Robust determinants of CO2 emissions

Abstract

We synthesize the findings of a large number of recent papers on the determinants of CO2 emissions and identify the most robust determinants accounting for model uncertainty using Bayesian Model Averaging. Our results show that GDP per capita, the share of fossil fuels in energy consumption, urbanization, industrialization, democratization, the indirect effects of trade (networks effects) and political polarization are the robust determinants of CO2 emissions per capita. All of these determinants negatively affect the environment, with the exception of greater political polarization, which reduces the level of CO2 emissions. We also find that the determinants of CO2 emissions depend on the level of income per capita of a country. In low-income economies, foreign direct investment increases environmental degradation, while tourist arrivals have a negative impact on the environment in high-income economies. The results are robust to the Cluster-LASSO model selection method and Weighted Average Least Squares.