The ups and downs of system prices13 August 2020
Electricity ‘system prices’ are used to price imbalance volumes for participants in the GB power market. Usually they are between £0/MWh and £100/MWh but occasionally can exceed £2000/MWh (eg, £2242/MWh on 4 March 2020) and go as low as -£150/MWh (with significantly more instances of negative system prices being recorded in 2020 than in previous years). What lies behind the peaks and troughs and what are the implications?
Elexon manages the balancing and settlement code (one of 11 major codes setting out the rules that underpin Great Britain’s gas and electricity wholesale and retail markets).
We calculate the difference between how much electricity generators and suppliers said they would produce or consume with what actually happens each day (an energy imbalance volume). We work out a price (system price) for these differences and ensure that generators and suppliers, pay (or are paid) the right amounts of money. This is known as settlement. We explain how settlement works and how the electricity system is balanced in chapters one and two of our new guidebook.
In the GB electricity market, participants buy and sell electricity for delivery in each 30 minute settlement period. Therefore, there are 48 settlement periods every 24 hours. Electricity can be bought years in advance, up to the start of each settlement period.
Part of our role is to calculate the system price every 30 minutes. We use system prices to settle energy imbalance volumes for electricity market participants. System prices reflect the price of electricity used by National Grid ESO to keep the system in balance. National Grid ESO does this by paying generators and suppliers to increase or decrease electricity generation and demand. National Grid ESO then sends the information about what actions they took and for what price to Elexon’s systems. Elexon takes this information and calculates the system price using an agreed calculation.
We have calculated system prices since 2001, however the way they are calculated and how they are applied to imbalance volumes has been modified 25 times since their introduction. These modifications have changed the calculation to support the changing needs of GB’s electricity market and system. For example, some modifications have sharpened price signals for imbalances, others have simplified settlement for new entrants.
System prices give a financial signal to electricity market participants to reduce their energy imbalance volume, or to have energy imbalance volumes that are helpful to the system.
For example, if you were a generator you might agree to sell 30 MWh of electricity to a supplier in advance. If you ended up generating 29 MWh, you would have a deficit imbalance volume of 1 MWh and be charged the system price for that 1 MWh.
On the supplier side, if the supplier bought 30 MWh to cover the forecasted electricity demand of its customers and they actually ended up using 25 MWh, then you would have a surplus energy imbalance of 5 MWh. In this example, the supplier would then be paid for the surplus energy at the system price.
Electricity traders can also buy and sell energy to deliberately incur a surplus or deficit imbalance volume. This would be done if they think they will make a profit from the system price compared to market prices (ie, the wholesale price of electricity).
High system prices
Higher system prices occur when there is a net deficit of electricity, ie, when there is more demand for energy than there is generation. National Grid ESO balances the system by asking generators to increase output or asking consumers to reduce demand. The costs of balancing the system (for example paying generators to produce more electricity, or paying users to trim their demand) are then reflected in the system price.
High system prices lead to high energy imbalance cash-flow for market participants. This cash-flow could either be a large charge if the participant had a deficit of energy, or a large payment to the participant, if they had an excess of energy.
Let’s look at an example from earlier this year (Graph 1). In the early evening of 4 March 2020, at 18:00, the system price spiked and reached £2242/MWh (the highest since 2001).
Graph 1. GB system prices, 4 March 2020 (source: Elexon)
Looking at the system prices earlier in the day, they had remained between £0/MWh and £72/MWh up to 16:00. The system price then increased between £119/MWh and £144/ MWh before going up to £2242/MWh at 18:00. In this instance, the system price calculation reacted to a perceived scarcity of energy reflected in the ‘reserve scarcity price’.
The reserve scarcity price is equal to the loss of load probability multiplied by the value of lost load, currently set at £6000/MWh.
The loss of load probability is calculated by National Grid ESO and reflects the probability that customers will have their electricity cut off. The value of lost load is a price used to represent the cost to electricity consumers of having their electricity cut off.
National Grid ESO forecasts the loss of load probability for a settlement period. We then publish these forecasts on our Balancing Mechanism Reporting Service (www. bmreports.com) as an information tool for industry. Using the loss of load probability calculated one hour ahead of the settlement period, we calculate a reserve scarcity price to be fed into the system price calculation.
