The rapidly growing literature on the relationship between energy consumption and economic growth has not univocally identified the "real" causal relationship yet. We argue that bivariate models, which analyze the causality at the level of the total economy, are not appropriate - especially in cases where both variables do not cover the same scope of economic activity. After discussing appropriate pairs of variables, we investigate Granger causality between energy consumption and GDP in the U.S. for the period from 1970 to 2007 for three sectors - industry, commercial sector, transport as well as for the total economy. The choice of additional variables is based on major findings from the Environmental Kuznets curve literature and its critical reflections. Using the recently developed ARDL bounds testing approach by Pesaran and Shin (1999) and Pesaran et al. (2001), we find evidence for long-run Granger causality for the commercial sector, in case energy is the dependent variable, as well as bi-directional long-run Granger causality for the transport sector. We conclude that controlling for trade as well as increasing energy productivity significantly improves the fit of several extensions of the bivariate model.