This paper by Jeffrey A. Dubin and Daniel L. McFadden, published in *Econometrica* in 1984, focuses on the econometric analysis of residential electric appliance holdings and consumption. The authors aim to test the statistical exogeneity of appliance dummy variables in demand equations for electricity, as ignoring this exogeneity can lead to biased and inconsistent estimates of price and income elasticities. They use a subsample of the 1975 survey of 3249 households conducted by the Washington Center for Metropolitan Studies (WCMS) for the Federal Energy Administration. The paper introduces a unified model of the demand for consumer durables and the derived demand for electricity, emphasizing the importance of considering the relationship between durable purchases and their use. The authors derive and estimate a joint water-heat space-heat choice model to determine the magnitude of bias resulting from estimating a unit electricity consumption (UEC) equation by ordinary least squares when unobserved factors influence both appliance choice and usage intensity. Key findings include: - The estimated discount rate declines with income, suggesting that lower-income households are more credit-constrained. - The price elasticity of demand for electricity is negative, indicating that higher electricity prices lead to reduced consumption. - The income elasticity of demand for electricity is smaller than previously estimated, and the cross-price elasticity with the price of gas is larger when estimated using consistent methods. The authors also perform specification tests to confirm the independence of choice variables and the error term, rejecting the null hypothesis that unobserved factors influencing appliance choice are independent of those influencing usage intensity. This suggests that estimating UEC by least squares can introduce significant bias, and appropriate statistical methods, such as instrumental variables, are necessary to correct this bias.This paper by Jeffrey A. Dubin and Daniel L. McFadden, published in *Econometrica* in 1984, focuses on the econometric analysis of residential electric appliance holdings and consumption. The authors aim to test the statistical exogeneity of appliance dummy variables in demand equations for electricity, as ignoring this exogeneity can lead to biased and inconsistent estimates of price and income elasticities. They use a subsample of the 1975 survey of 3249 households conducted by the Washington Center for Metropolitan Studies (WCMS) for the Federal Energy Administration. The paper introduces a unified model of the demand for consumer durables and the derived demand for electricity, emphasizing the importance of considering the relationship between durable purchases and their use. The authors derive and estimate a joint water-heat space-heat choice model to determine the magnitude of bias resulting from estimating a unit electricity consumption (UEC) equation by ordinary least squares when unobserved factors influence both appliance choice and usage intensity. Key findings include: - The estimated discount rate declines with income, suggesting that lower-income households are more credit-constrained. - The price elasticity of demand for electricity is negative, indicating that higher electricity prices lead to reduced consumption. - The income elasticity of demand for electricity is smaller than previously estimated, and the cross-price elasticity with the price of gas is larger when estimated using consistent methods. The authors also perform specification tests to confirm the independence of choice variables and the error term, rejecting the null hypothesis that unobserved factors influencing appliance choice are independent of those influencing usage intensity. This suggests that estimating UEC by least squares can introduce significant bias, and appropriate statistical methods, such as instrumental variables, are necessary to correct this bias.