Effect of board quality on the nancial performance of conventional and Islamic banks in a stable nancial context: Comparative study on the international evidence


 Returning to the literature of finance and banking governance, our article provides the first logical analysis that detailed the process of comparative analysis between the correlation of board determinants’ quality and the financial performance of conventional and Islamic banks. Previous research has always discussed the main role of the board as an internal mechanism of governance on the financial performance separately in each bank type. However, we have never encountered rewarding studies that compared these impacts. In our study, we distinguished between the impact of the board of directors on the financial performance in conventional and Islamic banks. Settings of the financial performance and board of directors of the conventional and Islamic banks are collected from 30 countries located in four continents: America, Europe, Asia, and Africa. Two equal samples were collected that each of them is composed of 112 banks. By using the GLS method, data were used to explore the impact of the board of directors on the financial performance between both types of banks over the period 2010-2018, giving us 1008 bank-year observations in each sub-sample. On the whole, empirical results have shown that in conventional banks the board of directors has negatively affected the financial performance, while the impact of the board on the financial performance of Islamic banks is ambiguous. Nevertheless, the degree of the positive impact on financial performance is more significant in Islamic banks.

institutions of all types. In other words, this study attempts to compare the degree of the impact of the board on IBs' FP of with the relative impact of its conventional competitors in order to answer the following question: in what type of banks, the impact of board quality on FP is more meaningful?
We synthesized the contributions of our research into two points. The rst contribution is that our results can be a valuable source of knowledge for investors, policymakers, funders, economic agents, and regulators, especially in the nancial services sector and in the development of nancial objectives and planning of strategic plans for boards' control. The second contribution is that we gave the board its real vital importance in each bank type and we revealed its insu cient role on the FP of large conventional and Islamic banks, future research can address the implications of board quality for monetary policy in some developed, emerging and developing countries.
The remainder of the paper is organized as follows. Section 2 reviews previous research on this topic and the hypotheses of the study. Section 3 describes the methodology which is employed, the data sources, variables, and the models' speci cations. Empirical results and discussion are presented in section 4. Finally, section 5 contains concluding remarks.

Literature Review: Board Determinants
The BOD is at the center of a banks' governance system. Since the board is the most in uential governance mechanism, in this section we looked at the impact of the board quality on the FP of conventional and Islamic banks. We examined whether the characteristics of the BOD affected the banks' FP in a period of nancial stability. Our objective is to compare the correlation between the board of Islamic and conventional banks on FP's indicators.
In our study, we insisted on four characteristics of a board composition which are board size, independent directors on board, separation of board chairman and CEO roles, and board meeting. Our choice is justi ed by two reasons; rstly, because of the availability of observations relating to these variables, secondly, the variables used are the most widespread and the most frequently used in the literature (Mollah and Zaman, 2015;Gebba and Aboelmaged, 2016;Vallascas et al., 2017;Farag et al., 2018;García-Ramos and Díaz, 2019;Naciti, 2019;Chen et al., 2020;Tran and Turkiela, 2020).
The formulation of our hypotheses is based on the results of previous research in the direction of stakeholder theory.

Board size
According to the literature review, the size of the board is an important characteristic to obtain an optimal governance structure (Tulung and Ramdani, 2018 andPaniagua et al., 2018). Large boards could include more representative directors in terms of quantity and quality. Multiple choices give more opportunities for senior managers to choose directors with more experience and specialized expertise (Williams et al., 2005;Kent and Stewart, 2008;Ntim et al., 2012). The board size improves governance reporting and is expected to impact the director's quality oversight (Ntim et al., 2012). As already noted, some studies have found that companies with large boards have a positive in uence on FP, especially companies with complex and large businesses operating in multiple segments (Coles et al., 2008). Besides, small boards lead to the creation of con icting relationships, either by aligning the interests of executives with those of shareholders or by aligning the interests of executives and directors.
In the same vein, Dalton et al. (1999) discovered a positive and signi cant relationship between the board size and the banks' FP. The signi cant effect added by the Big boards is due to the presence of agency problems, to the con icts of interest, to the communication challenges, to the coordination failures, to the transmission of accounting problems, to the transfer of nancial information, and the decision-making. In addition, Adams and Mehran (2012) carried out a study that aims to analyze the relationship between the BOD determinants and the FP based on a sample of 35 US banks listed throughout the period 1986-1999. In conclusion, they stated that the board size is positively correlated with FP. Similarly, Adams and Mehran (2003) completed a study aimed at verifying the impact of board size on banks' performance. They found that when the banks' board size is large, banks have high levels of performance, implying that they are necessarily associated with higher risks than banks with small boards. As a result, if the board size is small, members can easily misappropriate their own interests through manipulations, and directors are in uenced by their superiors.
In the case of the Middle East, other academics stipulate that the small board easily suffers the in uence of the leaders more than the large one, because it has a variety of experiences belonging to the different administrators (Gary and Gleason, 1999). Along the same lines, Naushad and Malik (2015) tested the impact of board size on FP in 24 banks covering the GCC region over the period 2012-2013. The results revealed that small-sized boards are signi cantly capable of monitoring the management of banks in the CCG region.
In Asia, by performing a panel study on a sample of Chinese and Indian banks listed during the Subprime nancial crisis, Battaglia and Gallo (2015) investigated the association between some governance mechanisms, risk management and the banks' performance in China and India during the period 2007-2011. They noted that banks with a large BOD are governed by a positive and signi cant impact on the FP indicators. Also, Yung (2009) analyzed the impact of banking governance arrangements in Hong Kong on bank performance. He monitored the governance quality by both the size and the composition of the board. He concluded that banks with large boards and low loan levels between related parties tend to perform better. To protect the shareholders' rights and to ensure that managers cannot defend their interests to the detriment of shareholders, a correlation between loans and related parties has been established as a parameter for valuing the governance system e ciency. The higher the level of loans granted, the more the bank exports a bad reputation to the banking market. This consequence can damage the degree of effectiveness of its existing governance system. As a result, this affects its FP because banks facilitate the ine cient allocation of resources. This nding re ects the importance of governance in improving banks' FP.
Other studies have invested in determining the optimal board size; they have found that there is a non-linear effect between board size and bank performance (Andres and Vallelado, 2008). As for Booth et al. (2002), their empirical results have shown that banks with large boards perform better than their counterparts with reduced boards. They suggested that the average size of banks' boards is 16 directors. This current research argued that a small directors' number is easily in uenced by the CEO. Moreover, Belkhir (2009) conducted a study to determine the preferable number of BOD, so that the bank can deliver the expected return. He showed that when the board size extends to about 19 members, performance increases. At the limit of this threshold, increasing the board size results in performance growth. Beyond 19 members, the performance decreases. While there is some debate about whether banks should be large or small, some other studies have suggested that institutions with more complex operations should have more members in their BOD (particularly banks). They veri ed that the board size has a positive effect on FP (Coles et al., 2008;Bassem, 2009). Within a comparative framework between banks, Wasiuzzaman and Gunasegavan (2013) conducted a comparative study between the performance of Islamic and conventional banks operating in Malaysia during the period 2005-2009. They found that the assets' pro tability, the board size, and the CBs' sizes are higher than those of the IBs.
After a rich literature exposure, we expected that a large board could ameliorate the parameters of pro tability, e ciency, liquidity, and solvency of the banks' board. For this reason, we have proposed a theoretical suggestion that con rms that the board size has a strictly positive impact on the banks' FP: Hypothesis 1: The board size has a positive effect on the FP of conventional and Islamic banks.

