Bayesian Hierarchical Modeling of Individual Effects: Renewables and Non-Renewables on Global Economic Growth

Abstract

Examining the relationship between renewable and non-renewable energy sources and economic growth is crucial for designing sustainable growth policies in the context of global sustainability efforts. Previous studies relying on frequentist inference have faced challenges in disentangling the individual effects of these energy sources on economic growth due to their high degree of correlation, often leading to biased results. The Bayesian approach offers an alternative estimation method to address this multicollinearity issue. This study aims to demonstrate one of the advantages of the Bayesian hierarchical framework in handling multicollinearity by using a sample of 72 countries to evaluate the distinct impacts of renewable and non-renewable energy on economic growth. By incorporating specific priors into a Bayesian model to guide the estimation process, the findings confirm that both energy sources play significant roles in driving economic growth, with renewable energy sources exhibiting a comparatively weaker effect. These results align with theoretical expectations, indicating that renewables make a limited contribution to economic growth due to high investment costs, intermittency issues, and supply chain constraints. This study establishes a solid foundation for sustainable growth policy formulation by providing robust evidence.