Macroeconomic drivers of CO2 emissions: a panel estimation for African countries

This paper analyses the long-run and causal linkages between macroeconomic variables and CO2 emissions at both panel and country levels for four African countries with the most production-based CO2 emissions from 1990 to 2021. The study employs first-generation panel unit root tests, structural break tests, cointegration tests, cointegration estimation methods, causality tests, and variance decomposition tests. The results of three cointegration tests showed a long-run equilibrium relationship between the variables. The coefficients obtained from the dynamic ordinary least squares (DOLS) and fully modified least squares (FMOLS) estimators showed a significant positive relationship between CO2 emissions and fossil fuel consumption, energy import, energy intensity, population, and GDP. The findings indicated a significant negative relationship between CO2 emissions and renewable energy consumption in South Africa and Algeria. The coefficient of the spatial spillover variable indicates a positive spatial spillover, where CO2 emissions from neighboring countries contribute to increased emissions in the local country. The study supports the environmental Kuznets curve (EKC) hypothesis, establishing an inverted U-shaped relationship between GDP per capita and CO2 emissions among these countries. The findings of causality tests show unidirectional short-run Granger causality from fossil fuel consumption to CO2 emissions and GDP. Finally, the result of the variance decomposition showed that GDP and fossil fuel consumption are the major contributors to CO2 emissions among these countries. The result of this study provides important insights for national and international policymakers pursuing climate mitigation and economic development projects in Africa.