This paper investigates the key factors affecting household energy expenditure in Egypt. Based upon the latest 2015 Egyptian HIECS Survey, we develop a quantile regression model with an innovative variable selection approach via Adaptive Lasso Regularization technique to untangle the spectrum of household energy expenditure. Unsurprisingly, income, age, household size, housing size, and employment status are salient predictors for energy expenditure. Housing characteristics have a moderate impact, while socio-economic attributes have a much larger one. The largest variations in household energy expenditures in Egypt are mainly due to variations in income, household size, and housing type. Our findings document substantial differences in household energy expenditure, originating from the asymmetric tails of the energy expenditure distribution. This outcome highlights the added value of implementing quantile regression methods. Our empirical results have various interesting policy implications regarding residential energy efficiency and carbon emissions reduction in Egypt.
Research Fellows
Fateh Belaïd
Full Professor of Economics, Lille Catholic University
Authors
Christophe Rault
Full Professor of Economics, University of Orléans