As a self-confessed imbalance pricing geek, one of my first questions as I stood stranded in St Pancras station on Friday evening 9 August, was: what was the imbalance price? And was it £6,000/MWh? Spoiler: it wasn’t, and it shouldn’t have been – but we might see imbalance prices of up to £6,000/MWh if demand control is needed to balance the system again.
My first role at Ofgem was looking at the imbalance prices as the mechanism for security of supply in the energy market. The story at the time was that cash-out prices should drive wholesale prices, and therefore incentives to invest in the right amount of capacity to meet the country’s demand. As part of this, the regulator decided in 2014, any demand disconnections and voltage reductions (i.e. blackouts and brown-outs) should be included as ‘balancing actions’ when we calculate the imbalance price, with a price of £6,000\MWh.
However, on Friday, despite the automatic load shedding of 941MWh of load at 15.54, the Balancing Mechanism Reports website showed system prices of between £64.50/MWh and £64.75/MWh in the settlement periods when demand was disconnected. These prices were revised upwards five working days later to account for the disconnected volumes (and other balancing actions taken outside of the Balancing Mechanism) but the impact was only a £0.25/MWh increase in the second two settlement periods.
The why and wherefore of VoLL?
Before the Capacity Market at least we once had an energy-only market, where wholesale prices were supposed to rise to reflect scarcity rent (the long and short-run costs of marginal generation being recovered) when margins were tight. This was the original vision of the New Electricity Trading Arrangements (NETA), but of course, this vision was ‘upgraded’ with the advent of Electricity Market Reform (EMR).
In an energy-only market, prices should be allowed to rise to the level that consumers are willing to pay for it or to put it another way the price of disconnections – the value of lost load (VoLL). This price should send the most efficient signal for security of supply in the energy market.
Even with the Capacity Market in play, the pricing of scarcity into wholesale prices has been a focus of the regulator, particularly as a means of encouraging parties to ensure they are in balance with the system. While I was at Ofgem we conducted a joint study with the former Department of Energy and Climate Change to determine the ‘correct’ level of VoLL. Unsurprisingly the answer was: ‘it depends’ with a range of values depending on the customer type, time of day, method of collection, how you asked the question etc.
A weighting of domestic and customers winter peak VoLLs resulted in a figure of £17,000/MWh, which was deemed to be the ‘true VoLL’. It was also deemed a “bit scary”, so an “administrative” VoLL of £6,000/MWh was chosen. It was higher than the VoLLs of most industrial and commercial customers surveyed (so as not to discourage Demand Side Response) and also higher than price caps in neighboring countries (so as to encourage interconnector flows when we needed them most).
How do we price in VoLL?
As part of its review of imbalance prices, Ofgem concluded in 2014 that not reflecting demand control actions in the imbalance price had been perversely dampening prices when the system was under stress. This – along with other factors such as inefficient reflection of reserve costs, and unnecessary averaging in the price calculation – was leading to a weak imbalance price, and therefore weak incentives to invest in flexible capacity. And this was part of the reason why there was a missing money problem that eventually needed to be solved by the Capacity Market.
And so, its changes to make imbalance prices ‘more marginal’ were introduced on 5 November 2015 (remember, remember) through the Balancing & Settlement Code (BSC) modification P305, and with it, pricing of demand control actions at administrative VoLL of £6,000/MWh (introduced in two phases starting at £3,000/MWh). In theory, this change should have seen the imbalance price rise to £6,000/MWh whenever a demand control action is needed to balance the system.
So why wasn’t the price VoLL?
From an imbalance pricing perspective, not all balancing actions are treated equally. Only those actions that resulted from ‘true’ energy imbalances are reflected in the imbalance price signal. Actions that were taken for other ‘system’ reasons (such as constraints on the transmission network) do not make up the price. Low-Frequency Demand Disconnections (LFDD), the automatic shedding of load should the frequency levels hit certain trigger level, are not reflected in the imbalance price at VoLL for this reason.
However, disconnections are effectively a balancing action, and so should be reflected in the overall measure of imbalance on the system, otherwise known as the net imbalance volume, or NIV. They are therefore included in the volume stack, but are ‘SO flagged’, meaning they are effectively un-priced.
Reflecting the volume of the disconnection action, but not as a driver of price, has an impact on prices by shifting the ‘merit order’ of balancing actions, so the price-setting action is a more expensive one. In the case of Friday 9 August, the merit order impact on price was only £0.25/MWh.
The existential question
In practice, whilst imbalance prices encourage trading to avoid costs and achieve system balancing, being able to benefit from unpredictable high system prices at times of scarcity is not, and has never really been the primary incentive to invest in capacity in GB.
While some parties have benefited from ‘NIV chasing’ to capture high imbalance prices, in practice this is just one (falling) revenue stream and is unlikely to grow in importance in the longer term. Wider access to the Balancing Mechanism is likely to drive more business models designed around playing in that market, in combination with a range of other revenue sources, to jump between the capacity market, wholesale markets and balancing services. At the end of the day, the extrinsic drivers of timing, frequency and scale of imbalance revenue streams don’t sit well with more risk-averse long-term infrastructure investors.
It is now a complex combination of signals – policy instruments, avoided costs, and system services, some national and some local – which developers are trying to build investment cases around. The success or failure of their attempts to do so will ultimately have a bearing on system reliability.
As the system becomes more complex to manage, the imbalance price calculation (and all the work we do to make sure it’s efficient) will no doubt remain an important component to achieving system balance through incentivising energy trading. But like the broader wholesale markets, it will probably need to adapt and changed to reflect a power market that is now anything but “energy-only”, and increasingly more local and distributed than national and centralised.