Investigating the validity of the agricultural-induced environmental Kuznets curve (EKC) hypothesis for Ghana: evidence from an autoregressive distributed lag (ARDL) approach with a structural break

Authors:
Publisher website
Google Scholar
Abstract
Purpose: The purpose of this research is to examine the validity of the agriculture-induced environmental Kuznets curve (EKC) hypothesis with evidence from an autoregressive distributed lag (ARDL) approach with a structural break including real income and energy consumption in the model for Ghana over the period 1980–2014.Design/methodology/approach: The ARDL approach with a structural break was used to analyze the agriculture-induced EKC model which has not been studied in Ghana. The dynamic ordinary least squares (DOLS), canonical cointegration regression (CCR) and fully modified ordinary least squares (FMOLS) econometric methods were further used to validate the robustness of the estimates, and the direction of the relationship between the study variables was also clarified using the Toda–Yamamoto Granger causality test.Findings: The ARDL results revealed that GDP, energy consumption and agricultural value added have significant positive effects on CO2emissions, while GDP2reduces CO2emissions. The Toda-Yamamoto causality test results show a bidirectional causality running from GDP and energy consumption to CO2emissions whereas a unidirectional long-term causality runs from GDP2and agriculture value-added to CO2emissions.Practical implications: This finding validated the presence of the agriculture-induced EKC hypothesis in Ghana in both the short run and long run, and the important role of agriculture and energy consumption in economic growth was confirmed by the respective bidirectional and unidirectional causal relationships between the two variables and GDP. Thus, a reduction in unsustainable agricultural practices is recommended through specific policies to strengthen institutional quality in Ghana for a paradigm shift from rudimentary technology to modern sustainable agrarian technologies.Originality/value: This study is novel in the EKC literature in Ghana, as no study has yet been done on agriculture-induced EKC in Ghana, and the other EKC studies also failed to account for structural breaks which have been done by this study. This study further includes a causality analysis to examine the direction of the relationship which the few EKC studies in Ghana failed to address. Finally, dynamic ordinary least squares (DOLS), canonical cointegration regression (CCR) and fully modified ordinary least squares (FMOLS) methods are used for robustness check, unlike other studies with single methodologies.

This publication has 93 references indexed in Scilit: