Five minute settlement rule change: What to expect in the energy market

May 30, 2018

Five minute settlement rule change

From 1 July 2021, the National Electricity Market (NEM) in Australia will be settled every five minutes. At present, generators in the NEM are dispatched on a five minute basis. However, these five minute dispatch prices are then averaged (on a time weighted not volume weighted basis) to produce the 30 minute Trading Price, which is then used for settlement. When the NEM commenced in the 1990s, the limitations in metering and data processing at the time led to these different periods for dispatch and settlement.

In May 2016, Sun Metals Corporation, which owns a zinc refinery in Northern Queensland, first initiated the rule change request to move to five minute settlement. Their reasoning was that the mismatch in dispatch and settlement:

  1. Allows generators to game the market, thereby increasing prices for consumers; and
  2. Impedes entry for fast response generation and demand side response.

In this article, we will take a deeper look at these claims.

Electricity prices may not reduce as much as people had hoped

Aligning dispatch and settlement is logical and will make for a more transparent market, but the big question for many in the industry is how will the change affect wholesale electricity prices? Many are hoping that the five minute settlement rule change will be a magic bullet to reduce energy prices. As usual, reality is considerably more complex and nuanced.

This RenewEconomy article illustrates how generators are able to game the market under the current rules. It gives the example of a 30 minute trading interval that contains a price spike of $13,800/MWh in only one of the five minute dispatch intervals. This means that the price averages out to $2,350/MWh over 30 minutes. The important thing to note is that the generator in the example had very low output during the price spike, and only increased output after the fact. The averaging effect means that the generator is able to receive $149,000 for this 30 minute period under the current 30 minute settlement. However, under five minute settlement, the generator would receive only $15,000.

So just how much does this averaging effect distort the market as a whole? As it turns out, not much. We used NemSight, a software developed by Creative Analytics (part of the Energy One group) to examine the difference in settlement values across the entire NEM from 2008 to April 2018. We found that the cost to consumers under a five minute settlement would have been over $100 million higher than compared with 30 minute settlement. In fact, every single year produced a cost rather than a benefit as seen in Figure 1. Even on a state level, every state had a cost in every year.

NEM cost of 5 minute settlement

Although $100 million may seem like a large sum, it is only 0.09% of the total value transacted in the NEM. We consider this to be a negligible impact. However, the real life difference could be much greater. We anticipate that the bidding behaviour will change significantly under five minute settlement. In particular, we expect that most existing generation in the NEM will struggle to respond to five minute price spikes, especially when they are unexpected. This means that it will be largely up to fast responding technology, such as batteries, to regulate prices. Therefore, for consumers to benefit from the change, there will need to be enough fast responding technology entering the market to offset the reduced response from the incumbents.

Furthermore, when it comes to pricing outcomes, there is a much bigger elephant in the room – the high concentration of market power in the NEM. Regardless of whether settlement is calculated on a five or 30 minute basis, certain generators still have sufficient market power to raise prices by restricting output and will be significantly better off by doing so.

In Figure 2, we examined the market share of participants in the NEM by generated energy (rather than capacity) for the 12 month period from 1 May 2017 to 30 April 2018. In NSW, the three major gentailers, AGL, Origin Energy, and Energy Australia control 80% of the state’s generation. AGL with 38% has the largest share. AGL also has the highest market share in Victoria (40%) and South Australia (32%). In Queensland, the state owned generators Stanwell and CS Energy dominate 68% of the market. State owned Hydro Tasmania is essentially a monopoly generator in Tasmania except for the imports that come into the state via the Basslink interconnector.

NEM generation market share

In this context, we believe that consumers will benefit most from the five minute settlement rule change if new entry fast response technology is owned by parties other than the current major players. If the major players end up in control of the new entry, there will be little incentive for these assets to be bid into the market in a way that will ease pressure on prices.

Batteries will be the big winners

We have already touched on the fast acting nature of batteries being well suited for a five minute market. But there is another reason why batteries will be the big winners under the five minute settlement rule change. It has to do with capacity.

