摘要

This paper applies an aggregate production function to examine the dynamic linkages among energy consumption, capital stock, and real income (real GDP per capita) in G-7 counties. We employ the Toda and Yamamoto (1995) Granger causality test, the generalized impulse response approach, and variance decompositions in a multivariate setting to uncover the extent and the magnitude of the relationship among variables. Our empirical results find evidence of a unidirectional relationship running from energy consumption to real income in Canada, Italy, and the UK, indicating that energy conservation may hinder economic growth in the three countries. Furthermore, the causality relationship appears to be unidirectional, but reversed, for France and Japan, implying that energy conservation in both countries may still be viable, but without being detrimental to economic growth. As to Germany and the U.S., there is no causality between the variables, which demonstrates the 'neutrality hypothesis', and in such a case economic growth will not affect energy use. We further see that the impact of capital stock is relatively higher compared with that of energy consumption. Transitory initial impacts of innovations in energy consumption on capital stock and real income are observed.