When one or two energy geeks meet it is rare that “security of supply” isn’t mentioned. “Thou shalt have security of supply” is one of the commandments of the energy system – keep those lights on!
But like most commandments, energy security means different things to different people and is employed to justify all sorts of interventions, and actions. It is also very difficult to critique a core tennet of the energy system unless you welcome the wrath of the priests that will surely follow when you challenge its centrality in policy making.
I am not in the business of questioning interventions that protect the security of our system. Nor would I promote any policy that could possibly impact keeping those all-important lights on. However, I would suggest that we have become very sloppy in how we use this emotive term.
As the energy sector’s catchphrase for everything from a mildly inconvenient service slip through to a total system failure, it is time that we examine what sort of security we are looking for – and how best to plug that risk. This means unpicking the term and analysing much more precisely the different levels of risk and costs sitting beneath it.
It would be good to start with distinguishing between “service” risk and “system” risk. These are distinctly different things and form the book ends of the catch-all phrase.
Practically all of the power cuts experienced by consumers today are the result of faults, or planned maintenance, on the low voltage distribution systems. The debate over security of supply, however, is dominated by questions of generation.
Policy makers believe that the central energy system must deliver a 100 per cent energy service at any time and to all consumers.
Government’s primary tool for securing capacity and delivering this service security is the capacity market, and though this does offer opportunities to other services, it is primarily focused on providing centralised generation capacity. To a much lesser extent, does policy take into account energy supply that is not on the system. Nor does it anticipate the technological interventions that could help us deliver that 100 per cent service without drawing all of this from centralised generation.
With distributed energy, grid storage, prosumers, industrial demand and storage beyond the plug, new options for responding to service shortfalls are increasingly available and should be encouraged. Actually, during the year we are generally oversupplied with generation and undersupplied with flexibility services.
So, as a consumer it should be possible for me to contract with a supplier for 80 per cent system service, while using my electric car, my local authority storage capacity or my in home battery for the additional 20 per cent.
Current calculations are that we need back up for 2 hours at a time, totalling 5GW across the system – it is not at all insurmountable for much more of this to be covered by new services. This would mean that we can take the certainty of a 100 per cent system service out of the security of supply paradigm and redefine it as a shortfall delivered from the system with other top up options available.
It is not that 100 per cent service isn’t important but, we must adapt to recognise it can be delivered via accessing non-system interventions.
In addition, we should ask if we are calculating the most important risk to the security of the system accurately or appropriately.
Currently this is calculated with very superficial data. While this will obviously improve with rollout of smart meters, the “SoS” risk is still predicted using an old fashioned and data poor metric – derated reserve margins.
As Edmund Reid, from Lazarus points out in his excellent security of supply paper, reserve margin is by definition subjective, and is “not telling us that there is insufficient generation capacity in general; it is telling us that in a scenario where it is dark, still, dry and cold there may be insufficient generation capacity.”
In fact, wrote Reid “during 2015/16’s supposedly ‘tight’ winter, when a number of commentators claimed a high risk of supply interruption, we have had 37 negative balancing price periods”. In short, the power system was so long on power that winter that it needed to constrain it off 37 times more often that it had to ask for more.
Even the IEA said that the reserve margin metric “is not well suited to taking into account the capacity of variable renewable energy”.
We compound this problem by having supply standards that are amongst the most demanding in Europe. Some might say that with this level of margin we will never demand greater productivity gains from the system and we will never allow the market to stimulate some really great efficiency gains. Some other national regulators believe that countries across Europe have become captured by a capacity paradigm that more closely reflects a fear policy rather than a real need.
It is time for an encyclical from ministers. Security of supply as a term should be banned. Instead, we should design polices that distinguish between service risk and system risk, whose different characteristics and needs, are conveniently disguised by our sector’s favourite catch-phrase. We will then be able to identify the most cost effective and system efficient measurement and responses to where risk really lies.