The Economic Value of Temperature Forecasts in Electricity Generation

Every day, the U.S. electricity-generating industry decides how to meet the electricity demand anticipated over the next 24 h. Various generating units are available to meet the demand, and each unit may have its own production lead time, start-up cost, and production cost. Total costs can be minimized if electricity demand is accurately forecast. Accurate demand forecasts, in turn, depend on accurate temperature forecasts. This paper estimates the cost savings (i.e., benefits) attributable to temperature forecasts used by the U.S. electricity-generation industry. It does this by establishing the relationship between the quality of temperature forecasts and the quality of electricity demand forecasts at six sites around the United States. It then draws on earlier work by Hobbs et al. on the relationship between the quality of demand forecasts and production costs to estimate the percentage of cost savings from different temperature forecasts. Finally, these cost savings are extrapolated to estimate the total benefits, and incremental benefits, for the United States as a whole. The total benefits of U.S. National Weather Service (NWS) forecasts are estimated to be $166 million. The additional benefits potentially obtainable from a perfect temperature forecast are $75 million per year. It is estimated that an incremental 1% improvement in the forecast quality (from the current NWS forecast) would be worth an additional $1.4 million per year. These numbers do not include other possible benefits of forecasts to the electricity industry, such as those from the improved scheduling of plant maintenance.