Table 1 shows a theoretical 30 minute period. It contains one price spike of $10,000/MWh while the remaining five minute intervals have a price of $100/MWh. Under five minute settlement, a 50 MW battery could discharge 50 MW during the price spike and stay idle in the remaining intervals to earn $41,667 for the half hour. Conversely, the averaging effect under 30 minute settlement means that the battery would need to be discharging at full power for nearly the whole 30 minutes to capture the same revenue. This means that a much smaller (and cheaper) battery could earn a similar revenue. In other words, each MWh of battery capacity will be able to capture more revenue under the five minute settlement rule change. This significantly changes the economic proposition of battery storage in the NEM.

Battery operation under five minute settlement rule change

Demand response will be harder for many industrials

Large energy users have been responding to spot pricing since the commencement of the NEM. Through our own experience, as well as discussions with other industry experts like Paul McArdle of Global Roam, we believe it is likely that there are thousands of megawatts of demand response in the NEM from the large energy user sector.

Interrupting an industrial process, in particular, is inherently more complex than controlling a hot water system or air conditioning. Firstly, most industrials prefer to perform a controlled stop of their process rather than an emergency stop. This is because an emergency stop puts the machinery at a higher risk of damage. Secondly, most industrial processes require a safety check before being interrupted. Because of these reasons, many industrials will struggle to respond in a faster, more volatile five minute market. Industrials will need to think about whether they can make any further improvements with automation and will likely need to revise their curtailment strategies to account for the higher risks involved.

Author: Marija Petkovic, Founder & Managing Director of Energy Synapse
Follow Marija on LinkedIn | Twitter

Subscribe to our mailing list

* indicates required

Jan 2017 records highest ever electricity price in QLD but flexible energy users could pay 37% less

February 1, 2017

Demand response

The average pool price during January 2017 in the QLD wholesale electricity market was $197.65/MWh. This is the highest monthly average since the commencement of the National Electricity Market (NEM). It exceeded the previous record for QLD of $192.46/MWh set in June 2007.

Figure 1 shows the average January pool price for QLD over the last two decades. Since 2013, January pricing has been at an elevated level, creating a challenging time for QLD energy users, particularly those who are exposed to wholesale prices. However, for energy users with flexibility in their operation, this high volatility presents an opportunity to save big and leap ahead of competitors. This can be achieved by engaging in a type of Demand Response, where energy users alter their electricity consumption depending on the price in the wholesale market.

To illustrate the opportunities that were available during January, we will use Facility A and Facility B as an example.

Facility A (Flexible/Price responsive operation)

  • Facility A is a manufacturer that operates 24/7. The entire load is exposed to the QLD pool price.
  • At maximum production output, power usage is 11 MW.
  • At minimum production output, power usage is 9 MW.
  • The production process can also be completely shutdown in 30 minutes. When the Facility is shutdown, it still consumes 0.5 MW.

It would be nice if Facility A had an unlimited ability to respond to every single event in the market. However, this is simply not feasible for most energy users. Facility A’s primary obligation is to supply its customers and hence it needs to manufacture enough product to do so. The flexibility of Facility A is also limited by how much product it can store, as storage can be used to buffer changes in production.

Given these constraints, Energy Synapse utilises in-house predictive modelling of energy markets to develop an operational strategy for Facility A with the goal of minimising the cost of electricity.

Facility B (Does not respond to price)

Facility B is the reference scenario. Facility B is identical in every way to Facility A, except that Facility B does not take wholesale electricity prices into account when developing its operational strategy. Facility B operates continuously at a level that is required to meet product demand. Over the long run, Facility B can be expected to pay the market average price of electricity.

January 2017 Results: Facility A pays 37% less than Facility B ($533k saving)

The strategies employed by Facilities A and B were tested during January 2017. The total pool cost for each Facility is in Figure 2. The flexible price responsive strategy used by Facility A results in a pool cost that is $533k (37%) lower than Facility B. Remember, this saving is just for the single month of January and both Facilities use exactly the same amount of electricity. The difference is all in the timing.

Savings that can be achieved by other facilities will vary depending on their unique operational characteristics as well as the characteristics of the electricity market at the time. Feel free to get in touch with Energy Synapse to discover the opportunities for your sites across the NEM and WA.

Author: Marija Petkovic, Founder & Managing Director of Energy Synapse
Follow Marija on LinkedIn | Twitter