Earnings Management, Earnings Quality, and Corporate Social Responsibility: A Panel Data Analysis

Zulu, Archbald

Abstract

Corporate Social Responsibility (CSR) has become a major priority for many companies in recent years. With increased attention placed on CSR, many questions have been raised as to whether it can help mitigate earnings management. The current research was carried out to examine the impact of CSR on earnings management by exploring the concept of earnings management, the different motivations behind it, and the role of CSR in mitigating earnings management, drawing on evidence from both theoretical and empirical studies. This research contributes to the ongoing discourse on the ethical dimensions of corporate behaviour. It provides empirical evidence on the role of CSR in mitigating earnings management, offering insights for practitioners, investors, and policymakers. Multivariate regression analysis is applied to panel financial data from 2010-2019, using a sample of the entire population of the FSTE-350 on the London Stock Exchange. Econometric analysis utilising the system General Method of Moments (GMM) addresses endogeneity concerns, while instrumental variables mitigate reverse causality issues, ensuring valid and robust empirical findings.

The findings of the study revealed that there is a positive relationship between earnings management and CSR, a negative relationship between earnings quality and earnings management, a positive relationship between financial performance and earnings quality, and a positive relationship between financial performance and CSR. The Conceptual model constructed suggests that CSR affects financial performance through its impact on earnings quality and directly through the theories of CSR. The study's conclusion emphasises that a high level of earnings management corresponds to symbolic CSR, leading to diminished firm performance within the FTSE-350 index. Managers might use CSR as a façade to conceal their earnings management practices, negatively impacting the company's long-term performance. Stakeholders are encouraged to discern between authentic CSR and symbolic CSR (greenwashing) to make informed assessments of a company's true social responsibility efforts. The conceptual model developed illustrates the non-monotonic nature of the concepts under consideration, implying a limitation in the predictability of models as the monotonicity of relationships between dependent and independent variables across the entire range is a fundamental aspect of linear regression analysis.

• Thesis Type: Thesis
• Deposit Date: Jan 25, 2024
• Publicly Available Date: Feb 27, 2024
• Award Date: Jan 26, 2024