A Bayesian network approach to understanding how rule of law shapes ease of doing business globally
摘要
This study addresses the challenge of understanding how rule of law dimensions collectively shape the ease of doing business (EoDB) across countries. Despite the recognized importance of governance for business performance, prior research often examines rule of law factors in isolation, overlooking their interdependencies. To fill this gap, this study develops two Bayesian Belief Network (BBN) models using data from 128 countries: one analyzing the relationships between the eight rule of law factors and EoDB (achieving 75% predictive accuracy), and another assessing the influence of key subfactors (85% accuracy). The findings reveal that order and security, regulatory enforcement, and open government are the most critical probabilistic determinants of favorable business conditions. Furthermore, the results highlight the strong interdependencies among rule of law components, demonstrating that governance, legal, and security factors operate as a network rather than as isolated influences. Issues such as corruption, judicial inefficiencies, and lack of transparency are probabilistically associated with lower EoDB, emphasizing the importance of holistic governance interventions. This study’s novelty lies in applying BBNs to capture these complex interdependencies, providing policymakers with a more comprehensive understanding of institutional factors that facilitate business environments.