Revisiting the EKC hypothesis in an emerging market: an application of ARDL-based bounds and wavelet coherence approaches

Abstract

This study aims to reexamine long-run and causal effects of trade openness, energy usage, gross capital formation, and real growth on CO2 emissions in Mexico utilizing recent econometric techniques. The study utilizes yearly data spanning between 1971 and 2016. No prior study has used the wavelet coherence approach to collect information on the correlation and/or causal relation between these economic variables at different frequencies and time frames. Thus, this study proposes to fill the gap in the literature. The motives of the study are to address the questions: (a) Is the EKC hypothesis valid for Mexico? (b) Is there a long-run and causal relationship between CO2 emissions and its determinants? (c) How are the indicators related at different frequencies and various periods? To capture long-run effects, the study utilized ARDL, FMOLS, and DOLS estimators, while wavelet coherence technique is utilized to explore causal effects among the variables. The empirical findings confirm that (a) EKC hypothesis is valid for Mexico; and (b) gross capital formation, energy usage, and economic growth impact CO2 emissions positively. The wavelet coherence technique revealed (a) bidirectional causality between economic growth and CO2 emissions; (b) unidirectional causality from CO2 emissions to energy usage; and (c) one-way causality running from CO2 emissions to gross capital formation. Based on these findings, recommendations were suggested.