After the oil crisis of 1973, the California Public Utilities Commission (CPUC) in 1976 ordered Pacific Gas and Electric Company (PGandE) to charge its large nonresidential customers with monthly billing demand of over 4000 kilowatts (kW) mandatory time-of-use rates. Using a translog (TLOG) specification attributable to Christensen, Jorgenson, and Lau (1973), Chung and Aigner (1981) estimate the electricity demand price elasticities by time-of-use for 64 of these customers in 13 Standard Industrial Classification (SIC) code groups. Own-price elasticity estimates are generally around -0.1 and at times can be as high as -0.5, or they have the wrong sign. Cross-price elasticity estimates indicate that electricity usages by time-of-use are mostly substitutes. However, the estimated price responsiveness typically is larger than observed usages (see below and the section, Experimental Design and Data). Moreover, positive own-price elasticity estimates, though not statistically significant, raise further doubts about the validity of empirical results.
[1]
Walter Diewert,et al.
An Application of the Shephard Duality Theorem: A Generalized Leontief Production Function
,
1971,
Journal of Political Economy.
[2]
Walter Diewert,et al.
Exact and superlative index numbers
,
1976
.
[3]
D. Aigner,et al.
An Analysis of Commercial and Industrial Customer Response to Time-of-Use Rates
,
1983
.
[4]
W. Diewert.
Intertemporal Consumer Theory and the Demand for Durables
,
1974
.
[5]
Douglas W. Caves,et al.
Econometric analysis of residential time-of-use electricity pricing experiments
,
1980
.
[6]
Industrial and Commercial Demand for Electricity by Time-of-Day: A California Case Study
,
1981
.
[7]
Jan Kmenta,et al.
A General Procedure for Obtaining Maximum Likelihood Estimates in Generalized Regression Models
,
1974
.
[8]
Costs and Benefits of Residential Time-of-Use Metering
,
1982
.