|As the energy trilemma – the need for decarbonisation, security of supply and affordability – looms in the UK, policy-‐makers scramble to compose an energy supply that makes sense. The electricity sector is at the heart of this effort, as it is hoped a growing proportion of supply can be delivered via this potentially low carbon energy carrier in the future.
The cost of electricity has become a pressing concern in recent years for many families, institutions and the government. Uncertainty is a key factor in determining generation costs and consequently consumers’ energy bills. Cost estimation, particularly aspects concerning magnitudes and methodologies, is a frequent topic for publication. However, uncertainty is rarely placed at the focus of these studies.
The Department of Energy and Climate Change (DECC) has produced Levelised Cost of Electricity (LCOE) estimates since 2010. Uncertainty has latterly come to be presented in the results using sensitivities; ‘high’ and ‘low’ figures presented alongside central estimates. This presentation of uncertainty, although simple and clear, is limited in its provision of context. It also fails to provide an overall picture of how costs and uncertainty are varying over time. This work explores a new way of analysing and communicating cost uncertainty, and assesses temporal variation in estimation.
Three analyses are performed using the DECC cost estimates for three electricity generation technologies – nuclear, offshore wind and Carbon Capture and Storage (CCS). Together, these technologies present a spectrum of uncertainty, and promising generation options for the UK. The first analysis composes cost trajectories from selected DECC LCOE estimates and presents them alongside contextual data, such as out-‐turn, historic and projected costs. This analysis produces cost landscapes, which form the foundation for the subsequent two uncertainty analyses. The second, bespoke analysis evaluates temporal estimate uncertainty in the decade 2020-‐2030; an approach aimed at capturing the temporal consistency of estimates, alongside variations in magnitude. The final analysis builds on the second, evaluating the temporal estimate spread in 2025; an approach aimed at partially translating the uncertainty assessment into a more familiar medium.
The results yield a number of interesting observations. Perhaps predictably, nuclear emerges as the cheapest technology with the most consistent set of estimates in temporal terms. The chronology of the estimates preceding the Hinkley Point C agreement and past experience of cost overruns, are the foci of the discussion. The offshore wind estimate trajectories exhibit huge temporal variation and substantial cost premiums. The imminently forecasted reversal of the current trend of decreasing returns to scale is a key discussion topic. The analysis of CCS confirms the unknown nature of the costs associated with the technology. Given its infancy as a concept and resultant unknown-‐unknown nature of the uncertainty, the degree to which it can viably compete with nuclear as a base load supply source is discussed.
This work underlines the need for uncertainty to be brought to the fore in discussions concerning electricity generation costs. DECC is to be lauded as a rare example of a body publishing cost estimates that attempt to quantify and evaluate uncertainty, rather than merely acknowledge it. A consistent approach whereby uncertainty is evaluated with a temporal dimension should be established, perhaps one similar to those developed in this study. Further work is required to refine the methodology and ensure its suitability for application across a wider range of technologies.