Rooting of the board chairman or the CEO / Chairman duality
Returning to the literature, good governance asserts the real separation between the positions of CEO and chairman to limit the power of the chairman (Daily and Dalton, 2015;Chang et al., 2018;Wang et al., 2019). As other studies show that cards with a very centralized power structure lead to a very volatile and highly variable FP. For example, Adams et al. (2005) revealed that the more powerful the CEOs, the more they lead to greater variability in company FP. In this topic, previous research has attempted to demonstrate the effectiveness of the board and its impact on the CBs' performance. However, in IBs the subject of rooting is not yet widely treated. This may be due to the weakness of the board impact or because of its limited power as a mechanism of governance in this banks' type. Theoretically, rooting means the occupation of the same post by a manager after the end of their rst xed-term contract. It manifests itself in two methodical forms leading to the same results. Either through the duality of the CEO and that of the board chairman by the same person. Either the same person in which designation as a board chairman will exceed his rst contract or he automatically renewing his mandate without verifying the conditions of his independence.
The separation of roles means higher board oversight and lower CEO power, while duality means lower board oversight and stronger CEO power (Krause et al., 2017 andChang et al., 2018). Nevertheless, If the CEO also holds the position the board chairman, monitoring and control processes cannot be effective, precise, and useful to monitor agent actions (Allegrini andGreco, 2013 andHussain et al., 2018). In this way, Silveira and Barros (2013) stated that if the CEO or chairman has strong power, they can get out of control, adopt reckless policies, and left unchecked by a board of directors which can lead to poorer performance. The same, Naushad and Malik (2015) examined the effect of the CEO duality on FP of 24 banks selected from the GCC zone for two years 2012 and 2013. The results revealed that the CEO, who also serves as the board chairman, is likely to improve the banks' FP. Owners of blocks have a positive and signi cant effect on the banks' performance. Duality can have important advantages for institutions, if the CEO is also the chairman, he will make his knowledge and experiences available to other administrators, which will allow them to produce more resources (Adams and Ferreira, 2007). Indeed, Al-Hawary (2011) studied the association between some mechanisms of banking governance measured by the ownership concentration, the majority shareholding, the capital adequacy ratio, the board size, the presence of external directors and the duality between the CEO and the board chairman on the performance of Jordanian banks during the period 2002-2009. Results showed that duality and external directors have a positive and statistically signi cant effect on Tobin's Q. In this way, he deduced that banks in Jordan need to arrange good quality of governance to positively and effectively in uence banking performance.
Based on the majority of research highlighted in this theme, we have predicted the presence of a positive correlation between the CEO rooting and the banks' FP (Donaldson and Davis, 1994;Kota and Tomar, 2010).
Hypothesis 2: There is a positive relationship between the rooting of the board chairman and the FP of conventional and Islamic banks.

