Banking disclosure and banking crises in Africa: does board gender diversity play a role?

Banking crises have become a recurring issue globally, causing significant economic disruption. Increased monitoring of banks is crucial to mitigate these crises, and market discipline can be enhanced by incorporating diverse perspectives into management. Corporate governance literature suggests that women bring unique skills and resources to the boardroom, improving decision-making and potentially mitigating crises. The threat of disciplinary action from market participants encourages greater prudence and efficiency among board members. Early detection of risks and governance weaknesses is vital to contain banking problems before they spread through the industry. The amount of information banks disclose is critical because a lack of information hinders market discipline. This study addresses the gap in the literature by empirically examining the impact of board gender diversity on the relationship between banking information disclosure and banking crises in Africa. It specifically investigates three hypotheses: 1. Board gender diversity increases bank disclosure (Wake-up call hypothesis). 2. Board gender diversity and bank disclosure reduce the predicted probability of banking crises (Resource dependency and signalling theories). 3. Board gender diversity enhances the negative impact of bank disclosure on the predicted probability of banking crises (Interaction effect). The study’s significance lies in its focus on the African context, where banking crises occur unpredictably, and governance standards and voluntary disclosure frameworks are evolving.

This study draws upon corporate governance, financial disclosure, and banking crisis theories, primarily the resource dependency theory and the signalling theory. Resource dependency theory emphasizes the board's role in connecting the firm to its external environment, providing counsel and meeting stakeholder needs. Board gender diversity is viewed as a mechanism to enhance corporate governance and disclosure, bringing diverse perspectives and problem-solving skills. Women on boards, due to their communication skills and networks, may enhance the information environment. Prior studies support the positive impact of women on boards on corporate disclosure, linking it to financial performance and banking system stability. Stakeholder theory is also relevant, suggesting that companies with more women on their boards better protect stakeholder interests. Resource dependency theory suggests women's participation enhances closer monitoring, committee involvement, and banking system stability. The signalling theory is employed to explain the relationship between disclosure and banking crises. It suggests that firms with good performance willingly disclose more information to distinguish themselves, thereby reducing crisis probability. While some argue that disclosure improves bank prudence and attracts deposits, others suggest that it may lead to costly misinterpretations, triggering panic and failure. The existing literature presents conflicting views on the impact of disclosure on bank performance, especially during crises. The increasing presence of women in top management necessitates further research into board gender diversity's role in the disclosure-banking crisis nexus, particularly in Africa, given the lack of empirical studies in this specific context. This study adds to the literature by examining board gender diversity's impact on bank disclosure and its moderating role in the relationship between disclosure and banking crises in Africa.

The study uses an unbalanced panel dataset of banks in 42 African countries from 2006 to 2018. It employs various econometric techniques including pooled ordinary least squares (OLS) panel, random effect model, fixed effect model, two-stage least squares (2SLS), dynamic system generalized methods of moments (GMM), and logistic regression models, as appropriate. Country-level macroeconomic data were obtained from the World Development Indicator (WDI), while bank-specific data came from Bankscope and the Global Financial Development Databases. Three models were specified to test the hypotheses: 1. The effect of board gender diversity on bank information disclosure was examined using a regression model (Equation 1). Board gender diversity was measured by the number of women on the board and the presence of women on the board. The model also included control variables such as the Lerner index (competition), ownership structure, GDP per capita, exchange rate, and inflation rate. 2. The independent effect of board gender diversity and bank disclosure on banking crises was analyzed using a panel logistic regression model (Equation 2), where the dependent variable was a dummy variable for banking crises (1 for crisis, 0 otherwise). The model also included the control variables. 3. The interaction effect of board gender diversity and bank disclosure on banking crises was analyzed using a dynamic panel estimation (system-GMM) model (Equation 3) to address endogeneity concerns. This model included the previous year's banking crisis as an explanatory variable. Diagnostic tests were conducted to check for normality, autocorrelation, homoscedasticity, and multicollinearity. The study utilized Pearson’s correlation analysis, Shapiro-Wilk test, and variance inflation factor (VIF) to assess the reliability and validity of the data.

Descriptive statistics reveal an average return on assets (ROA) of 9.1%, an average of one woman on boards, and an average bank disclosure score of 5.19 (on a scale of 0–10). Pearson correlation analysis showed no multicollinearity. The results across various econometric models consistently showed that: 1. The number of women on boards had a negative relationship with bank disclosure, while the square of the number of women on boards had a positive relationship. This suggests a U-shaped non-linear relationship, indicating an optimal number of women on boards to maximize disclosure. 2. The presence of women on boards had a positive impact on bank disclosure, supporting resource dependency theory and stakeholder theory. 3. Both the number of women on boards and bank disclosure independently reduced the probability of banking crises. The negative relationships between board gender diversity and banking crisis probability, and bank disclosure and banking crisis probability, held robustly across various models including the logistic regression, pooled OLS panel, fixed effect, random effect, and 2SLS. 4. The interaction effect revealed that board gender diversity enhanced the negative impact of bank disclosure on the predicted probability of banking crises. The net effect of bank disclosure on banking crisis probability was significantly more negative when women were present on the board, indicating a strong moderating effect. The results were consistent across different model specifications, including System-GMM, which addressed endogeneity concerns. The previous year’s banking crisis significantly predicted the current year’s crisis, highlighting the persistent nature of banking instability.

The findings support the hypotheses, showing that board gender diversity positively influences bank disclosure and negatively affects the probability of banking crises. This aligns with resource dependency and signalling theories, where diverse boards improve information environments and enhance monitoring. Women on boards seem to contribute to more transparent reporting, reducing information asymmetry and increasing market discipline. The significant interaction effect highlights the synergistic relationship between board gender diversity and bank disclosure in mitigating banking crises. The robust results across various econometric models, including the System-GMM which addresses endogeneity, strengthen the study’s findings. The study’s focus on Africa contributes substantially to the existing literature, which largely focuses on developed economies.

This study adds to the literature by providing empirical evidence on the positive impact of board gender diversity on bank disclosure and the subsequent reduction in the likelihood of banking crises in Africa. It emphasizes the importance of optimal levels of both women on boards and bank disclosure to ensure banking stability. Future research could investigate the impact of specific regulatory frameworks and varying levels of economic development on these relationships in other regions.

The study’s cross-sectional nature limits the ability to establish causality conclusively. The data used, while comprehensive for the African context, might not fully capture the nuances of individual banking systems. External shocks and unforeseen events not included in the model could also influence banking crises. While the System-GMM addressed potential endogeneity, other forms of unobserved heterogeneity may exist.