Interfuel substitution and changes in the way households use energy : the case of cooking and lighting behavior in urban Java

Abstract

A major shortcoming of standard methods for estimating household fuel demand as a function of fuel choice is that, for cooking, end-use services are treated as constant, while not taking into account behavioral changes that accompany interfuel substitution. For lighting, the amount of electricity required to displace kerosene is often estimated on the basis of assumptions about unmet lighting demand in households lighting with kerosene. This paper develops a statistical procedure to analyze interfuel substitution based on the actual behavior of urban households. Applied to data from a recent survey of energy consumption in urban households on Java, the amount of energy that is actually consumed by comparable households using different fuels (fuel substitution ratios) can be estimated. The analysis of fuel use by urban households on Java provides a sound basis for projecting the substitution potential of kerosene for wood and liquified petroleum gases (LPG) for kerosene in cooking end-use, thereby clarifying the range of estimates commonly derived from various assumptions about average stove efficiencies. Results also show that the assumptions traditionally used to estimate the benefits of electric lighting are far off the mark. For households lighting with electricity versus those lighting with kerosene, it was found that electricity is less costly to the country and to households, and households also enjoy roughly six times more light.