Board Independence
The independence of directors is considered to be the most important measure of board effectiveness (Aggarwal et al., 2008;Jackling and Johl, 2009;Bruno and Claessens, 2010;Pombo and Gutiérrez, 2011;Liu et al., 2015). Are considered as independent directors all members who do not have any a liation with the bank, except for their membership in the board as directors (Brown et al., 2011). A non-executive board member only has a business relationship with the bank director who is quali ed by his/her integrity and expertise to protect the interests of all stakeholders (Young, 2000). This director type should not be a current employee, a former employee of the bank, or a member close to the top of the hierarchy and has no signi cant business relationship with the bank (Pathan, 2009). Previous research supposed dependent all directors having a previous executive function, familiar with the bank manager, or have other business links with the bank (Aebi et al., 2012). According to Cotter et al. (1997), the board is assumed to be independent when the number of independent directors within it exceeds 50% of the total members. Some studies have found that the presence of independent directors among board members is not associated with any effect on FP. To test this observation Yermack (1996); Bhagat and Black (2002) and Hermalin and Weisbach (2003) have pointed out the absence of a signi cant relationship between the quali ed composition of the board, which has been measured by the directors' independence degree, and the FP. They argued that the lack of a signi cant correlation between board composition and FP is evident. This relationship is generally consistent with the relative economic situation and taken into consideration during a nancial crisis. Other studies have associated market measures to detect the independent directors' impact on the FP. They found no signi cance between the composition of the board and the FP (Hermalin and Weisbach, 1991;Bhagat and Black, 2002). For this reason, Gebba and Aboelmaged (2016) indicated that a typical board should contain a heterogeneous mix of executive and nonexecutive directors. The revised Malaysian Corporate Governance Code (CGCM, 2007) also provided that all publicly traded companies should have a BOD, which should have a balance between executive and non-executive directors to avoid decisions dominated by one group or by some board members (Kallamu, 2016).
However, several researchers have adopted another discordant explanatory approach which stipulates that external directors are not able to understand the complexity of the banks' activities. They considered outside directors to be agents with insu cient expertise to carry out their stakeholder control missions, to detect the opportunistic behavior of the managers, and to monitor the overruns against the sense of increased performance. Besides, due to stakeholder theory, the independence of BOD can be negatively associated with the sustainability of banks' FP, as they have less information about the situation of the bank, hence they will be those most subject to pressure from shareholders. As a result, a board composed of a signi cant portion of independent directors may lose the ability to oversee management and protect the interests of shareholders and stakeholders. Moreover, within this opposite argument, some researchers have found that the presence of foreign directors on the board has a negative and signi cant effect on banks' future FP (Adnan et al., 2011). In the same topic, Coleman and Biekpe (2006) examined the effect of the board composition on the nancial decisions of 47 companies listed on the Nairobi Stock Exchange over the period 1995-2004. They found that larger rms were more in debt. However, short-term debt is negatively correlated with board independence. This explains why this type of directors is very likely to make consecutive and massive debt decisions to exploit them for purposes prejudicial to the bank interest, which con rms the presence of opportunistic behaviors. Even more, Minton, et al. (2010) investigated the correlation between risk-taking, the performance of US banks, and the board directors' independence in nancial crises. The results indicated that board independence is positively related to risk-taking and bank performance before the crisis, but it has shown a poor impact on banks' performance during the crisis. Also, Michelon and Parbonetti (2012) and Allegrini and Greco (2013) found no relationship between the number of independent directors and the disclosure of information on sustainable development. Furthermore, independent board directors may receive misleading and manipulated nancial information from other executives (Kravet and Volkan, 2013).
Independent directors with con icting interests lead to poor governance practices, as the situation is favorable for the appearance of interest between the BOD and managers (Hermalin and Weisbach, 1991) leading to a decline in performance (Yermack, 1996;Bhagat, and Black, 1999;Ghazali, 2010). They observed a negative relationship between the proportion of independent directors on the BOD and the performance per share. In the same research path, Erkens et al. (2012) studied the association between banks' corporate governance and FP in 30 countries in the Subprime Crisis period 2007-2008. They found that board independence and ownership concentration are associated with poor rates of pro tability during the crisis. They argue that the independent boards raised a lot of equity during the crisis, banks with a high concentration of property took more risks before the crisis, the risks translate into signi cant losses for shareholders during the crisis, the wealth of existing shareholders was transferred to the creditors thereafter. Similarly, Adams (2012) showed that banks with more independent members on their boards performed poorly. Also, Minton et al. (2010) and Hoque and Muradoglu (2015) found that the nancial expertise of the independent directors of commercial banks is negatively related to the variations of their values. They went through nancial troubles which led to a decrease in the banks' performance. Moreover, Mollah and Zaman (2015) analyzed the effect of the governance effectiveness on performance in conventional and Islamic banks during the period 2005-2011. They found that in IBs, the size and independence of the board signi cantly and negatively in uence their FP. Besides, they noted that the duality and the internal recruitment of the CEO have a negative and signi cant impact on the IBs' FP. Empirically, they have proven that the Sharia Committee contributes positively and signi cantly to the IBs' FP during a supervision mission. This impact becomes negligible when it is an advisory role. Even more, Bansal et al. (2018) put forward another explanation for this negative impact, they demonstrated that the behavior of independent directors can be in uenced by reputational risk during CSR disclosure.
In the Asian context, referring to more in-depth research, Kallamu (2016) studied the moderating role of independent directors in the interaction between the ownership structure and the FP of Malaysian banks between 2007 and 2011. He noted that the existence of independent directors among board members adversely affects FP. In Malaysian banks, the majority of owned directors have mixed and confused their interests with the interests of other shareholders. On the contrary, the presence of independent directors on the BOD constitutes a positive and signi cant moderating effect on the relationship between the directors' ownership and the FP. This re ects that independent directors positively moderate the relationship between directors' ownership and improved the FP of the institution, as there will be an alignment between directors' and shareholders' interests. In conclusion, he stated that the independent directors exercise a guiding force on the banks' FP.
Therefore, based on previous selective studies, we formulated our third research hypothesis in the following form: Hypothesis 3: There is a negative relationship between the proportion of the board's independent directors and the FP of conventional and Islamic banks.

