Is there a relationship between natural gas consumption and the environmental Kuznets curve? Empirical evidence from Bangladesh

Bangladesh has significant natural gas reserves, and total demand has climbed substantially in recent years. The study uses the autoregressive distributed lag (ARDL) model for cointegration and the vector autoregressive(VAR) Granger causality model to analyze a long-run link between natural gas (NG) consumption, economic development, urbanization, and CO₂ emissions. The objective is to investigate the relationship between the environmental Kuznets curve (EKC) and Bangladesh’s NG consumption using data from the years 1990 to 2018. According to the ARDL model, economic growth, urbanization, and NG consumption, all have a positive and significant influence on CO₂ emissions. Despite having a negative coefficient, the square of economic development has a significant impact on CO₂ emissions. In the long run, it verifies the EKC hypothesis in Bangladesh. Both linear and nonlinear economic development determinants display statistically significant positive and negative signals in the short run. From Bangladesh’s perspective, this also demonstrates the presence of an EKC. The impact of NG consumption in the short run is insignificant; nevertheless, urbanization has a significant effect. The VAR Granger causality demonstrates that economic development and urbanization have a bidirectional response; however, NG consumption and CO₂ emissions have just one-way causality. The key policy implication of the study is that NG use is expected to raise emissions. Increasing the share of clean energy in the energy utilization system, such as nuclear power and renewable energy, is a plausible policy choice.