Response of Residential Electricity Demand to Price: The Effect of Measurement Error

In this paper we present an empirical analysis of the residential demand for electricity using annual aggregate data at the state level for 48 US states from 1995 to 2007. We estimate a dynamic partial adjustment model using the Kiviet corrected LSDV (1995) and the Blundell-Bond (1998) estimators. In addition to the lagged dependent variable, our equation includes energy prices, income, cooling and heating degree days, and average household size. We find that the short-run own price elasticity of consumption is similar across LDSV, bias-corrected LSDV and the variant of the Blundell-Bond where we instrument for price. The short-run elasticity is the lowest when we use the Blundell-Bond GMM approach that treats the price of electricity as exogenous. The long-term elasticities produced by the Blundell-Bond system GMM methods are largest, and that from the bias-corrected LDSV is greater than that from the conventional LSDV. From an energy policy point of view, the results obtained using the Blundell-Bond estimator where we instrument for price imply that there is room, in an electricity system mainly based on coal and gas power plants, for discouraging residential electricity consumption and curbing greenhouse gas emissions by imposing a carbon tax.