Corporate Social Responsibility Disclosure as an Agency Cost Reduction Mechanism for Investment Decisions in the Apparel & Textile Sector

Abstract

This study examines the role of Corporate Social Responsibility (CSR) disclosure as a mechanism to reduce agency costs and improve investment decisions in Indonesia’s apparel and textile sector. Using data from publicly listed companies on the Indonesia Stock Exchange (IDX) from 2020 to 2024, the research employs a quantitative approach with panel data regression and mediation analysis conducted using EViews software. CSR disclosure, measured through a CSR Disclosure Index, serves as the independent variable, while agency costs, assessed via operating and administrative expense ratios, function as the mediating variable. Investment decisions, the dependent variable, are evaluated using financial indicators such as Net Present Value (NPV), Internal Rate of Return (IRR), Return on Investment (ROI), and capital expenditure allocation efficiency. The findings reveal that CSR disclosure significantly reduces agency costs by enhancing transparency and accountability, thereby mitigating conflicts between shareholders and managers. Additionally, CSR disclosure positively impacts investment decisions, with agency costs partially mediating this relationship. These results align with agency theory, demonstrating the value of CSR practices in improving corporate governance and investment efficiency. However, the partial mediation effect suggests that other factors, such as market dynamics and regulatory frameworks, may also influence the CSR-investment relationship. This study contributes to the literature by validating agency theory in the context of CSR and investment decision-making in an emerging market setting. It provides practical implications for policymakers and practitioners, emphasizing the importance of standardized CSR reporting to foster better governance and financial performance. Future research should explore additional mediators and expand the scope to other industries and regions to generalize the findings further