Residential Energy Demand: a Multiple Discrete-Continuous Extreme Value Model using Italian Expenditure Data

The economic analysis of energy consumption is mostly focused on single components of total expenditure in energy-consuming services. The discrete-continuous models, following the formulation of Hanemann (1984), consider the case of perfect substitute goods: the maximization process leads to extreme corner solutions in which only one alternative is selected. According to this model the literature on energy consumption is limited to study some components of total energy consumption, i.e. space and water heating or transportation. Following the path opened by Pinjari and Bhat (2010), the goal of this paper is to build a multiple discrete-continuous model of residential energy demand based on Italian expenditure data. A non-linear utility structure, originally used in Kim et al. (2002) and extended in Bhat (2005), is implemented within the Kuhn-Tucker multiple-discrete economic model of consumer demand proposed by Wales and Woodland (1983). The paper here presented is the first application of this model to Italian expenditure data. The model predict parameters stability over time and low price elasticities for electricity (0.56) and natural gas (1.17). Considerble variations in natural gas expenditures (+54%) are predicted in case of climate changes measured of increases in Heating Degree Days (+15%).