STRUCTURAL BREAKS AND ASYMMETRIC ANALYSES OF CO2 EMISSIONS, ENERGY CONSUMPTION AND ECONOMIC GROWTH NEXUS IN NIGERIA

Abstract

Energy consumption is a leading cause of carbon dioxide (CO2) emissions. Previous studies have addressed either the link between economic growth and energy consumption or economic growth and CO2 emissions and largely ignored the investigation of both relationships in a single framework. Also, the structural break and asymmetric effects of the relationship at the disaggregated level remained largely unexplored. This study was therefore designed to investigate the structural breaks and asymmetric nature of economic growth and energy consumption on CO2 emissions in Nigeria at the aggregate and disaggregated level in a single framework. The Environmental Kuznets Curve hypothesis provided the framework. It considered the dynamic effects of economic growth, energy consumption on the emissions from aggregate CO2 and the disaggregated components of coal, natural gas and petroleum products emissions using annual data between 1970 and 2017. Data were sourced from the World Bank World Development Indicator. The Dickey-Fuller with Generalised Least Squares detrending (DFGLS) and Ng-Perron (NP) tests were utilised to investigate the order of integration of the variables. The auto-regressive distributed lag (ARDL) and the Non-linear Auto-regressive Distributed Lag techniques (NARDL) that took cognisance of short and long run relationships were employed. The analysis employed structural break unit root test and the multiple break date regression test to reflect the potentials of the structural shift. The asymmetric effects of economic growth and energy consumption on CO2 emissions were analysed by differentiating between the partial sums of positive and negative shocks using the Wald tests (WSR and WLR). The coefficients were analysed at . The DFGLS and NP stationarity tests revealed a mixed order of integration of the variables. While some variables were stationary at levels, others were found stationary at first difference. The symmetric result showed that higher economic activities were responsible for higher CO2 emission in the short run ( =0.21) and long run ( =0.71), while energy consumption increased emission only in the short run ( =0.24). There was significant and increasing impact of economic growth on emissions from coal in the short run ( =1.60) and long run ( =2.96). The asymmetric result showed that economic growth was responsible for higher CO2 emission in the short run ( =0.59) and long run ( =2.24), while energy consumption increased emission in the short run ( =1.87) and long run ( =5.24). However, economic decline was responsible for reduced CO2 emission in the short run ( =0.86) and long run ( =2.74). There was significant and increasing impact of economic growth on emissions from coal only in the short run ( =1.54). Accounting for structural break, economic growth increased CO2 emission in the short run ( =0.15) and long run ( =0.24). The Wald test revealed the presence of asymmetries in the aggregate economic growth, energy consumption and CO2 emission nexus (WSR=4.74; WLR=3.21) and coal emission (WSR=5.39). Economic growth and energy consumption exhibited significant asymmetric and structural breaks on CO2 emission at the aggregate and disaggregated level in Nigeria. Appropriate domestic policies and institutional arrangements are required to reduce carbon emission in the production process.