Corporate social responsibility (CSR) is increasingly crucial in the business world, emphasizing ethical conduct and consideration for operational environments. Stakeholders expect companies to disclose their CSR practices regarding employee welfare, environmental impact, community engagement, governance, and human rights. Existing research demonstrates the significant influence of CSR practices disclosure (CSRPD) on corporate financing decisions, audit risk, investment efficiency, financial and market performance, and foreign direct investment, establishing it as a valuable intangible asset. However, the quality and quantity of CSRPD vary significantly. Some studies suggest that corporate structure, particularly family versus non-family ownership, might explain these differences. The role of family companies in CSRPD is unclear; their level of social consciousness and transparency may vary considerably. Country-level institutions also impact CSR practices and the family company-CSR disclosure link. National or cultural differences may explain family companies' socially conscious behavior, especially in developing countries with policy and legal constraints. Family companies' behavior might also vary depending on the extent of their relationships with authorities, potentially leading to both positive and negative extremes in sustainable behavior. This study focuses on Saudi Arabia, a country with a significant presence of family-owned companies (70% of publicly listed companies are family-controlled) and where CSR practice and reporting are still developing. The Saudi Code of Corporate Governance (SCCG) of 2017 emphasized CSR initiatives, but the quality of CSRPD remains low despite a high quantity of disclosed information. Saudi Arabia's economic structure, with many family-controlled companies and high potential for environmental pollution, makes this investigation particularly relevant. This study examines the governance structure, particularly board independence, in family companies to understand how their CSR behavior manifests. The research questions are: 1) Does family control affect CSRPD for Saudi Arabian companies? and 2) What role does board independence play in shaping the behavior of family control companies towards CSRPD?

The stakeholder theory posits that companies should integrate stakeholder interests into their strategies and decision-making, with CSR representing a strategic plan for fulfilling moral, ethical, and social obligations. Family companies differ from non-family companies by considering both economic and non-economic (socioemotional wealth, SEW) goals, where SEW focuses on benefits for family members. The SEW theory explains family companies' unique behavior, emphasizing the preservation and enhancement of socioemotional values and family needs. This can lead to greater concern for stakeholders through actions such as community investment. However, counterarguments suggest that family companies may prioritize self-interest over societal well-being, potentially exploiting employees or expropriating resources at the expense of minority shareholders. Empirical evidence on family companies' CSR behavior is mixed. Some studies show family companies exhibiting greater social consciousness, supporting legitimacy, and preserving SEW through CSR engagement. Others find family companies engaging less in CSR activities, potentially due to focusing on family interests or considering CSR as costly, short-term investments. Studies using international samples have shown varying results, with some demonstrating higher CSR levels in family companies and others indicating lower levels. These inconsistencies highlight the need for further research, especially in understudied contexts like the Arab world, to understand the nuances of family influence on CSR behavior. The role of board independence, a key governance mechanism, needs further examination in the context of family companies and their CSR practices. Specifically, the moderating role of board independence on the relationship between family control and CSR disclosure is critical to understand.

This study uses a sample of 152 non-financial companies listed on the Saudi Arabian capital market (Tadawul) between 2016 and 2021 (837 company-year observations), spanning 11 industry sectors. The sample period is chosen because it covers the implementation of Saudi Vision 2030, a strategic plan emphasizing sustainable social development, and the 2017 update to the SCCG, both of which aimed to improve CSR reporting. Data on CSR disclosures, family control, and corporate governance were collected from annual reports and the Thomson Reuters DataStream database. **Dependent Variable:** CSRPD was measured using a self-constructed index based on a 37-item checklist covering six categories: community, customer, employees, energy, environment, and products and services. The checklist incorporates previous research in the Saudi context and guidelines from the Global Reporting Initiative (GRI-G4), ISO 26000, and the Capital Market Authority (CMA) ESG disclosure guidelines. Each item was scored from 0 to 3, depending on the quality of information provided, resulting in a CSRPD score for each company. **Independent Variables:** Family control was measured as the percentage of shares held by family members, using various ownership thresholds (5%, 10%, 15%, 20%) to assess robustness. Board independence (BIND) was defined as the percentage of independent directors on the board. **Control Variables:** Several corporate governance variables were included: board size (BSIZE), ownership concentration (OWCO), institutional investor ownership (IOW), foreign ownership (FOWN). Financial variables were also used: company age (FAGE), sales growth (SGROWTH), return on assets (ROA), company size (FSIZE), debt ratio (LEVGE), market-to-book ratio (MTB), loss (LOSS), and systematic risk (BETA). Industry and year fixed effects were also included. **Model Specification:** A pooled ordinary least squares (POLS) regression model was estimated, examining the relationship between Family, BIND, their interaction, and CSRPD, while controlling for other variables. Various econometric techniques were employed to address potential issues: two-way cluster-robust standard errors to account for heteroscedasticity and autocorrelation, winsorizing to mitigate outliers, and alternative regression models (Newey-West, PCSE, logistic regression) to test robustness. Heckman’s two-stage model and 2SLS regression were used to address the potential endogeneity of family control. **Alternative CSRPD Measurement:** An alternative CSRPD measure, focusing on the quantity of disclosure rather than quality, was also created using a dichotomous approach (1 if disclosed, 0 if not). The model was re-estimated using this alternate measure. Subsample analysis was conducted to separately examine companies with high and low board independence, using the median as a cutoff.

