On the exploration of causal relationships between energy and the economy

Abstract

The paper provides an overview of theoretical and methodological aspects of approaches used to test the interaction between energy use and economic development, and discusses the appropriateness of drawing policy conclusions on the basis of energy-economy Granger causality tests. Methodological issues, arising from the number and type of variables or the test methods used, are outlined. Problems associated with the use of bivariate models are illustrated by performing energy-economy causality tests for Germany and the US, using aggregate and sectoral data and three different econometric methods. The empirical examples presented illustrate that one should be cautious when drawing policy implications with the aid of bivariate causality tests on small sample sizes. Besides econometric problems, bivariate approaches may suffer from additional shortcomings, and the paper underlines the importance of using multivariate models, which accommodate several mechanisms and causality channels and provide a better representation of real-world interactions between energy use and economic growth.