Inputs: storage

The FES Consumer Transformation scenario will put 23GW of storage on the system by 2050, but we think that's not enough.

The latest CM auctions (a good way of assessing how many projects are at a stage where the developers are confident they'll get built as non-delivery penalties are high) show 12GW of battery storage by 2027.

Battery storage is the only non-subsidized, commercially proven, scaled technology that can manage renewable intermittency, grid constraints, balancing, etc. We think only doubling this from 2027 to 2050 is a huge underestimation of the build-out, and we believe 50GW is closer to the mark.

In addition, when we run the model with 23GW of storage (and the FES capacities 'as are'), we get significant hours (around 20 hours per year) of lost load, particularly post-2035. This leads to revenues for storage of upwards of £300k/MW/year (or more). If the grid is that constrained, and storage revenues are (as a result) very high, more storage will get built - so the buildout figure is higher.

July 11th edit: The new 2023 FES figures put the buildout of storage at 49GW by 2050, including 35GW of battery storage, so we're still 15GW ahead.

The duration of storage is fit to previous behaviors. Longer duration assets will be built to deal with longer periods of low or high prices driven by renewable intermittency.

Pumped storage capacity will grow

The FES says that an additional 3GW of pumped storage will come online in the early 2030s. This is largely driven by a significant new project (Coire Glas) in Scotland and the conversion of smaller hydro sites to pumped hydro.

Other storage will come online

Mid-duration (4 hours +) will come online in the late 2020s. The 2022 FES scenario does not have any long-duration storage in it.