This paper evaluates the convergence of energy intensities for a group of 97 countries in the period 1971–2003. Convergence is tested using a recent method proposed by Pesaran (2007) [Pesaran, M.H., 2007. A pair-wise approach to testing for output and growth convergence. Journal of Econometrics 138, 312–355] based on the stochastic convergence criterion. An advantage of this method is that results do not depend on a benchmark against which convergence is assessed. It gives more robust results. Applications of several unit-root tests as well as a stationarity test uniformly reject the global convergence hypothesis. Locally, for Middle East, OECD and Europe sub-groups, non-convergence is less strongly rejected. The introduction of possible structural breaks in the analysis only marginally provides more support to the convergence hypothesis.