrastructure was seen as critical to state economic development. Over supply in capacity improved reliability and enabled state governments to negotiate attractive deals with large industrial customers. The problem was that all other customers, in particular businesses, had to take whatever was on offer by a state government monopoly.
Real costs of electricity continued to fall for much of the second half of the 20th century, mainly due to the benefits of scale enjoyed by the state owned utilities, abundant and cheap coal and gas and the absence of climate risk. By the end of the 1980s these benefits had plateaued. Inside the Government owned monopolies little had been done to improve labour productivity or other efficiencies. The sector was not exposed to either domestic or international competition.
In the early 1990s the Council of Australian Governments (COAG) decided to open up these sectors of the economy to competition. This was part of the microeconomic reform agenda that had seen the opening of the Australian economy through the floating of the dollar, reduced trade restrictions and increased autonomy of the Reserve Bank in setting monetary policy.
Introducing competition into sectors like electricity was a crucial step in delivering greater efficiency and innovation in a key input into the Australian economy. Through the 1990s the five eastern states were connected via transmission lines, state owned utilities were broken up and network monopolies were separated from generation. Competitive and contestable retail markets were established.
The NEM was formally established in 1997 as the market for wholesale electricity, or generation. Its implementation, led by the Keating Labor Government and supported by Coalition state governments, was relatively non-contentious. The big winners were commercial and industrial customers who were now able to negotiate significantly better prices for electricity from a competitive market. This reduced their costs and increased their competitiveness. The residential market (about 30 per cent of electricity in Australia) also benefitted, in particular behind deregulation of retail prices and the increased competition that followed.
The NEM is a complex but elegant design. It operates like a Swiss watch, every second of the day, every day of the year. Essentially it performs two roles: it operates a dispatch function to signal generation to enter and exit the market in real time, and it also performs an investment role, to create a price signal for old or redundant generation to exit the market and new electricity suppliers to enter.
The NEM is indifferent to technology type and to greenhouse emissions. It seeks to find the most efficient way of supplying electricity demand at any given time, given the suite of available generation bidding in. Generators bid into the market in five minute intervals. From these bids a spot price is determined by the market operator (the Australian Energy Market Operator). This is the price paid to all generators.
Behind this another complex set of forward contract markets has evolved, as both generators and buyers seeks to hedge the risk of these often volatile spot markets. The spot price of electricity can reach as high as $14,000MWh and as low as -$1,000MWh. This volatility is necessary because electricity is a commodity that cannot be stored (at least not cost effectively yet, and not at scale).
Low prices in the spot price encourage generators to stop generating. High prices encourage them to enter. And they do. When we see reports of wholesale spot prices more than $1000MWh during a heat wave, that is the market doing its job.
This sort of short run volatility is normal. If higher prices are sustained, then that becomes a signal for potential investors to build new generation to access these higher prices. Similarly if low spot wholesale prices are sustained, this is a signal for marginal generators to exit the market.
A useful test of the real success of major economic reform is that, once implemented, it is rarely challenged. We don’t question the appropriateness of a floating exchange rate or an independent reserve bank setting interest rates. Debate over the merits of a GST has collapsed since its introduction. The NEM is a success because for nearly 20 years no one questioned its creation. So if the problem isn’t the NEM, what is it?
Reducing greenhouse gas emissions has the potential to disrupt the smooth functioning of the NEM in two ways. First it requires a mechanism that considers not only the most efficient dispatch profile at any given time, but also the lowest emissions. Second, as reducing emissions will involve increased use of intermittent generation technologies like wind and solar, the NEM needs to adapt to manage the disruptive nature of these technologies at scale. These disruptions impact both the day to day operational role of the NEM, and its longer term role to signal new investment and closure of redundant generation.
The policy antidote to the first disruption has always been some variant of a price or constraint on greenhouse emissions that flows through to increase the unit cost of generation for higher emitters. Operationally and for investment signalling, this requires a carbon price (or its equivalent) that is sufficient to change the way in which generation is dispatched such that lower emissions technologies (gas) runs ahead of higher emissions technologies (coal). Renewables like wind and solar are not affected by a carbon price as they are zero marginal cost generators who run first regardless of any carbon constraint.
With a carbon constraint in place, the NEM can operate as designed. The price will shift the merit order and a new lower emissions dispatch will emerge. This will, at least in theory, incentive more lower emissions generation and accelerate higher emissions generation. High cost of generation from gas (as we have now) means the carbon price signal needs to be so much higher to achieve this shift in the merit order. This is why tight gas markets and high gas prices is such an impediment to decarbonisation of the NEM.
The stop-start-stop-start-stop discussion over implementing a carbon constraint has paralysed investment and neutered the NEM’s price signalling function. While the carbon price risk is real, it cannot be monetized by new investors.
The disruptive nature of intermittent generation is a newer challenge, laid bare by the South Australian experiment. As a result of a range of factors (higher wholesale prices, streamlined approval processes, good wind resources) South Australia has attracted around half of all the investment in wind energy resulting from the national Renewable Energy Target (RET) legislated in 2009. As a result South Australia now has approximately 50 per cent of its generation coming from intermittent sources.
The NEM and its operating rules were not varied to accommodate this development. The only substantive change made was to upgrade the interconnector between South Australia and Victoria. The NEM did what it has always done and dispatched the zero marginal cost renewables first. As they grew in scale the firm (gas and coal) generators in South Australia began to operate less like conventional generators, and more like wind turbines in reverse: they ran hard when there wasn’t much wind, and turned off when there was.
As we have seen, the effect of this was to marginalise the viability of these generators. Both Torrens Island and Pelican Point power stations were, or were in the process of being mothballed. These were only deferred by the closure of the Northern Power station last year. The increased fragility of this grid has been manifest by the significant increase and scale of reliability events in the past 18 months.
While the RET exacerbated the problem by subsidising the cost of renewables, the underlying challenge for the current NEM design remains. Zero marginal cost intermittent generators of any type are indifferent to the NEM’s spot price signals. They run when they can and turn off when they can’t, not when the market needs them. In effect, they run outside the NEM’s current design.
Of course, the situation in South Australia is not a creation of the market, but a distortion to it. The energy policy debate in Australia is now how we retrofit efficient and workable adaptions to the investment and operational roles of the market. The problem has been exacerbated by a decade of governments abrogating fundamental and decisive reform in favour of more populist and more distorting measures.
It’s hard to lay blame for the resulting crisis at the feet of the NEM. It is even harder to suggest that the NEM is somehow “broken” as a result of a decade of multiple government interventions in its efficient operation. If stakeholders claim the NEM is broken, perhaps they should be required to elaborate which part, and what remedial action they recommend to return it to its central place in Australia’s otherwise successful energy market reforms.
[i] CIGRE and AHEF, 1996, “A dictionary on electricity”, April 2017
AEMO received headlines with its goal of delivering capability for the grid to manage 100 per cent renewables at any moment by 2025. The announcement was generally well-received but, unfortunately, misreported as targeting the running of the NEM entirely on renewables. Industry participants may ask why AEMO would be indicating a 100 per cent renewables capability by 2025, which is not to say that progress towards a lower emissions market isn’t welcome.
For the first time in at least a decade the cost of solar PV to end users has been rising. This has been driven by an increase in raw material costs, and polysilicon in particular, which has reversed the trend of continuously falling lower prices for solar that we have come to expect.
State governments increasingly accept that the uptake of zero emissions vehicles requires their support to create a critical mass, with NSW the latest state to unveil a range of measures to speed up the purchase of ZEVs.
Send an email with your question or comment, and include your name and a short message and we'll get back to you shortly.