The NEM - Part 3: The wholesale market design
A series explaining Australia's National Electricity Market
This is the third post in a series of explainers covering the NEM. Last week we looked at how the wholesale market works, and this week we’re looking at some of the key design decisions in the NEM. If you haven’t read Part 2, you might find it useful for understanding some of the discussion this week.
In particular, we’ll look at two topics:
Why does the wholesale electricity market have a single marginal price?
The energy-only market vs. a capacity market debate.
These topics are a bit harder to cover concisely and are probably another level of detail deeper than Part 2. However, I hope you still find them helpful for putting the wholesale market in context, particularly during these turbulent regulatory times.
In part 4, we’ll focus on the role of the demand side.
A marginal clearing price
In the NEM, there is a single marginal clearing price in each state.1 All large generators2 in that state earn this clearing price when they’re dispatched. It’s called a marginal price because it’s set by the generator producing the marginal unit of electricity (i.e. the generator that would produce the next bit of electricity if we needed a tiny bit more).
Why do we use a marginal price for all generators? Sometimes, it can seem counter-intuitive or even wrong. For example, if the wholesale electricity price reaches $16,000/Megawatt-hour (MWh), more than 100 times greater than average prices, why should a solar generator earn this price when they produce electricity for free?
There are other models that could be used in the NEM. For example, a pay-as-bid style market, where dispatched generators are paid the price they offer instead of the marginal price.3
To answer the question of why we have a marginal price paid to all generators, you need to start with an understanding the purpose of the wholesale market.
The point of a wholesale market
The wholesale market is a focal point of the NEM. It is designed to simultaneously achieve multiple related aims:
Provide long term benefits to consumers: The people want electricity - the wholesale market is about efficiently giving it to them. This means giving consumers enough electricity supply now and into the future, at a price they’re willing to pay.4
Coordinate supply and demand: There are multiple different generators and energy users spread across the NEM. As we discussed in Part 2, the wholesale market is responsible for coordinating the cheapest combination of supply and demand every five minutes of every day.
Encourage investment: The wholesale market also needs to signal to investors, developers and all other electricity market participants when and how to invest. The market needs to reward astute investors who make good investment decisions, and discourage poor investments. For example, if the electricity market needs more storage to keep the system reliable, the price signals in the wholesale market should communicate this.
So why use a marginal price?
To have a market design that achieves what the wholesale market is meant to achieve is nothing to sneeze at. The marginal pricing system has multiple advantages:
It’s simple and transparent. Every five minutes there is a single price (per region). All generators are paid this price and all retailers pay this price. This pricing mechanism is (relatively) easy to understand and is observable to all interested parties – this is important for a multitude of reasons.
It helps with forecasting electricity prices. Forecast prices are needed to plan future investments, help electricity retailers offer pricing to customers and help all other manner of long-term decision making.
It underpins a more complex contract market based on the electricity price, which will be covered later in this series.
It’s a clear price signal. The wholesale price tells us the value of electricity every five minutes. This is happening in the wholesale market now - looking at the prices will tell you that electricity is abundant in the middle of the day and in greater demand on hot afternoons. These prices will reward technologies that can store cheap electricity and release it later in the day (e.g. batteries).
Back to our solar example - if the price goes to $16,000/MWh, why should a solar generator who’s producing this electricity basically for free get this price? Because it’s valuable! When the price goes to $16,000/MWh, electricity is scare and we don’t really care where we get it from as long as we get it. This very high price for the solar farm communicates a few things: 1) Please please please solar farm, keep generating; 2) You energy user, do you really want to use electricity right now? How about turning things down for a few minutes…; and 3) Hey investor, would you like to build a generator that’s available during these events, because you’ll be well paid!
A single price communicates all of these important things.
It works for both sides of the market. The single marginal price works for both supply and demand. It encourages generators to compete and it allows electricity users to dip out when prices are higher than what they’re willing to pay. For example, some large electricity users will cease using electricity when the wholesale price passes a threshold. This is a good outcome! Instead of producing more electricity from expensive generators, the system clears at a lower level of demand. We’ve talked a little bit about this previously and in a future part we’ll talk more more about how electricity users interface with the NEM.
