“Corporate governance dynamics in financial institution performance: A panel data analysis”

The study aims to identify the effect of corporate governance factors on financial institution performance in Bangladesh. This study employs annual data for 20 financial institutions, including banks, NBFIs, and insurance companies, data is collected from 2011 to 2022. Here, three corporate governance indicators are utilized – board size, board independence, and director’s ownership. The performance of the financial institutions is measured using return on assets (ROA), return on equity (ROE), and net asset value (NAV). Apart from the corporate governance variables, three company-specific factors, i.e., firm age, financial leverage, and firm size, are used as the control variables. Panel data analysis is conducted through the dynamic Feasible Generalize Least Square (FGLS) method, and the robustness is performed using the random effect model. The results show that corporate governance parameter such as board size has a significant positive influence on financial institution performance in Bangladesh, where board independence and director ownership do not have a significant influence on the performance of financial institutions. Thus, the performance of financial institutions increases when board size increases. This indicates that board members are actively engaged in strategic decision-making and ensure the rights of all stake-holders, which helps improve financial institutions’ overall performance. Therefore, financial institutions may increase their board size to the maximum level to ensure better corporate governance practices in the organizations, which ultimately increases performance.


INTRODUCTION
Good corporate governance is crucial for a company's sustainability, image, and long-term growth.Financial institutions tend to be more volatile and in turmoil and thus require a more critical analysis of corporate governance indicators within the financial institutions than other industries.Corporate governance is widely recognized as a vital determinant of the performance and trustworthiness of financial institutions within the global financial market (Bhagat & Bolton, 2019).In the corporate governance process, corporations are legally bound to work in the stakeholder's interest.It synchronizes the interests and motives of all the business stakeholders, including managers, shareholders, directors, and employees (Demb & Neubauer, 1992).A business entity must be well-governed and controlled for its efficient operation by maintaining transparency, accountability, and predictability so that the concept of corporate governance is established (Cadbury, 2002).
To mitigate the risk and ensure sustainable growth, investors increasingly recognize the role of good corporate governance because there is a dynamic relationship between a firm's performance and corporate governance (Anderson & Campbell, 2004).Researchers contend that ideal board size and a standard form of outsider directors upgrade straightforwardness and responsibility and increase the administration quality of a firm (Adams & Ferreira, 2007).Additionally, the alignment between the interests of investors and directors, as indicated by stock ownership, has the potential impact on the long-term value creation of a firm (Fama & Jensen, 1983).
The financial sector of Bangladesh, with its expanding economic area, presents a convincing field for exploring the connection between firms' stock performance and corporate governance practices.Regardless of the generally perceived idea that in the context of Bangladesh, there is a gap between the interests of the chief's stock possession investors and board members (Ahmed et al., 2016;Rouf, 2012;Deb et al., 2017).Although firm age is commonly used in the USA and other countries (Mester, 1996;Chen, 2012), this variable is quite uncommon in the Bangladeshi context.Stock price measures such as ROA, ROE, and NAV have been explored individually in the context of financial performance (Willim, 2015;Fiador, 2013).Thus, a comprehensive analysis of the Bangladeshi financial sector is required to identify the impact of corporate governance on its performance by utilizing ROA, ROE, and NAV as performance measures.

LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT
Corporate governance practices are continuously enhancing, driven by factors such as the global financial crisis, the rapid expansions of privatizations, and the evolution of financial institutions.Skillful management of the corporate governance process is pivotal in strengthening corporate performance.The improvement of a company's image, reducing the uncertainty of fraud, and increasing shareholders' confidence are possible by practicing good corporate governance (Jesover & Kirkpatrick, 2005 (Madura, 2015).
By regulating financial institutions, corporate governance implies a fair impact on financial performance.As the stock price reflects investors' perceptions of the company's value in the financial markets, the stock price of a company is frequently regarded as a performance indicator (Drobetz et al., 2003).It reflects the market's assessment of the company's overall health, growth potential, and current and future financial prospects.
Numerous studies have investigated the relationship between the performance of financial insti- In a study on director ownership, Bhagat and Bolton (2019) stated in their study that director ownership is significant and positively correlated to bank performance, and a higher percentage of ownership of directors lowers the firm's risk.Andreou et al. (2016) found the opposite scenario, which showed that a higher percentage of directors' ownership raises future stock price crashes.
In addition, the total time of operation of a firm, known as firm age, is a crucial factor that affects firm performance (Chang & Chiu, 2006).Chen (2012) found a strong influence of age on firm performance, while Mester (1996)

