RESIDENTIAL CONSUMPTION OF ELECTRICITY AND ENERGY EFFICIENCY A quantile regression approach

We study the residential consumption of electricity by considering the relevant socioeconomic characteristics of households, their location, and the use of electric appliances. Our empirical exercise uses microdata for the Metropolitan Area of Buenos Aires and the estimation strategy applies a quantile regression approach which enables us to identify the characteristics associated with high consumption levels, and presumably, with high energy inefficiency. The two main results of the paper are the relative low importance of income and the high importance of the lack of access to natural gas, i.e. the lack of access to one kind of energy good (natural gas) leads to inefficiencies in the consumption of another energy good (electricity). This evidence suggests that efficiency programs should be targeted to such household groups. Other interesting results are the relevance of air conditioning units as one of the main determinants of high electricity consumption and the preliminary evidence that a move from houses to apartments leads to lower consumption (and very likely, to better energy efficiency levels).