The National Electricity Market (NEM) and the emerging 'seasonality paradigm'

Australia’s National Electricity Market (NEM) is well and truly into its second phase of evolution. In 2024, 39% of the NEM’s electricity supply came from renewable sources, with about 19% coming from solar (utility-scale and rooftop). While coal remains the largest single source of supply, the market dynamic is increasingly one where coal and solar generators compete vigorously during the middle of the day, driving prices to very low or even negative levels. But, overall prices remain high because prices soar when the sun sets, especially when wind generation is low.  

This dynamic was barely present ten years ago, which begs the question – what market dynamics will we see in ten years’ time?  

Before we discuss the future, let’s first briefly discuss the first phase of the NEM’s evolution, roughly from market start in 1998 to the late-2010s. During this time the NEM operated under the ‘merit order paradigm’, where the price of electricity was primarily determined by the level of demand. 

This arose because the NEM’s market design (sensibly) required generators to sort themselves into a ‘merit order’, stacked from lowest cost to highest cost. The lowest cost generators fulfilled the ‘baseload’ role, providing the bulk of energy supply, while higher cost ‘mid-merit’ generators and the highest cost ‘peaking’ generators stepped into the market as demand and prices rose.  

We left the simple merit order world many years ago. We are now in the second phase of the NEM’s evolution, the ‘duck curve paradigm’, where the price of electricity is fundamentally determined by fluctuations in supply. Fluctuations in demand are still important but increasingly play a secondary role.  

The term ‘duck curve’ describes a daily pattern starting with moderate demand and prices in the morning, plummeting prices as the sun rises, and a sharp rise in prices in the evening as the sun sets. When drawn the pattern of prices looks something like a duck, moving from the tail (the morning), dipping down to the low of the back during midday and rising to the head in the evening.  

A key response to today’s duck curve is large-scale investment in batteries to soak up excess solar energy at low prices in the middle of the day for re-sale in the evening at much higher prices. Indeed, 4 GW of utility-scale batteries are presently under construction or financially committed and the NEM market operator anticipates at least a further 5 GW to be built in the near-term.  

While it is easy to think of the duck curve as a fundamental feature of a renewables-heavy grid, ACIL Allen’s long-term NEM modelling simulations strongly suggest that it is really a transitory phase and will soon be overtaken by a new dynamic: the seasonality paradigm.  

As its name suggests, the seasonality paradigm will mean that electricity prices are highly seasonal. We expect that prices will be consistently high in winter (other than at times of high wind output) and generally low in summer and the shoulder seasons.  

We see this emerging as a result of four key factors: 

• Reducing costs of lithium-ion batteries and other storage technologies will reduce (though not eliminate) intraday price variations characteristic of today’s duck curve 

• Continuing coal generator retirements will reduce the severity of midday over-supply, while removing the NEM’s primary source of bulk energy supply 

• As solar and wind replace coal as the primary sources of energy, supply will become increasingly driven by seasonal variations in output, particularly of solar  

• Solar’s winter decline will coincide with consistently high winter demand, meaning that natural variations in wind output will regularly result in extended periods where supply is tight and prices are high.  

ACIL Allen’s market modelling highlights a range of important questions that should be front of mind for everyone in the Australian electricity industry as we transition from today’s duck curve paradigm to the emerging seasonality paradigm: 

• What is the role of gas-powered generation (GPG) in the NEM?
  While batteries are an effective source of short-duration capacity to ‘flatten the duck curve’, they are too expensive to transfer energy from summer to winter on the scale the NEM will require. Australian pumped hydro projects have experienced very high construction costs, which makes the economics of using this technology for seasonal storage similarly challenging. This means that to eliminate coal we may have to accept ongoing use of GPG (and its associated emissions) for the foreseeable future. In our view the high cost of eliminating GPG means that the most economic approach is likely to be offsetting GPG’s residual emissions using credible carbon sinks to achieve the policy objective of net zero emissions. Blending renewable gases such as biomethane into the gas supply can also play an important role in reducing emissions from the GPG fleet.  

• Is market design really a key barrier to the NEM’s transition?
  While various market reform proposals have advocated for a capacity market that would provide fixed annual payments for firm generation capacity irrespective of whether it operates, our modelling shows that the NEM will have plenty of firm capacity in the future (especially batteries). The key driver of reliability will be the ability of generators to operate for extended periods of low wind in winter (ranging from days to weeks), rather than their ability to provide short sharp bursts of power. In this environment, payments for capacity are unlikely to effectively target the market’s key source of reliability risk.  

• What is the role of demand side response?
  During both the NEM’s first and second phases, there has been significant focus on the benefits of consumers reducing power demand for a few peak hours times to reduce infrastructure needs. However, we are moving into a world where periods of tight supply can persist for days or weeks. Small customers cannot realistically be expected to restrict electricity usage for such an extended period, and so the focus of demand response is likely to shift to very electricity-intensive businesses that have the strongest economic incentive to reduce output to avoid prolonged high prices.  

• What is the flow-on effect on retail pricing structures?
  In recent years policy-makers have spent significant effort convincing customers to accept ‘time-of-use’ pricing, where the price of electricity varies based on the time of day. However, we expect that intra-day price variations will soon become less important than seasonal variations. In this world, will electricity consumers accept paying higher prices in winter than summer, like they do for gas today? 

While we do not pretend to have all the answers to the above questions, we consider that policy-makers and NEM participants will need to consider these types of questions over the coming decade as they grapple with both the challenges and opportunities of the emerging seasonality paradigm.