Determinants of CO2 emissions in upper middle-income group countries: an empirical investigation

This study investigates the kinked exponential growth, degree of association, and causation between economic growth, energy consumption, population, trade openness, and carbon dioxide (CO2) emissions in 25 upper middle-income group countries spanning data from 1985 to 2014. The study employed first-generation and second-generation unit root tests; prior to that, the cross-sectional dependence test is also applied and panel cointegration techniques, panel FMOLS and DOLS, and panel causality techniques are employed to test the degree of association and causation among the variables. The study reveals a long-run cointegration among the variables. Results of FMOLS declare that there are negative associations between economic growth and CO2 emissions, trade openness, and carbon dioxide emissions respectively, whereas it was found that there are positive relations between energy consumption and CO2 emissions, population, and CO2 emissions. While analyzing the association through DOLS, we find that all the selected determinants of carbon dioxide emissions are directly proportional to CO2 emissions in these countries. The panel Granger causality test indicates that there is bi-directional causality between population and economic growth and between trade openness and economic growth. Finally, the study ends with some policy suggestions and new avenues for future research.