The role of technology in interfuel substitution : a combined cross-section and time series approach

This paper describes interfuel substitution for coal, oil, gas and electricity at a level of 12 activities. We use cross section data in each activity for appliance technologies (heating/cooling, steam generation, industrial processes, motors and lighting/computing) to estimate fuel input demand equations by appliance technology in a panel estimation with fixed effects for activities and a uniform effect of technical progress across appliance technologies. In a synthesis with the time series approach we estimate fuel input demand equations at the "aggregate" level of activities as the weighted sum of appliance technologies by inserting parameters from the panel estimation. In this "disaggregated" model the impact of prices and of technical progress in each activity can be decomposed into two effects: (i) changes in the share of appliance technologies and (ii) fuel switch within appliance technologies.