Impact of Corporate Social Performance on Financial Performance: Evidence from Islamic Banks, Conventional Banks and Social Banks

Purpose — This study aims to assess the impact of banking models on the relationship between corporate social performance (CSP) and corporate financial performance (CFP) in determining a viable model for sustainable banking. Design/Methodology/Approach — The study uses a cross-country sample of 117 financial institutions across 36 countries over an 8-year observation period between 2013 and 2020. To address heterogeneity and endogeneity issues, the authors use the System Generalised Methods of Moments (GMM) estimation models. The study also constructs a novel CSP Index as the independent variable for the research. This CSP Index comprises six indicators reflecting dimensions of financial inclusion and intermediation, serving as proxies for sustainable banking. Findings — The findings reveal that the distinct banking models have a significant impact and can alter the direction of the CSP-CFP relationship. Specifically, the conventional banking (CB) model exhibits a statistically significant negative association between CSP and CFP. Conversely, the Islamic banking (IB) model emerges as a promising avenue for sustainable finance, indicating that increased corporate social responsibility (CSR) activities within Islamic banks (IBs) lead to greater profitability. This difference arises from the inherent strengths of the IB system in conducting financial intermediation and inclusion activities. This contrasts with the CB model’s reliance on debt-based instruments, which exacerbates risk and detrimentally impacts financial performance. The findings also show that the social banking (SB) model has a significant effect on the CSP-CFP relationship. Originality/Value — The findings give new insights into the longstanding debate on the CSP-CFP relationship by examining the impact of banking models. Introducing a novel CSP Index, characterised by its objectivity and verifiability, addresses the prevalent issue of bias inherent in the CSP indices of previous studies.