TY - JOUR
T1 - The moderating role of energy consumption in the carbon emissions-income nexus in middle-income countries
AU - Ehigiamusoe, Kizito Uyi
AU - Lean, Hooi Hooi
AU - Smyth, Russell
PY - 2020/3
Y1 - 2020/3
N2 - We examine the role of energy consumption in moderating the carbon dioxide emissions-income nexus in 64 middle income countries and compute the marginal effects of real GDP per capita on carbon dioxide emissions at various levels of energy consumption. To do so, we employ multiplicative interaction models because the traditional quadratic Environmental Kuznets Curve model cannot capture the marginal effects. We also use empirical techniques that can account for cross-sectional dependence, such as the Westerlund cointegration test and the Augmented Mean Group estimator. While the marginal effects of GDP on carbon dioxide emissions at the minimum, mean and maximum levels of energy consumption are 9.996, 9.210 and 8.452, respectively, we find no significant evidence that energy consumption moderates the relationship between income and carbon emissions in the panel. However, when we focus on specific countries, we find that energy consumption moderates the nexus between carbon emissions and income in roughly one-third of our sample and that the moderating effect is negative in about one fifth of the sample. We conclude with a discussion on why the moderating effect of energy consumption on the carbon emissions-income nexus differs between countries and offer some policy recommendations that are grounded in the main findings.
AB - We examine the role of energy consumption in moderating the carbon dioxide emissions-income nexus in 64 middle income countries and compute the marginal effects of real GDP per capita on carbon dioxide emissions at various levels of energy consumption. To do so, we employ multiplicative interaction models because the traditional quadratic Environmental Kuznets Curve model cannot capture the marginal effects. We also use empirical techniques that can account for cross-sectional dependence, such as the Westerlund cointegration test and the Augmented Mean Group estimator. While the marginal effects of GDP on carbon dioxide emissions at the minimum, mean and maximum levels of energy consumption are 9.996, 9.210 and 8.452, respectively, we find no significant evidence that energy consumption moderates the relationship between income and carbon emissions in the panel. However, when we focus on specific countries, we find that energy consumption moderates the nexus between carbon emissions and income in roughly one-third of our sample and that the moderating effect is negative in about one fifth of the sample. We conclude with a discussion on why the moderating effect of energy consumption on the carbon emissions-income nexus differs between countries and offer some policy recommendations that are grounded in the main findings.
KW - Carbon emissions
KW - Economic growth
KW - Energy consumption
KW - Environmental Kuznets Curve
KW - Interaction models
UR - http://www.scopus.com/inward/record.url?scp=85076865041&partnerID=8YFLogxK
U2 - 10.1016/j.apenergy.2019.114215
DO - 10.1016/j.apenergy.2019.114215
M3 - Article
AN - SCOPUS:85076865041
SN - 0306-2619
VL - 261
JO - Applied Energy
JF - Applied Energy
M1 - 114215
ER -