Econometric analysis of the residential demand for electricity in the Pacific Northwest
This study using household-level data examines two sources of potential bias in ordinary-least-squares (OLS) estimates of residential electricity-demand equations. First, if appliance-choice decisions are related to short-run utilization decisions, as theory implies, then bias may arise from endogeneity of appliance dummy variables in the equation. Second, bias may arise from endogeneity of electricity-price variables under multi-step block pricing. To examine the first question a joint model of appliance ownership and utilization is considered. Ownership or fuel-choice probabilities are estimated for various electricity-using appliances, using a multinomial logit specification. The electricity demand equation is estimated by OLS and by instrumental variables (IV), using fitted-ownership probabilities as instruments for the appliance dummy variables. To examine the second question, actual rate schedules are used to compute various price measures. Significant bias occurs in OLS estimates of elasticities with respect to observed marginal price and block-rate premium, the consumer's surplus captured by the seller under the block regime. The joint treatment of appliance ownership and utilizaation permits estimation of both long-run and short-run elasticities. Elasticities are considerably higher in the long run than in the short run, due largely to space- and water-heat fuel switching.
- OSTI ID:
- 5832987
- Resource Relation:
- Other Information: Thesis (Ph. D.)
- Country of Publication:
- United States
- Language:
- English
Similar Records
Modeling the residential demand for energy
Fuel choice and aggregate energy demand in the residential and commercial sectors of the US