Understanding CSR in Complex Supply Chains: Insights From Bayesian Models and Machine Learning

Abstract

This study examines the relationship of key variables in corporate social responsibility (CSR) in supply chains, with the focus being buyers' operational slack, customer bargaining power, and industrial dynamism as a moderator. Employing Bayesian Hierarchical Models and Stacked Ensemble Learning on data of 314 Chinese manufacturing firms covering 2000–2020. Results indicate a positive link of operational slack and supplier CSR performance and a negative one linking customer bargaining power to supplier CSR performance. Industrial dynamism, on the other hand, negatively moderates the association between buyers' operational slack resources and supplier CSR performance. Similarly, industrial dynamism has a detrimental moderating effect on the relationship between powerful customers and suppliers CSR performance. The findings clarify the importance of slack resources, bargaining power, and industrial dynamism, offering valuable practical implications for corporations wishing to enhance CSR strategies. This study, therefore, fills the gaps in existing literature as it explores the implications of CSR in complex supply chain networks and intends to contribute to the increased understanding of supplier CSR. This study is particularly useful for practitioners and policymakers aiming to undertake sustainable supply chain management.