Modelling the Value of Reliability for Residential Customers

Kenneth H. Tiedemann

Keywords

Reliability, Willingness to pay, Electricity

Abstract

Beginning in the early-1990s, electric utilities, regulators and governments in Canada and the United States became increasing interested in the security of electricity supply. From a cost-benefit perspective, investments in electricity system reliability should be undertaken up to that point where the marginal cost of reliability equals the marginal benefit of reliability, so that estimating the value of reliability is a key parameter in electricity system planning and operations. In this study, survey data is used to estimate residential customer willingness to pay for increased reliability in British Columbia, Canada and to model the determinants of the value of reliability using White’s heteroscedasticity-consistent estimator. Key findings are as follows. First, costs per outage vary by length of the outage, time of day and season of the year. Average costs vary from $2.75 for a 20 minute outage at 4pm in the winter to $19.60 for an 8 hour outage at 8am in the summer. Second, regression models successfully explain 84% of the variation in costs per outage. Third, costs per kWh lost also vary by length of the outage, time of day and season of the year. Average kWh costs vary from $0.73 for an 8 hr outage at 4pm in the summer to $31.75 for a 20 minute outage at 8am in the winter. Fourth, multivariate regression models successfully explain 66% of the variation in cost per lost kilowatt hour.

Important Links:



Go Back