The relationship between energy consumption and economic growth has created a large body of research in the energy-economics literature. In this paper, we investigate such a relation in the case of Chinese regions from 1995 to 2009. The majority of previous studies have ignored the regional dimension and the cross-sectional dependence of provinces. Besides, different energy policies adopted by the government have influenced energy intensity over time, showing improvement in the 1990s and deterioration from 2000 onwards. Thus, it is necessary to examine these two periods separately. Moreover, a detailed disaggregation of total energy consumption into electricity, coal, coke, and crude oil consumption and its linkage with economic growth may provide new insights for the design of energy policy across Chinese regions. We use panel techniques to test the direction of the causality in the long- and short-run between these different types of energy consumption and economic growth. Our results are mixed from 1995 to 2009 due the aforementioned break around 1999. However, in all cases our estimations provide empirical evidence that from 1999 to 2009 there is unidirectional causation from economic growth to energy consumption in the long-run. Therefore, energy-saving policies can be adopted without interrupting the path of growth.