Planning and adaptive thinking needed to navigate renewable energy network investment

12 April 2021

Portrait of Maxine Frerk

by Maxine Frerk
Director, Grid Edge Policy

Maxine is an experienced regulatory and consumer expert with a particular focus on energy. She has recently left Ofgem after 15 years during which time she had responsibility for a variety of areas including consumer policy, retail markets and smart ...

Renewable Planning Adobe Stock 144687551
One of the challenges facing Ofgem as regulator, in planning for integration of renewable energy sources into our energy supply, is determining how much network investment to allow, given the uncertainty around the pace and direction of energy transition.

Too much, and customers end up paying for network that is not needed. Too little, and renewables and other low carbon technologies can’t connect, jeopardising achievement of net zero carbon emissions targets.

My recent working paper, published as a Visiting Fellow on the Oxford Martin Programme on Integrating Renewable Energy, argues that Ofgem (and industry) need a new way of thinking about the problem.

The first step is to identify the uncertainties that exist, where scenario planning is an established technique. National Grid produces a set of Future Energy Scenarios (FES) each year that underpin many of the decisions taken in the energy sector. These scenarios are hugely valuable but were developed for a particular purpose. As such they represent a relatively narrow range of core scenarios. In contrast the original idea behind scenario planning was to encourage decision makers to think the unthinkable. What really would make this investment a bad idea?

That points to the importance of focussing the scenarios on the particular question at hand. For example, for electricity distribution networks the level of demand from more electric vehicles (EVs) is a crucial driver of investment. The FES scenarios assume that by 2040 the number of EVs is roughly the same under any scenario. That suggests that the worst that happens is that the networks will be investing a bit earlier than necessary. However, if you throw in autonomous vehicles, micro and public transport, together with different approaches to smart EV charging, you could come up with some very different scenarios that the networks will have to cope with.

This takes us to the second challenge, which is deciding between the different options based on how they perform under different scenarios. A lot of emphasis is placed on “no or low regret options”. The danger here is that the risks of underinvesting and failing to meet net zero are not built into the equation and that “low regret” simply becomes “low cost”. A better framing is to talk about decisions that are “robust” in the face of a range of scenarios. There is also an argument for a more iterative approach, looking to improve on options that perform badly in particular scenarios but do well otherwise, rather than simply discarding them. That requires judgment not rigid decision rules.

Finally, this leads to an argument for more use of “real options thinking”. The idea comes from the financial world where you can have an option to buy or sell shares. In the real world you want to be able to defer decisions until more information is available – doing whatever is necessary at this stage to keep the options on the table. A number of academic papers have been written which show that conventional cost-benefit analysis undervalues demand side response - a change in the consumption of electricity or fuel by customers in order to help match demand with the available supply - and other forms of “flexibility” because it ignores the “real option” value they provide. The choice is not between conventional network reinforcement and flexibility but also “flexibility then invest” – moving from a one-off decision to a multi-stage process. Solutions that can adapt to changing circumstances should be preferred to those that are irreversible.

Ofgem has talked about moving to more adaptive regulation. This is the right language but it has to be about more than the timing of decisions. It requires a different way of thinking – about scenarios, about what constitutes a robust decision and about the real option value of flexibility.

This opinion piece reflects the views of the author, and does not necessarily reflect the position of the Oxford Martin School or the University of Oxford. Any errors or omissions are those of the author.