Issue Institutionalising ex-post evaluation and periodic recalibration of demand forecasts

Summary

The sector comparison suggests that aggregate forecast error reflects different forces acting on industry, transport and buildings rather than one uniform bias. The author therefore recommends systematically evaluating error by plan round and sector, and whether key technology variables were reflected, at regular intervals after each plan. Findings should feed back into the next plan's input assumptions, additional demand estimates, demand management effects, efficiency accounting and self-supply planning. Demand management in particular needs a stronger method for quantitatively estimating and verifying actual savings rather than merely stating targets. The goal is a virtuous cycle in which demand forecasting repeatedly learns from its errors and recalibrates.