RESULTS
The descriptive statistics for the chosen variables are presented in  Notes: ***, **, and * represent the significant level at 0.001, 0.01, and 0.05, respectively.The table shows the results of the FGLS panel data regression model to identify the impact of corporate governance on the performance of financial institutions using annual data from 2011 to 2022 for 20 financial institutions in Bangladesh.Here, columns indicate the performance measures, and rows indicate the corporate governance factors and company-specific factors.The data are transformed using a two-step data normalization method to improve the normality of the data (Templeton, 2011).The company-specific factors are lagged by 1 year to exclude the effect of endogeneity.
Generalized Least Square (FGLS) is applied to generate bias-free results if there is any multicollinearity, heteroscedasticity, and autocorrelation problem in the data.The FGLS is performed for three regression analyses such as ROA, ROE, and NAV.The result of the regression analysis is given in Table 4.
For hypothesis H 1 , overall board size has a significant positive effect on performance.The results show that the coefficients of board size have a statistically significant positive relationship with return on assets (ROA) ( β = 0.001, p < 5%) and net asset value (NAV) ( β = 0.159, p < 1%).However, board size exhibits no significant relationship with return on equity ROE ( β = 0.008, p > 5%).
The The robustness of FGLS regression results has been checked by using the Random Effect Model.(Templeton, 2011).The company-specific factors are lagged by 1 year to exclude the effect of endogeneity.
The findings of the panel random effect model show the same findings that have been identified in FGLS regression at the same significance level.
To analyze the random effect model, it is explored that the analysis of FGLS regression (Table 4) is justified.Table 5 shows the regression output of the random effect model.

DISCUSSION
The In summary, the evidence highlights a mixed impact of corporate governance on the performance of financial institutions.The significant effects of board size on ROA and NAV underscore its importance, while board independence and director ownership do not show a meaningful impact on firm performance.This indicates that in the context of financial institutions in Bangladesh, board size plays a pivotal role in performance outcomes, whereas other governance factors may have less influence.

CONCLUSION
This study examines the influence of corporate governance on financial institution performance in Bangladesh using data from 20 financial institutions spanning the years 2011 to 2022.Both the feasible generalized least square (FGLS) method and the random effects model of panel data analysis exhibit that board size is the only component of corporate governance that demonstrates a statistically significant positive link with the performance of financial institutions in Bangladesh.Nevertheless, board independence and director ownership do not exhibit a statistically significant relationship with financial institution performance.This study suggests that increasing the size of the board could be a wise strategy for enhancing financial institution performance since it could help to get a greater variety of perspectives and specialties in the firm operation.However, the lack of a significant relationship between board independence, director ownership, and performance measures raises the possibility that these aspects of corporate governance may not matter for the performance of financial institutions in Bangladesh.
The study's findings will be helpful to financial institution investors in Bangladesh because investors get specific insight into the important corporate governance factors that influence financial institution performance.Policymakers might also utilize these findings to make operational and management decisions during periods of economic uncertainty in Bangladesh.The study contributes to the existing knowledge by using comprehensive panel data analysis methodologies, challenging the conventional emphasis on board independence and directors' ownership, and emphasizing the importance of board size in maximizing financial performance.
Future studies can design upon the limitations of this research by focusing on several key areas.Firstly, future researchers can extend the sample size beyond 20 financial institutions and also extend the study period study for make the findings more generalized.Secondly, other corporate governance factors, like CEO compensation and board diversity, can be considered to get a more comprehensive understanding of the impact of these variables on financial success.Lastly, an in-depth understanding of the influence of corporate governance factors on financial institutions' performance might be obtained by employing monthly data frequency rather than annual data, which would help to capture short-term variations.
(Ahmed et al., 2016ists of the company's governing board members(Ahmed et al., 2016).Deb et al. (2017) explore that board size significantly correlates with a firm's performance.Cheema and Din (2013) find the same results.On the other hand, Eisenberg et al. (1998) and Willim (2015) note that board size negatively influences the firm's performance.Deb et al. (2017) identify the same results.However, Hasan et al. (2023) explore that board size has no significant relationship with a firm's performance.The research by Ahmed et al. (2016) found the same results.
Rouf (2012).(2023)6) and Financial Innovations, Volume 21, Issue 3, 2024 http://dx.doi.org/10.21511/imfi.21(3).2024.24ityandindependentdirectors.Also, this shows that a firm's performance has a negative relationship with the audit committee and board size.A study byRostami et al. (2016)shows that board independence, CEO duality, CEO tenure, and ownership concentration have a significant correlation with return on assets.Similarly,Kumalasariand Pratikto (2018) conducted another study showing that return on assets has a significant relationship with corporate governance factors.Contrarily, findings by Deb et al. (2017) demonstrate that within the context of their study, there is no statistically meaningful link between the board size and the firm's performance.Similarly, Eisenberg et al. (1998) found the same results.In the Ghanaian market, Net asset value per share holds significant esteem, particularly when the CEO serves as a board member, or the firm has a small board size.If there are no non-executive directors, which indicates the board's independence, Investment Management and Financial Innovations, Volume 21, Issue 3, 2024 http://dx.doi.org/10.21511/imfi.21(3).2024.24itindicates a negative impact on the market value of shares(Fiador, 2013).A study byHasan et al. (2023)shows that board independence is the only corporate governance factor significantly correlated with firms' performance.Rouf (2012)verified the same results.Rostami et al. (2016) and Deb et al. (2017) also found that board independence significantly correlates with ROA.Fiador (2013) found that board independence negatively influences net asset value.Erkens et al. (2012) consider that higher board independence leads to firm failure.

