International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, v. 26, n. 3, pp. 397-428. Abstract: Fuzzy models have been increasingly used in decision-making in the energy sector to deal with many uncertainties such as lack of data and climate change. This paper presents a global energetic efficiency analysis based on the time series data of 91 countries from 1960 to 2010, using an integrated two-stage fuzzy approach. More precisely, Fuzzy DEA models for traditional constant and varying returns to scale assumptions are employed in a first stage to assess the relative efficiency of these countries over the course of time. In the second stage, fuzzy regressions based on different rule-based systems are used to predict the impact of a set of demographic and socio-economic variables on energy efficiency. Energetic efficiency appears to be explained by the countervailing forces of urbanization, wealth inequality, and social development. Thus, a transition to a more energetic efficient lower carbon society will depend on how we address certain socio-political factors, such as pursuing a more sustainable urbanization, reducing inequalities and taking into consideration socio-environmental aspects in trade agreements.