On top of the reasons why it’s a good system, the marginal price system also has the benefit of inertia. Its been around for a long time and the whole system revolves around it. If someone were to propose changing it, even if they could come up with a compelling alternative, it would be a complex, expensive and drawn-out process.
In summary, the pricing used in the wholesale market is what the people who designed the NEM thought would best achieve their objectives: efficiently coordinate supply and demand, support investment and ultimately deliver long-term benefits to consumers.
Energy-only vs. capacity markets
In electricity markets, we want people to invest money to build more generators when they’re needed. This means, investors need to think they’ll make their money back and then some before investing.
How to best design electricity markets to do this is the subject of hotly contested debates. There are two main schools of thought: energy-only and capacity markets. Most electricity markets throughout the world that allow private investment fall into one of these schools.
And while the NEM is an energy-only market, it’s not immune to suggestions that we need to shift to a capacity market. There are regular throwdowns on whether the energy-only market works,5 whether we need to a capacity market or whether we end up somewhere in the middle…
What investment signals do we need?
There’s two buckets of money that a generator in the NEM would need to be confident they can recover in order to decide to invest. There are capital costs - building the generator, buying land, etc; and there are operational costs - fuel, maintenance, staff etc. To invest, you would need to be confident you can recover these costs and make a profit.
Energy-only markets
In an energy-only market, as implied by the name, generators are paid only for the electricity they produce (and consumers pay only for the energy they consume).6 For each five minute dispatch interval (as covered in Part 2), each generator will be paid for their generation at the wholesale electricity price.
When wholesale electricity prices in an energy-only market are at “normal” levels, generators are not normally recovering enough revenue to cover their operating and upfront costs.7 If prices stayed at this level at all times, we would have problems - there wouldn’t be enough revenue available for generators to invest when we need them to.
So generators can make enough money, wholesale prices in energy-only markets can climb to very high-levels (while a normal wholesale electricity price might be $50-150/MWh, prices can jump to $16,600/MWh). The wholesale price will typically jump to these levels when there is very little spare supply. This might happen when demand is very high or if there have been generation outages. The potential for prices to climb to very high levels (which can be referred to as volatility) is critical – repeated and sustained high prices are signals from the market that there is a requirement for additional generation.
In fact there’s an argument within energy-only designs around how high the price should be able to climb – most markets have a maximum price cap. In the NEM this is currently $16,600/MWh which is very high by international standards.
Perhaps counterintuitively, a higher price cap allows the system price to be lower more of the time, with short periods of high volatility allowing investors to earn back their investment in storage or peaking generation.
Within energy-only markets there can still be more than one market price – for example ERCOT a day-ahead market and the real-time market. Both of these markets serve different functions, but they value only energy.
Capacity markets
Capacity markets on the other hand introduce a new price into the market design – a price based not on the energy generated, but on the potential energy generated – i.e. the size or capacity of the generator.
A capacity price explicitly pays for the capacity costs of values new generation, and provides a clear signal for future investment. Depending on the design, a capacity price can also send other signals, like what type of generation (or where) by having modified prices.
The typical mechanism for a capacity market is a capacity auction – a reverse auction run a number of years in advance of delivery, where existing generators and potential investors, can bid.
Of the competitive energy markets around the world, this type of market is the dominant design; there’s only about half a dozen energy-only markets internationally.8
There is still an electricity price in capacity markets. Generators will get paid for each the electricity that produce - the key difference is that these prices are not allowed to go as high as they are in an energy-only market.
What’s better?
Both market designs have pros and cons, and how you think about these pros and cons depends on your framing.
Efficient (winner: energy-only): energy-only markets tend to leave the decisions on how and when to invest in the electricity market to market participants. In a capacity market, a central planner has a greater role in determining how much generation we need. From an efficiency perspective, the energy-only market puts more of the investment risks onto generators which is better than a single centralised operator (who will tend to be risk-adverse) deciding how much capacity is needed.
