Can we use the energy intensity indicator to study “decoupling” in modern economies?

The Economic Energy Intensity of a country, defined as the ratio between the energy consumed and the Gross Domestic Product of a country, is a widely-adopted indicator of economic performance, often applied to sustainability analysis. The usefulness of this indicator has been challenged on the basis of the fact that the two variables making up this ratio are correlated. This implies that, when considered in isolation, each one of these two variables (Energy Consumption p.c. and GDP p.c.) maps on to typologies of socio-economic systems (e.g. developed countries vs. less developed countries), however calculating their ratio produces an indicator which does not map on to any meaningful typology of economic system (a number without an external referent). This paper checks the validity of this criticism by looking at the Economic Energy Intensity (EEI) of 133 countries over the period 1960–2009. Two approaches are used for this task: (1) an analysis at a given point in time (year 2009): after excluding countries with real GDP p.c.<5000 US$ I considered three clusters of countries with similar values of EEI. Within each cluster, when looking at similar values of EEI, we find extremely diverse typologies of economies; (2) a video showing the movement in time of 133 countries between 1960 and 2010 over the plane determined by the two variables composing the indicator. These results confirm the doubts about both the meaning and usefulness of the indicator itself. In particular this analysis indicates that the problems with the EEI indicator are generated by the attempt to compress into a single number different pieces of information referring to different dimensions (monetary and energy flows) and to different scales (the characteristics of the whole country are determined by the specific characteristics of different sectors).

► It provides a brief overview over the controversy over the usefulness of the EI indicator (applied to the national level).
► It provides an analysis based on data of energy intensity for 133 countries in year 2009 (tables and figures) and over the period 1960–2010 (video on an external website).
►  It shows that, above the threshold of GPD p.c. >5000pppUS$, clusters of countries expressing similar values of EI are very diverse in their socio-economic characteristics.
► It provides empirical evidence that shows the EI indicator should not be used to study dematerialization and decoupling when applied at the national level.