Table 1 .
Variable descriptionNet asset value (NAV) per share reflects the equity position of each shareholder.It is defined as the total assets minus the total liabilities of the firm in the period divided by the number of shares outstanding Reed (2015)-3; foreign commercial banks -9; non-scheduled banks -6; non-bank financial institutions -43, and 83 insurance compa-Investment Management and Financial Innovations, Volume 21, Issue 3, 2024 http://dx.doi.org/10.21511/imfi.21(3).2024.24nies(BangladeshBank 1 , Bangladesh Securities and Exchange Commission 2 ).Among these, ROA, ROE and NAV), β is the coefficient, t is the time, i is the unit of firm.BI indicates the board's independence, BS indicates the board size, and DO indicates the director's ownership, which are independent variables.Controls indicate the control variables such as firm age (FA), firm size (FS) and financial leverage (LEV), and t-1 indicates the lag for period 1 because all company-specific variables are lagged by one year to mitigate potential endogeneity problems, according toReed (2015).

Table 2 .
Descriptive statistics Notes:The table provides a detailed snapshot of the descriptive statistics encompassing all variables, drawing from data spanning the years 2011 through 2022.Here, ROA, ROE, and NAV are performance measures used as dependent variables, BS, BI, and DO are corporate governance factors used as independent variables, and FA, LEV and FS are company-specific factors used as control variables.Data are normalized by applying the two-step data normalization method inverse document frequency (IDF).

Table 3 .
Correlation matrixNotes: ***, **, and * indicate the level of significance at 0.001, 0.01, and 0.05, respectively.The correlation matrix shows the relationship between the dependent and independent variables.Here, ROA, NAV, and ROE indicate the return on assets, net asset value, and return on equity.Also, BS is board size, BI is board independence, DO is director's ownership, FA is firm age, FS is firm size, and LEV is leverage.

Table 4 .
Regression results of impact of corporate governance on financial institutions' performance (feasible generalized least squares (FGLS) regression model) results corresponding to hypothesis H 2 show that board independence does not have a significant effect on performance measures.Results show that the coefficients related to board independence with ROA, ROE, and NAV are positive but not significant at the 5% level.

Table 5 .
Regression results of impact of corporate governance on financial institutions' performance (random effects model regression) The table shows the results of the Random Effect Model of panel data analysis to identify the impact of corporate governance on the performance of financial institutions using annual data from 2011 to 2022 for 20 financial institutions in Bangladesh.Here, columns indicate the performance measures, and rows indicate the corporate governance factors and company-specific factors.The data are transformed using a two-step data normalization method to improve the normality of the data Investment Management and Financial Innovations, Volume 21,Issue 3, 2024 24tp://dx.doi.org/10.21511/imfi.21(3).2024.24