Descriptive statistics reveal that the mean CSRPD score for Saudi companies was 0.991 (33%), with family-controlled companies scoring significantly lower than non-family companies across all CSR dimensions. Family-controlled companies also had significantly lower board independence, board size, ownership concentration, institutional ownership, and foreign ownership. Correlation analysis showed a significant negative correlation between family control and CSRPD, but no significant multicollinearity among variables. POLS regression results (Table 4) show a statistically significant negative relationship between family control and CSRPD (-0.004, p<0.001). This suggests that family-controlled companies disclose less CSR information than non-family companies, contradicting predictions based on SEW theory but aligning with the agency theory. However, when the interaction between board independence and family control was included (Table 4, Column 2), the results revealed a significant positive interaction effect (0.040, p<0.05). This indicates that a higher percentage of independent directors on the board of a family-controlled company significantly increases their CSRPD. This finding is robust to the use of an alternative measure of CSRPD based on the quantity of disclosure (Table 5). The findings remain consistent when using different thresholds for defining family control (Table 6). Subsample analysis (Table 7) further supports these findings, showing a negative association between family control and CSRPD only in companies with low board independence. Alternative regression models (Newey-West, PCSE, logistic regression, Table 8), and methods to address endogeneity (Heckman two-stage and 2SLS, Tables 9 and 10), all confirm the main findings. Other control variables showed significant effects: foreign ownership was negatively associated with CSRPD, while firm size was positively associated.

The findings address the research questions by showing that family control is negatively associated with CSRPD in Saudi Arabia, but this relationship is significantly moderated by board independence. The negative association between family control and CSRPD aligns with the agency theory, suggesting that family owners may prioritize their own interests over broader stakeholder concerns. However, the significant positive interaction effect of board independence indicates that the presence of independent directors can mitigate this negative effect and encourage higher levels of CSRPD in family-controlled firms. These findings contribute to the literature by providing detailed empirical evidence from a significant sample of Saudi Arabian companies, a region underrepresented in previous studies. The interaction effect highlights the importance of corporate governance mechanisms in shaping CSR behavior, specifically in family businesses, where SEW might otherwise outweigh stakeholder considerations. The study's robust results across various econometric techniques and alternative measurements enhance its reliability and generalizability.

This study provides strong evidence that family-controlled companies in Saudi Arabia report lower levels of CSR practices compared to their non-family counterparts. However, the presence of independent directors on the board significantly moderates this relationship, leading to increased CSR disclosure in family firms with higher board independence. This highlights the vital role of corporate governance structures in promoting responsible behavior in family businesses. Future research could expand this study to other Arab countries, explore the specific mechanisms through which board independence influences CSR disclosure, and investigate the influence of other governance mechanisms or specific SEW dimensions.

The study's focus on Saudi Arabia limits the generalizability of findings to other contexts. The reliance on self-reported data from annual reports may also introduce potential biases. Future research could explore other data sources, such as surveys or interviews, to triangulate findings and enhance the study's validity. Additional research into the specific mechanisms by which board independence influences CSR behavior within family businesses would also provide additional valuable insights.