Estimation of Household Energy Demand and Its Elasticities in Urban Ethiopia: A Case Study of Nakamte Town

Abstract

This study looks into households’ energy demand and its elasticities in Nakamte town. The objective of the study is to estimate household demand for the main energy sources (fuel-wood, charcoal, kerosene and electricity), and their elasticities of demand. The study used household-level survey data collected from 120 sample households in Nakamte town. A two- stage sampling procedure which consists of simple and systematic random sampling techniques was employed to select the sample households. The study used a Linear Approximated Almost Ideal Demand System model (LA/AIDS) to estimate household energy demand system. Estimates of the model were constrained to comply with neoclassical theoretical restrictions on demand, and the model was estimated using Iterative Seemingly Unrelated Regression (ISUR). The result shows that prices of energy sources significantly influence the energy expenditure shares of all energy sources. Demand for all energy sources are price elastic and all energy sources are substitutes. Except for charcoal, household total energy expenditure significantly influence energy expenditure shares of energy sources. All energy sources considered have expenditure elasticities greater than zero indicating no energy source is an inferior good for households in the study area. Furthermore, energy expenditure share of a particular energy source is significantly influenced by one or more of household characteristics such as sex and age of household head, education level of the wife and refrigerator ownership. Policy interventions aimed to change the household energy consumption pattern of the town need to focus on the estimated elasticities and household characteristics that significantly influence household energy demand.

Keywords: Household energy, Iterative Seemingly Unrelated Regression, Linear Approximated Almost Ideal Demand System, energy source, Nakamte town