Number of meetings held by the board of directors
Based on the literature review, several studies have identi ed the importance of the frequency of board meetings as a mechanism able to in uence the governance quality in different contexts (Sánchez, 2010;Choi and Lai, 2014;Thu et al., 2016) or also as a control parameter of the FP (EL-Maude et al., 2018). The effect added by this governance mechanism has led us to distinguish two groups of previous studies. Although many studies discussed the effect of the meetings' number held by the CBs' board, the studies discussing the effect of the board meetings' number on FP in Islamic banking are almost non-existent.
In contrast, an intermediate stream has established coordination among governance mechanisms to determine whether the quality of one mechanism affects or enhances the quality of the other. These researchers found no correlation. In a recent study, Thu et al. (2016) have indicated that the performance of commercial banks in Vietnam does not depend on the board meetings' frequency.
Nevertheless, some studies have found that there is a negative association between the governance quality and the meetings' number held by the BOD. In this sense, Sánchez (2010) analyzed the impact of some governance determinants on the governance quality in Spain banks over the period [2004][2005][2006]. He studied the in uence of board characteristics, including board size, independence, and diversity of mechanisms on the governance quality. He found that board activity is limited by a small number of annual meetings.
Moreover, the frequency of meetings has been considered by many authors as empirical evidence that slows down the FP of institutions. As stated by Vafeas (1999), one can expect a negative relationship between board meetings and FP. If outside directors do not take advantage of the time, they spend together to exchange meaningful ideas with or among directors, the frequency of meetings becomes unnecessary (Lin et al., 2014). Besides, stakeholders cannot wait for the board members' meetings to address emergency issues, prepare strategic programs, monitor stakeholder interests or make effective decisions since they do not have su cient time at meetings to discuss all the alternatives (Huse, 2009). Also, Jensen (1993) indicated that routine tasks absorb the time of board meetings, and reduce the quality of board control to leaders. Even more, considering the time costs, travel costs, and bene ts paid to directors because of meetings, the high frequency of meetings is not bene cial and probably a response to poor FP (Lin et al., 2014). Furthermore, Choi and Lai (2014) analyzed the effect of bank governance on the FP of Asian banks during the period (2007-2012) by using multiple linear regression analyses. The system of bank governance has been assessed by three characteristics, a variable re ects the governance quality, a variable measures the board size, and a variable symbolizes the frequency of board meetings. The ndings indicated that there is a negative and statistically signi cant relationship between assets quality and board meetings.
We have drawn from the foregoing the following hypothesis: Hypothesis 4: There is a negative relationship between the frequency of board meetings and the FP of conventional and Islamic banks.

Empirical Method: Distinction Between The Effect Of The Board Quality On The Financial Performance Of The Islamic And Conventional Banks
To investigate the relationship between the board quality and the FP in the two banks' types, we employ the multivariate regression framework using the GLS method. This method is the best which allows to concretize the comparison between the individual effects and to show the variability between these effects. Moreover, this method is the most suitable for drawing detailed conclusions about the effect of each characteristic of counseling on each measure of FP. Indeed, we have worked on panel data with which the chosen method is complemented by also indicating that it is the most appropriate with the general approach of our study.

Methodological aspects
The methodology applied in our exploratory study is a demonstrative comparison by resorting to modeling. This helps to identify the impact factors affecting the relationships between basic FP measures and the in uences due to the BOD. The research plan to be followed to answer the questions already mentioned began with the clari cation of the data sources, then we quoted the variables to be modeled, nally, we exposed our models.

Data collection
Two samples were taken from two reference populations. The choice of banks is limited to countries whose banking systems incorporate both Islamic and conventional banks over the period 2010-2018 regardless of the proportion of each model in each country's banking market. These populations are made up of 2,974 conventional nancial institutions and 683 Islamic nancial institutions. The countries part of our study are USA, France, Singapore, Algeria, Thailand, India, Egypt, Bangladesh, Indonesia, Pakistan, Tunisia, Malaysia, Canada, Sudan, Turkey, United Kingdom, Luxembourg, Bahrain, Jordan, Kuwait, Oman, Qatar, Saudi Arabia, United Arab Emirates, South Africa, Senegal, Nigeria, Sri Lanka, Kazakhstan, and Lebanon. However, we have excluded all speci c nancial institutions subject to particular regulations. The tested samples include only purely conventional or Islamic banks. Besides, due to di culties in collecting information on FP and BOD, we excluded banks marked by some missing observations, variables, or data. We also removed the multi-type mutated banks (Islamic-conventional window banks and conventional-Islamic window banks). These three conditions led us to eliminate 2862 conventional nancial institutions and 571 Islamic nancial institutions. Subsequently, we have reduced the banks' number remaining for each bank type based on qualitative and quantitative ltering criteria (samples equality, activity type, similarity of origin country, bank width), each CB has its Islamic equivalence in terms of capital and size taken from the same country. This restriction reduced the size of our samples to 112 banks each. Finally, after several elimination and deletion steps, we obtained two pairs of equal samples (n1 = n2).
3.1.2. The measurement of the variables to be tested 3.1.2.1. Endogenous variables In this sub-section, we presented FP measures. The main variable to explain was represented by four dependent variables: pro tability, e ciency, liquidity, and solvency. Table 1 shows the parameters we worked on, the symbols, and the relative reports.

Exogenous variables
Throughout the remaining part of this work, banks' FP is explained by four determinants of BOD. Referring to the review of the previous literature, the predominantly independent variables have been described in Table 2 as follows: 3.1.2.3. Measurements of control variables Table 3 displays the list of control variables supported by some previous studies that employed the same variables and their measures.

Presentation of models to estimate
Before proceeding to the estimations, it is necessary to present the typical models to reassess several times the FP and each time the dependent variable will be changed according to the FP measures and the bank type.
Conventional models of multiple regressions are of the following form: Model n1: Association between CBs' pro tability and board quality: LnRtc= +LnTCONSc+ENRADIRc+LnINDCONSc+LnREUCONSc+

