The aim of this paper is to re-examine the relationship between coal consumption and real GDP of China with the use of panel data. This paper applies modern panel data techniques to help shed light on the importance of the heterogeneity among different regions within China. Empirical analyses are conducted for the full panel as well as three subgroups of the panel. The empirical results show that coal consumption and GDP are both I(1) and cointegrated in all regional groupings. Heterogeneity is found in the GDP equation of the full panel. The regional causality tests reveal that the coal consumption–GDP relationship is bidirectional in the Coastal and Central regions whereas causality is unidirectional from GDP to coal consumption in the Western region. Thus, energy conservation measures will not adversely affect the economic growth of the Western region but such measures will likely encumber the economy of the Coastal and Central regions, where most of the coal intensive industries are concentrated.