Certainty (winner: capacity market): The biggest selling point for capacity markets is certainty. A central planner will pay for capacity several years into the future, which gives comfort to governments and regulators. Energy-only markets are much more dynamic, which might excite economists but doesn’t necessarily do it for our political leaders!
The evolving market (winner: energy-only): the electricity market is going to look very different in 10 years compared to 10 years ago. New technologies, distributed energy resources and more decentralisation point to a very different power system. We will need a market design that accommodates this new system. For example, soon we’ll need a market design that appropriately rewards people with EVs to plug in at home and send electricity into the grid. This is incredibly difficult for a central planner to properly value in a capacity market. In an energy-only market, they just need to respond to the electricity price.
It’s a generalisation, but I think it’s fair to say the pro-energy-only market camp is the “faith-in-markets”/decentralisation side and the capacity market camp is the “faith-in-central planning.”
Something else to keep in mind - no matter what market design is in place, it is a gargantuan effort to change it. So even if you ardently believe that energy-only is the better design, shifting from a capacity market to energy-only will disrupt all of the established processes for building generation, paying for power and much more.
Where is the NEM headed?
There seems to be a “grass is greener” vibe to market design. Those regulating energy-only markets (including the NEM) constantly bemoan the uncertainty of the market and the need to bolster it through capacity payments. At the same time, capacity markets are grappling with a market design that works for big, chunky power plants, not household batteries.
In the NEM, we’re headed for a halfway house. The energy-only market design put up a valiant fight, but governments are pushing for a faster transition of the power system. While this might have been achieved through a carbon tax, this was also too politically difficult.
It looks like we’ll end up with an energy-only market overlaid with significant government underwriting of new capacity.
Teaser for Part 4 - The role of the demand side of the market
If you’ve been wondering throughout these posts why there’s so much focus on the supply side of the market (i.e. generating electricity), you’re spot on. We don’t talk about it nearly enough!
Part of the reason is that its more complicated. Big machines sending power through poles and wires to electricity users is how the power system has always worked. When trying to simply explain how the NEM works, its easier to focus on the supply side because there are fewer moving pieces. But, it is a very incomplete way to think about the NEM.
In our next explainer article, we’ll dig into the role of the demand side and why its going to be increasingly important.
Please share this post if you’ve found it helpful!
Alex and I write this blog in our spare time and have really enjoyed the feedback we’ve had from readers. To date, most of our growth has come from people sharing posts they enjoyed.
In the NEM we have “regions”, and each region gets its own price every five minutes. These regions are the states with the exception of the ACT, which is included in NSW.
For brevity, generator will refer to large generators. A future post will be dedicated to looking at the differences between these large generators and the smaller generators like the PV that might be on your roof.
This came up in European electricity markets in 2022, when gas prices reached extremely high levels, pushing up electricity prices with it.
The notable commentary on electricity market design included European Commission president Ursula von der Leyen’s 2022 State of the Union address including:
“The current electricity market design – based on merit order – is not doing justice to consumers anymore. They should reap the benefits of low-cost renewables. So, we have to decouple the dominant influence of gas on the price of electricity.”
Whether you believe it does this, or how well it does this, is a debate for another time.
There are other revenue streams available like helping to keep the system stable. These revenue streams have their own markets which will be covered throughout this series.
This is sometimes referred to as the “missing money” problem.
The NEM, the NZEM in New Zealand, the NEMS in Singapore, ERCOT in Texas, CAISO in California and AESO in Alberta, Canada. AESO looked carefully at introducing a capacity mechanism in 2021 but the decision was scrapped by the government before the regulator was able to hand down a decision. Did we miss any?
Great read Declan, your explanation of some of the more complex nuances of the NEM are clear and simples (best meerkat voice). Making this information far easier to consume for people less actively involved in the energy sector.
Thanks for the great article.
Japan has a wholesale energy market.
South Korea, China, Vietnam, India and Philippines also have markets at slightly higher levels of regulation, and not sure if they are energy only markets.