Assessing the Environmental Impact of Economic and Urban Factors Using Bayesian Two-Stage Least Squares

The rapid industrialization and urbanization in South Asia have led to significant environmental challenges, particularly in the form of rising CO2 emissions. This study examines the relationship between economic factors such as trade, financial development, GDP per capita, and urban population growth, and their impact on CO2 emissions in South Asia. Using secondary panel data from international organizations and government publications, covering the years 2000 to 2020, we apply a Bayesian Two-Stage Least Squares (2SLS) method to address endogeneity and improve estimation accuracy. Our key findings indicate that trade and GDP per capita are positively correlated with CO2 emissions, while financial development has a negative effect, suggesting that more advanced financial systems may aid in emission mitigation. Urban population growth, however, does not exhibit a statistically significant effect on CO2 emissions. This study contributes to the understanding of the economic-environmental nexus in South Asia and emphasizes the need for policies that balance economic growth with environmental sustainability. The novelty of this research lies in its use of Bayesian 2SLS to address the endogeneity problem in estimating the impact of economic activities on CO2 emissions in the region. Graphical abstract This research investigates the links between trade, financial development, CO2 emissions, and urban population growth using Bayesian analysis. It concludes that trade positively affects CO2 emissions whereas financial development does not significantly influence them. Further, increase in population residing in urban areas contributes to higher emissions. In dealing with the problem of endogeneity, the research applies Bayesian 2SLS estimation alongside a fixed-effect model which ensures unobserved heterogeneity bias control, yielding robust results. The study aims to resolve endogeneity employing instrumental variables (IVs) and provides credible estimates for these relational factors. Findings reveal that trade is a considerable driver of emissions whereas financial development contributes negligibly. The model performs appropriately capturing a large share of emissions variation explaining this phenomenon. All in all, the research showcases the role trade plays on environmental impact while underlining the need for advanced statistical techniques when exploring such relationships.