On 4 March 2020, between 18:00 and 18:30, the loss of load probability was 0.3705 (37.05%). It may have been this high due to a low forecast of wind generation availability, combined with a higher forecast for electricity demand.
This meant that the reserve scarcity price was calculated to be £2223/MWh (£6000/ MWh multiplied by 0.3705). There was also an adjustment of £19/MWh added to get to the final system price. This adjustment was to account for the additional cost of warming a power plant to provide balancing energy.
There have only been nine other occasions between 2002 and 4 March 2020 where system prices have been higher than £1000/ MWh. All of these high prices were in November 2016 and May 2017. The November 2016 high prices were caused by French nuclear outages. The May 2017 high prices were due to low wind and solar generation.
High system prices encourage a market reaction, eg, a supplier needing to ensure it has contracted to buy more electricity, or a generator generating more electricity. This demonstrates the role that high prices can play in helping to keep the system in balance.
Negative system prices
When there is a net surplus of energy (more generation than demand) we can get negative system prices. A negative system price flips the cash-flow direction. Using the same examples from before, if a generator sold 30 MWh and generated 29 MWh, it would be paid for its 1 MWh deficit of energy. If an electricity supplier bought 30 MWh of electricity and their customers only used 25 MWh then they would have to pay for the 5 MWh surplus of energy.
There have been more instances of negative system prices between 1 January and 31 May 2020 than in any other year. The 175 instances of negative system prices in 2020 represent 2.4% of the settlement periods between 1 January and 31 May 2020, while over the period 2014-2019 negative prices occurred 0.5% of the time (Graph 2).
Graph 2. Percentage of settlement periods with negative prices (source: Elexon)
Negative system prices tend to occur when demand is low and a high proportion of generation is from weather dependent renewables.
Over the period 1 January to 31 May 2020, 32% of GB electricity was generated by solar and wind. For the 2.4% of settlement periods when the system price was negative, 55% of electricity was generated by solar and wind.
During April and May, COVID-19 lockdown restrictions have meant that demand for electricity was 13% lower in 2020 than in 2019. There has also been a higher proportion of electricity output from renewables, due to weather conditions over that period. This has led to negative system prices for 5% of the settlement periods during these months.
Very high system prices tend to occur in short spikes (as on 4 March 2020), while negative system prices can last for longer.
The longest stretch of negative GB system prices was for 11 straight hours over the night and morning of 7 December and 8
Graph 3. System prices on 7 and 8 December 2019 (source: Elexon)
December 2019 (see Graph 3). On this night there was an excess of wind generation in Scotland (when there is more wind generation than demand for electricity, National Grid ESO pays wind generators not to generate electricity in order to keep the system in balance). The cost of paying generators to curtail their wind generation during this period is reflected in the system price.
While system prices are negative the market is encouraged to reduce electricity generation and increase electricity demand. This helps to keep the electricity system in balance.
Elexon’s role is to administer and support the electricity market and the operation of the electricity system in Great Britain. These rare instances of high and low system prices support the operation of the electricity system by providing a price signal to electricity market participants.
Market participants that can forecast what the needs of the electricity system will be can make money by having a surplus of energy when the rest of the market has a deficit, or a deficit of energy when the rest of the market has a surplus. This opportunity is also why traders operate in the electricity market.
The supply/demand positions of the system can be difficult to predict, given the varying factors that influence supply and demand of electricity.
We make large amounts of data and training materials available to electricity market participants, so that they understand system prices and the implications of having energy imbalance volumes.
Clearly, increases in instances of negative prices are to be expected as Britain moves to a lower carbon electricity system, with increased amounts of renewable generation and declining output from some traditional power generation sources (especially coal).
Greater amounts of renewable generation make National Grid ESO’s job in balancing the system more challenging. However, in future, if more battery storage can be deployed, excess renewable generation could be stored when it is not needed, rather than being curtailed.
This electricity will then be available for re-use at times of higher demand and allow generators to release the electricity from storage at an optimum price. In addition, this could also provide extra levers for National Grid to use when managing the system instead of calling for more output by generators when there is a shortage of generation. Equally, the ability to store renewable generation, rather than stopping production at times of low demand could provide a significant contribution from the electricity sector to net-zero.
Author: Emma Tribe, market advisor, Elexon, UK