Econometric validation of models
The FP of conventional and Islamic banks depends on the systematic use of explanatory variables. Interdependence between the board determinants forced us to test also the correlation between the explanatory variables one by one. If the probability associated with the test is below the tolerance threshold (5%), we reject the hypothesis of heteroscedasticity (H0). Nevertheless, if the probability is greater than (5%), the null hypothesis will be retained, it is possible in this case to con rm the heteroscedasticity of the residues.
Before starting the analysis of the results, rst, we referred to the agreement of choice between the applied tests. Depending on whether it is xed or random-effect, for models that are con rmed xed, such as those a liated with the e ciency and solvency of IBs (Table 5), we compared these models to the Modi ed Wald test. Nevertheless, in the case of models relating to the pro tability, the e ciency, the liquidity, and the solvency of CBs (Table 4) and the models speci c to the pro tability and liquidity of IBs (Table 5), the Breusch-Pagan test proves more appropriate to detect heteroscedasticity. Table 4 illustrates the heteroscedasticity test results of CBs for random-effect models (LnRtc, Etc, Ltc, and LnStc). For this effect type, the randomeffects models have highlighted the need for the Breusch-Pagan test. We speak of heteroscedasticity when the magnitude of the error risk is constant over time. The χ 2 statistics of these models showed risks associated with the rejection of the heteroscedasticity hypothesis well below the desired threshold of 5%. The probabilities of χ 2 test of banks' pro tability, e ciency, liquidity, and solvency models are lower than the minimum risk acceptance rate of the null hypothesis, leading to the rejection of the proposal of heteroscedasticity problems at the level of these models. Therefore, we concluded that the CBs' models are not heteroscedastic.
Similarly, Table 5 showed the heteroscedasticity tests' results of the IBs sample for xed-effects models (Eti and LnSti) as well as for random-effects models (LnRti and Lti). Furthermore, the Modi ed Wald test gives a probability of the χ 2 lower than the predetermined threshold of 5%, the xedeffects Eti and LnSti models recorded values of χ 2 equal to (0.0000). Therefore, we rejected the heteroscedasticity problems hypothesis, the variance of errors in these two models is not the same for all IBs.
Moreover, results of the Breusch-Pagan test revealed χ 2 statistics below 5% for the models assumed until now random, more speci cally, the probabilities of the models relating to the pro tability and the liquidity of IBs equal respectively to (0.0006) and (0.0000). Therefore, we have been able to reject the null hypothesis, LnRti and Lti models are not heteroscedastic and the variance of their errors is not the same for all IBs. As well as we have deduced that our data do not have a heteroscedastic structure.
In the case of xed-effects models, whatever the model is heteroscedastic or not, if it does not contain individual effects, reasoning procedure requires direct navigation to analyze the correlation. However, with a random-effect model, we must check before, if the square of the residuals can be explained by the model variables. In this case, we con rm that there is a problem of heteroscedasticity. The choice of one test or another depends on the need, the type of variables, and the econometric effect of the models.

Residual autocorrelation test
The autocorrelation of residues test is used to detect for each individual in the sample whether the errors of a period are in uenced by the errors of the previous period. The autocorrelation hypothesis of the residues is a necessary condition before the validation of the estimated results. It also facilitates the choice of the best modeling method that lends itself to our data after the classi cation of existing effects.
There are several autocorrelation errors' tests, among which we have chosen the Wooldridge test and the Durbin-Watson test. Each of them is employed in a well-de ned econometric situation. In our research, we have insisted on the most often known: the Wooldridge test is generally the most suitable when the model is subjected to a xed-effect, however, the Durbin-Watson test is the most applicable in the case of random-effects models because of its ability to detect particular forms of autocorrelation.
The hypotheses of the autocorrelation test are as follows: H0: The errors are not autocorrelated.
The decision rule of this test is as follows: if the risk obtained is greater than the critical value (Prob> F) < (5%), we reject the null hypothesis, in this case, the errors of the individuals (conventional or Islamic bank) are considered autocorrelated and vice versa.
The results of the autocorrelation tests for residues are shown in Tables 6 and 7 below: For the models provisionally classi ed with xed-effects, the (Wooldridge, 2002) test results referring to the e ciency and the solvency of IBs (Table  7) have shown two values greater than 5%. They are illustrated with respective risks equal to (0.000) for Eti and LnStc. Since the risk of rejecting the null hypothesis is not high, therefore, it was rejected to con rm that the errors are autocorrelated. Based on these results, we concluded that their errors are autocorrelated, hence, we have recti ed the class of these two models to random-effects.
After the Durbin Watson test, Fisher statistics on the pro tability and the solvency models of CBs (Table 6) and the liquidity model of IBs (Table 7) generated p-values strictly greater than 5%, the risks being respectively equal to (0.0841) for LnRtc, (0.8583) for LnStc and (0.5246) for Lti. For this reason, we con rmed the absence of autocorrelation of errors, which indicates that in these models, the errors are dependent on each other. Therefore, these models have purely random-effects. Furthermore, for the speci c model to the e ciency and the liquidity of CBs and the pro tability of IBs recorded risks of less than 5%, equal to (0.0003) for Etc, (0.0000) for Ltc, and (0.0338) for LnRti respectively. The Durbin Watson test displayed a rate of risk null, which indicates the presence of autocorrelation between the errors. Hence, we have recti ed the class of these two models to xed-effects.
Given these results, 3 nal models took the form of heterogeneous panels with xed-effects that vary from one individual to another (Etc, Ltc, and LnRti) and 5 models took the nal form of heterogeneous panels with random-effects (Etc, LnStc, Eti, Lti, and LnSti). The second models' types assume that the relationships between the FP measurement of conventional or Islamic banks and the ownership structure determinants in each sample are not identical for all individuals of the same sample.
Apart from the distinguishing features and differences between the organizational structures of Islamic and conventional banks, there are nancial peculiarities that separate each banking model from the other and prevent the merger and proximity of the two models (Nganga, 2013). Each CB in our rst sample is governed by its own effect regardless of individual temperaments, con icts of individual interests, internal or external environmental factors of the bank, location of governance mechanisms, regions, etc. Similarly, identifying the effects associated with IBs implies that each of them is governed by a varying effect among banks, but it is unchangeable over time.

