Ongoing scrutiny of rising energy bills, coupled with the Government’s proposals for Electricity Market Reform, mean it is crucial that generation costs are predicted accurately, and policy makers understand the strengths and weaknesses of cost estimates and forecasts.
A team led by Dr. Robert Gross of Imperial College London looked at six different technologies – onshore wind, offshore wind, nuclear power, CCS, CCGT and solar PV – to try and find out which factors affect costs over time, and examine the methods that have been used to predict future costs.
They report that a major problem in getting the predictions right has been the tendency for much past research in the field to err on the side of optimism. A commonly held belief that prices would fall over time, as they had in the past, was knocked off course by the escalating cost of fossil fuels, raw materials and building from the mid 2000s onwards, leading to wide disparities between projections and outcomes.
Some of the more contemporary studies were found to take greater account of the variety of factors likely to drive costs in the wrong direction. But others were still anticipating future reductions even where costs and estimates have risen. In part this reflects an expectation that factors such as supply chain constraints will ease and that learning effects, as technologies mature and become more widespread, will lower capital costs and improve efficiency and performance.
Lead author Dr. Robert Gross says: ‘There are different types of uncertainty, and projections need to make that clear. Some are inherently unpredictable and volatile, such as natural disasters, which can overwhelm cost projections in the best of analytical worlds, and others, such as learning effects, ongoing innovations, scale effects and standardisation, lend themselves more readily to future projection. It’s also possible to anticipate and manage factors such as short-term bottlenecks and supply chain constraints, which have implications for policy design such as time horizons and sequencing’.
‘Predicting future generation costs is inherently uncertain, and forecasts often turn out to be wrong. Analysts and policymakers need to better appreciate this uncertainty, and understand that it is not always possible to seek certainty in an uncertain world’.
Presenting the Future: an assessment of future costs estimation methodologies in the electricity generation sectorby Rob Gross, Phil Heptonstall, Philip Greenacre, Chiara Candelise, Felicity Jones, Arturo Castillo Castillo.