A latent class approach for estimating energy demands and efficiency in transport: An application to Latin America and the Caribbean

In this paper, we use a stochastic frontier analysis approach to estimate demand functions for energy in the transport sector. The estimation of frontier functions allows us to obtain energy efficiency measures that are a robust alternative to the energy intensity indicators that are commonly used for international comparisons. Due to the likely unobserved heterogeneity among countries, we propose the use of a latent class model that allows to test for the existence of groups of countries with clearly differentiated demands that are associated with distinct price and income elasticities. This study is the first to use a latent class stochastic frontier approach in the estimation of energy demand functions. The proposed procedure is applied to Latin America and the Caribbean, where the transport sector represents a large share of the total energy consumption. These energy demand functions call for the inclusion of energy price in their estimation. As the transport of both goods and passengers implies the consumption of different types of energy, an index that aggregates various energy prices is thus required for the analysis. However, international agencies do not provide specific indicators of aggregate energy prices in transport for the majority of the countries analysed. For this reason, in this paper we propose the construction of a transitive multilateral index which, in contrast to those frequently presented by the aforementioned agencies, facilitates international comparisons over time.