Veri cation of multicollinearity problems
The multicollinearity test is performed to prevent the instability risk of the coe cients estimated by the GLS method. It also makes it possible to see if the matrix of the exogenous variables is regular. Any linear regression calls for the presence of collinearity and multicollinearity problems will be integrated into the same model, between the exogenous variables.
To verify the degree of correlation of the independent variables between them, when the majority of the explanatory variables did not satisfy the normality condition, we examined Spearman's correlation coe cients, the non-parametric version of Pearson (Ricco, 2015). The solution consists in eliminating the collinear variable or the block of exogenous variables containing the same types of information, affecting the quality of the regression. The correlation that deviates from the signi cance level of the Spearman test meaning is considered useless in the model. Tables 8 and 9 illustrate the matrices of Spearman's correlation coe cients measuring the degree of linear connection, on the one hand, between the variables to be explained and the explanatory variables, and on the other hand, between the explanatory variables between them. Table 8 displayed the CBs' Spearman matrix, whereas Table 9 displayed the IBs' matrix. Each coe cient is between -1 and +1, the sign takes into account the meaning of the relationship.
There is collinearity between two independent variables when the correlation between them is greater than 0.8 (Gujarati, 2004). The appearance of collinearity makes the signs and values of the estimated coe cients seem contradictory; in this case, their variances are unstable. Since collinearity leads to redundant estimates and misleading signi cance, it must be detected by performing the necessary corrections and treatments before completing the analysis of the results.
Correlation measures the intensity of the relationship between variables. The link strength between the variables is classi ed into three types, there is a strong, medium, or weak correlation. It is also positive if the correlation coe cient is positive and negative if this coe cient is negative. From Table  8, analysis of the Spearman correlation matrix of the CBs' sample revealed that the signs of the explanatory variables on the FP vary from one model to another. As a result, the separate interpretation of each model gave us a clear idea of the individual effect of each variable.
Beginning with the CBs' pro tability (LnRtc), the matrix conceded that (LnAGc) and (LnTAc) have acted positively on the CBs' pro tability. This leads us to argue that the more the effects of the cited variables develop, the better the CBs' pro tability will progressively improve. However, (LnTCONSc), (LnREUCONSc), and (LnINFc) are negatively affected the CBs' pro tability. This led us to conclude that the greater the impacts of these variables are, the more the CBs' pro tability deteriorates.
The shift to e ciency analysis (Etc) revealed that only variables related to (TYc), (LnAGc), and (LnINFc) are positively correlated with the CBs' e ciency. This has indicated that as the value of these variables grows, more CBs will become more e cient. Nevertheless, (LnTCONSc), (ENRADIRc), (LnINDCONSc), (LnREUCONSc), and (LnTAc) re ected the negative and destructive impacts of the bank e ciency.
Concerning the liquidity (Ltc), the correlation matrix arrangement found that (LnREUCONSc) and (LnAGc) are positively correlated with the CBs' availability. This summarizes that the more the tendency of these variables is evaluated, the more CBs increase their wealth. However, other variables demonstrated the opposite, an inverse impact was generated by (ENRADIRc), (TYc), (LnTAc), and (LnINFc), they generally recorded negative effects on bank performance and speci cally their liquidity powers. This explains why any improvements in these variables lead to the weakening of bank liquidity.
Similarly, after an inventory of solvency (LnStc), we attributed to the conclusion that (LnREUCONSc) supported the CBs' solvency. The more this variable has protected the CBs' solvency, the more its effects are propagated within the decision-making nodes, the more they guarantee its solvency capabilities. However, (LnTCONSc), (TYc), (LnAGc), (LnTAc), and (LnINFc) harmed the sustainability of the bank solvency, therefore, the evolution of the impact associated with these variables contributes, no doubt, to the deterioration of the CBs' solvency.
Finally, the analysis of the correlation between the explanatory variables did not reveal any inter-variable correlation coe cient greater than 0.8. For this, we admitted the absence of collinearity problems in all the models reasoning the board quality's impacts on the FP parameters.
Based on Table 9, an overview of the Spearman correlation matrix that collects with the IBs sample showed a heterogeneous mixture of the various signs, sometimes the variables' signs coincide with those detected by its conventional counterparts and sometimes they diverge. In what follows, we have listed the conclusions obtained on the effect of each variable on each FP measure.
First, the analysis of the IBs' pro tability (LnRti) highlighted the presence of three variables playing an important role in the process of creating pro tability; (LnINDCONSi), (TYi), and (LnAGi). However, (LnTCONSi), (LnREUCONSi), (LnTAi), and (LnINFi) revealed an opposite effect, the more the impacts of these variables grow, the more the IBs' pro tability decreases.
Then, the correlation coe cients relative to the IBs' e ciency (Eti) signed a very important cumulation of the positive effects generated by some variables mainly due to (LnTCONSi), (ENRADIRi), (LnAGi), and (LnTAi). This has proven that the more the effect of its organizations extends, the more IBs improve its short-term returns, and the more they guarantee its sustainable returns.
Turning to the impact of the explanatory variables on the IBs' liquidity (Lti), we pointed out that (ENRADIRi), (LnINDCONSi), and (LnTAi) played a favorable role of protecting their liquidity capacity. These mechanisms have all been factors for improving monetary dependence. On the contrary, (LnTCONSi), (LnAGi), and (LnINFi) have an adverse effect on the effectiveness of IBs' liquidity.
Finally, we looked at the dependency analysis between solvency (LnSti) and another explanatory variable, the Spearman test revealed a positive correlation between (LnINDCONSi) and IBs' solvency. In contrast, Spearman matrix testi ed to the presence of a considerable but suspicious cumulative effect on the part of (LnTCONSi) and (LnINFi). This led us to conclude that the more signi cant its effects, the more IBs will keep its solvencies.
The correlation coe cients between the independent variables of the 2nd matrix revealed that the intersection between all the variables did not generate any correlation coe cient greater than 0.8. Hence, there are no collinearity problems between the variables.

