Investigating the EKC hypothesis with disaggregated energy use and multi-sector production

Abstract

The Middle East and North African (MENA) economies experienced substantial economic fluctuations due to variations in carbon emissions and energy use. For this purpose, the present study examines the factors influencing carbon emissions in MENA economies, particularly economic growth and energy use. To this end, this study disaggregates economic growth into three sectors (agriculture, industry, and services) and energy use into renewable and non-renewable, and examines their environmental impacts by including the roles of urbanization and trade openness in the environment Kuznets curve (EKC) framework. This study uses panel data from 16 MENA countries over the period 1990-2018 to estimate the short-run and long-run coefficients and the Granger causality between the variables. The empirical results using the Mean Augmented Group (AMG) and the Common Correlated Effects Mean Group (CCE-MG) revealed that (i) the signs of GDP per capita and its squared scores validate the EKC hypothesis only at the aggregate level; (ii) the coefficients of sectoral GDP show that the industry and services sectors have the highest contributions to carbon emissions in the MENA region; (iii) non-renewable energy increases emissions, whereas renewable energy lessens them. The outcomes of the Granger causality confirm (i) a bidirectional relationship between emissions and per capita GDP as well as sectoral GDP, between CO₂ emissions and renewable energy, and between per capita GDP and renewable and non-renewable energy however; (ii) a unidirectional causal impact running from non-renewable energy to CO₂ emissions is found in the short term. The study calls for effective policies to focus on curbing emissions in secondary and tertiary economic sectors by growing the part of renewable energy in the total energy mix.

Keywords: Carbon emission; Economic growth; Environmental Kuznets curve; MENA countries; Renewable energy and non-renewable energies.