Several papers have examined the empirical link between energy consumption and economic growth. Most of these studies differ in their methodology and data used. While some are country specific studies, others used several countries in their study at the same time. Even with recent literatures, empirical evidence is still mixed, and whether energy consumption Granger causes economic growth or the reverse is still to be debated. The non--consensus on the causality between energy and output might be because different economies have different energy consumption patterns as well as different sources of energy which might have varying impacts on economic growth.
As documented in Chontanawat et al. (2006), Jumbe (2004) stated that if causality runs from energy consumption to economic growth, then it implies that an economy is energy dependent. On the other hand, if causality only runs from economic growth to energy consumption, it means that an economy does not depend on energy. Hence as noted by Masih, A. M. M. and Masih, R. (1997) and Ighodaro and Ovenseri-Ogbomo (2008) amongst others, energy conservation policies may be implemented with no adverse effect on growth and development as well as employment. Furthermore, if there is 'neutrality hypothesis' that is, if there is no causality in either direction, it means that energy consumption is not correlated with economic growth, so that energy conservation policies may be carried out without adversely affecting the economy.
Most previous studies on causality relationship between energy and economic growth used bivariate Granger causality tests to draw inference including previous study by Ighodaro and Ovenseri-Ogbomo (2008) for Nigeria. This approach, although appealing for its simplicity, is problematic on at least two accounts. First, single-equation ECMs are only valid given exogeneity assumptions (Bannered et al. 1993). To test the relationships between two or more series, bivariate specifications cannot capture all relevant information. For example, Glasure (2002) argued that the empirical examination of the association between energy and real income must also include money, government spending and the price of energy. Hence Glasure used Gross Domestic product, energy consumption, government expenditure, money supply and oil price in his study. To solve this problem of bivariate specification, the test for this paper is carried out in a multivariate framework proposed in the Johansen test (Johansen 1991).
Contrary to the earlier study of Ighodaro and Ovenseri-Ogbomo (2008), Omotor (2008) and Olusegun (2008) for Nigeria and follow from the studies of Yang (2000) and Glasure (2002) for Korea, the study re-examined causality relationship between energy consumption and economic growth for Nigeria by including monetary policy variable and another representing government activities as well as using the Johansen co-integration technique and the multivariate Granger causality method. Section two considers literature review on causality relationship between energy consumption and economic growth, energy situation in Nigeria is discussed in section three while section four examined methodology and discussion of results. Section five concludes.
2. Literature Review on Causality Relationship between Energy Consumption and Economic Growth
Previous researchers used different econometric methodologies to test the relationship among growth, income and energy consumption. Rasche and Tatom (1977), Kraft, J. and Kraft, A. (1978), Berndt (1978), Akarca and Long (1980), Proops (1984), Yu and Hwang (1984) as well as Nachane et al. (1988) were the earliest researchers in this area. Rasche and Tatom (1977) specified a production function for the United States. They included energy as explanatory variable along with land, labour and capital. They showed that energy price increases induced declining trends on potential gross national product over the study period. …