Results
Before judging the impacts of the board quality, we should estimate the separate impacts provided by the board determinants and the effects generated by the other control variables on the FP measures. To do this, we have established multiple linear models.
To correctly decide the individual signi cance of the variables, we referred to the Student statistics. When the estimated statistic's probability is less than one of the reference signi cance thresholds, we have selected the variable in question. Otherwise, the effect of the variable is considered insigni cant. The list of tables from 10 to 17 summarizes the coe cients of the different explanatory variables estimated by the model of each sample.
The BOD may have a positive or negative in uence on the banks' FP depending on the situations encountered. So far, we have checked the signi cance of the variables that explain the quality of the BOD in each model. In the next step, we established a comparative study between the same impacts of similar models, which highlights the importance of the board in their existence. Finally, we made a comparison between the preestablished signs (expected) and the signs already found.
4.1.1. Impacts of the board quality on the pro tability of conventional and Islamic banks Based on the Table below, the CBs' pro tability coe cients saw two conclusions. About the variables LnTCONSc, LnINDCONSc, and LnREUCONSc, they deteriorated the CBs' pro tability, but only the impacts of LnTCONSc and LnREUCONSc are signi cant at the threshold of 1%. By contrast, ENRADIRc showed a positive but no signi cant effect on the CBs' pro tability. About control variables, TYc, LnAGc, and LnTAc reported positive impacts on pro tability, but only LnAGc and LnTAc recorded signi cant impacts at the levels of 5% and 1%. Nevertheless, LnINFc negatively affected the CBs' pro tability signi cantly at the 5% level. As shown in Table 10, hypothesis n°4 has been con rmed in the case of CBs. Thus, we successively rejected hypotheses n°1, n°2, and n°3.
The results in Table 11 have distinguished two determinants of board quality adopting signs of negative and signi cant impacts on pro tability, LnTCONSi is signi cant at the 10% threshold and LnREUCONSi is signi cant at the 5% threshold. Nevertheless, only LnINDCONSi recorded a positive and signi cant impact on the IBs' pro tability at a rate of 1%. For the control variables, TYi and LnAGi revealed positive and signi cant impacts at the 1% level. However, LnTAi and LnINFi negatively affected the IBs' pro tability. As LnTAi is signi cant at the level of 5%, but LnINFi is signi cant at the level of 1%. Consequently, we adopted hypothesis n°4. However, contrary to what is anticipated, according to Table 11, the hypotheses n°1, n°2, and n°3 were rejected in the case of the Islamic banking model.

Impacts of the board quality on the e ciency of conventional and Islamic banks
As revealed in Table 12, the correlation between board determinants and the CBs' e ciency showed that ENRADIRc and LnINDCONSc negatively affected bank e ciency at the 1% level. Similarly, LnTCONSc and LnREUCONSc affected the e ciency of BCs, but at the 5% threshold. Concerning the additional variables, LnAGc and LnINFc generated a positive sign to reinforce the CBs' e ciency at the level of 1%. Also, TYc revealed a positive impact on the e ciency of BCs, but only at the limit of 10%. Notwithstanding the three mentioned variables, LnTAc adopted a negative effect that has acted signi cantly at the 5% level. Compared to previous results, we con rmed the hypotheses n°3 and n°4. However, referring to the decision rules, we rejected the hypotheses n°1 and n°2 (Table 12).
After analyzing Table 13, in IBs we mentioned that the same variables have another combination of impacts on the bank e ciency of our sample. LnTCONSi and ENRADIRi recorded a positive and signi cant effect on the BIs' e ciency at the threshold of 1%. However, LnINDCONSi and LnREUCONSi negatively correlated with the e ciency of BIs, but only the impact of LnREUCONSi is signi cant at the level of 5%. About the impacts of control variables, all the variables acted negatively on the BIs' e ciency, among which only the impacts relating to LnAGi and LnTAi have recorded negative and signi cant in uence effects on e ciency at the rate of 1%. Therefore, the hypotheses n°1, n°2, and n°4 were accepted. Subsequently, we rejected the third hypothesis speci cally in the case of IBs (Table 13).

Impacts of the board quality on the liquidity of conventional and Islamic banks
According to the Table 14, the speci c coe cients of the board effect on the liquidity available within CBs or produced by the operating work showed that LnTCONSc and LnREUCONSc play a fundamental role in forcing the liquidity production, whereas, this effect is only partially signi cant with LnREUCONSc at the level of 1%. However, ENRADIRc and LnINDCONSc negatively affected availabilities but not necessarily signi cantly, with the exception made, the incidence received by ENRADIRc is signi cant at the level of 5%. Regarding the additional variables, TYc, LnTAc, and LnINFc showed negative effects that are signi cant only with LnTAc and LnINFc at the rate of 1%. Nevertheless, LnAGc revealed a positive and signi cant impact at the 1% threshold on the CBs' e ciency. Therefore, we explicitly rejected all our hypotheses (Table 14).
Also, Table 15 revealed that the estimated model of the IBs' liquidity reported an improvement of the board characteristics that gave coe cients with positive signs just as ENRADIRi, LnINDCONSi, and LnREUCONSi. The analysis showed positive effects on liquidity, among which only LnINDCONSi recorded a signi cant impact at a rate of 1% and ENRADIRi scored a signi cant impact at the level of 5%. Conversely, LnTCONSi acted signi cantly on liquidity at the level of 1%. Focusing on auxiliary variables, we noted that LnINFi negatively affected the IBs' liquidity at the 10% signi cance rate, while the LnAGi effect is negative and signi cant at the 5% rate. However, TYi and LnTAi generated a positive and signi cant impact at the 1% threshold on the IBs' liquidity. As pre-established, based on Table 15 we validated the hypothesis n°2, for the IBs' model. Nevertheless, contrary to what is planned, we rejected the hypotheses n°1, n°3, and n°4.
4.1.4. Impacts of the board quality on the solvency of conventional and Islamic banks In the same line, through Table 16 we noticed that the solvency coe cients associated with the board determinants revealed that ENRADIRc, ENRADIRc, and LnREUCONSc participate in the preservation of the bank solvency rates, but more speci cally, only the role of LnREUCONSc is signi cant at the threshold 1%. However, LnTCONSc raised the CBs' solvency signi cantly to the 1% threshold. For the secondary variables, we noticed that TYc and LnINFc contributed to the decline in the solvency rates of banks at the signi cance level of 1%, the impact relative to the LnTAc is signi cant at the 5% level and the effect of LnAGc just at the 5% level. Therefore, referring to Table 16, all our hypotheses were rejected.
As shown in Table 17, through the estimation of the CBs' solvency model, we found that there are three board characteristics involved in the protection of the solvency of this bank type, like ENRADIRi, LnINDCONSi, and REUCONSi, by which only LnINDCONSi and LnREUCONSi reported signi cant impacts on solvency at the levels of 1% and 10%. However, LnTCONSi has an unfavorable and signi cant impact on the IBs' solvency at the threshold of 1%. Symmetrically, other additional variables reported negative effects. More particularly, we cite, TYi, LnAGi, and LnINFi, but only the effect of LnINFi on solvency is signi cant at the 1% threshold. However, LnTAi positively and signi cantly in uenced the solvency of BIs at the rate 5%. Hence, according to Table 17, these results allowed us to rejected all the hypotheses in the case of IBs.

Discussion
From the foregoing, the mono-analysis already carried out has shown an ambiguity of con rmation or assertion of the hypotheses from a single FP measure. First, not all the board determinants of the two bank types revealed signi cant effects on FP measures. Besides, not all board determinants of each bank type recorded the same signs. To overcome the problem of inconsistency of the non-signi cant impacts, we were limited to the variables that revealed signi cant signs. Then, the incompatibility of the signs led us to establish a state of reconciliation between the signi cant effects of the board determinants speci c to each bank type. To better appreciate the difference in board effects on the FP of each bank type, we compared the individual effects of each board determinant on the same FP measure for each bank type. Table 18 illustrates the reconciliation results speci c to CBs with their Islamic counterparts.
From Table 18, we noted that the exclusion of non-signi cant impacts and the aggregation of signi cant impacts clari es and simpli es the vision. On the one hand, the bringing of all the BOD impacts on FP together has shown that the CBs' board quality has deteriorated their pro tability, their e ciency, their liquidity, and their solvency even though the board features have improved some of their liquidity and their solvency. On the other hand, the IBs' board quality has revealed a non-decisive impact on FP. BOD has improved their e ciency, their liquidity, and their solvency, but at the same time, it has affected their pro tability and some of their liquidity, and their solvency. In conclusion, the degree of signi cance of the negative impact on the advice of CBs is more in uential on FP than the degree of signi cance of the positive impact on IBs.

Conclusion
This article examined the relationship between the composition of a board of directors and the FPs of conventional and Islamic banks within the framework of stakeholder theory. We have noticed that our empirical results are not conclusive in the case of two models of banks, we pointed out that the board quality in the banking environment represents a two-way destiny that does not have a decisive impact on FP. By evidence, we have estimated that board size and board independence have normally contributed to improving the FP of conventional and Islamic banks. Also, we have planned that the rooting of the board chairman and the number of meetings held by the board have in uenced negatively their FP. However, the reality has revealed mixed results, all depend on the FP measure. The degree of partial impact varies from one board determinant to another and from one FP variable to another for the same control variable. Besides, the results vary from one FP measure to another for the same board characteristic, depending on the study context, the accounting standards applied and the governance approach followed.
Due to previous factors, we discovered a new classi cation method of the heterogeneous and detailed impacts of the board quality on the FP through which we argued that the board quality has revealed a driving impact on FP in each bank type. The development of new accounting and auditing aspects aimed at steering the control of the BOD has become a necessity.
The global uni cation of auditing standards allows us to adjust and arrange boards of directors, while respecting the particularities of each banking model, to facilitate the supervision and monitoring of the governance systems introduced in banks independently of their types. Based on this proposal, the creation of an international institute for monitoring bank boards is possible. The initiative also opens the horizon for highlighting an international non-pro t organization with the power to the mandatory application of the standards of organizing the boards of directors and improving their working qualities.
Conventional or Islamic banks of large sizes are too big to fail, but also too easy not to go bankrupt. Everything is proportional, as long as there are failures in their governance systems, FP remains a relative dependent variable. The complexity of the large banks' activities goes beyond the traditional role of the BOD in that the degree of risk is higher, independently of the bank type. Nevertheless, the collective competence to identify and monitor challenges, overruns, con icts, manipulations, agency relationships, opportunistic behaviors, and banking risks is lower. This raises an important question about the feasibility and su ciency of the traditional board structure.
Board administrators of conventional and Islamic banks should develop an information system and a risk rating system that facilitates for them credit monitoring. Also, banks are expected to implement credit restrictions and limits, new credit approvals, and debt rescheduling. Credit management should be subject to a robust and well-de ned credit-granting process that allows them to detail their target markets, establish a thorough understanding of the borrower, their sources of repayment, and the purpose of the credit. Moreover, the credit system should determine the adequacy of loan loss provisions and reserves. When assessing credit risk exposures, banks should take into account potential future changes and economic factors, such as in ation, gross domestic product, and growth rate. Bank boards should install more stringent credit distribution policies for the reduction of defaults that result in an accumulation of defaulted loans and non-performing loans in order to maintain the coverage of loan loss provisions and distinguish between high-risk portfolios and low-risk portfolios.
Our study enriched this literature by examining how the attribution of the power of the board in uences the volatility of the FP of conventional and Islamic banks to be able to distinguish between these two banks' types through the impact of their board quality on their FP. For future lines of research, this study can be extended by using unbalanced panel data to include all conventional and Islamic nancial institutions. Other econometric techniques can be applied to compare the importance of BOD on the FP of conventional and Islamic banks.    (2002)