The impact of the supervisory board structure on bank performance

Thus, through this research, the authors aim to provide answers (justified by the results of an empirical research) to the question “Can the structure of the supervisory board affect the financial performance recorded in the Romanian banking system?” The research methodology used is mainly quantitative, based on a statistical deductive analysis, testing and identifying the type cause – effect connections, and in the same time with the assessment of the degree of their significance, representing the main objective of this research.


Introduction
As a result of the global financial crisis, in the banking industry there was registered a significant increase in the degree of harmonization and regulation of processes and corporate governance systems.Given all these symptoms, this research aims to investigate whether the structure of the banks' supervisory boards (characterized by size, structure and gender diversity) can impact on their performance and, if so, what is their degree of influence.
The failure of the corporate governance processes can generate significant costs for the banks.The occurrence of such costs is due to the banks' role in the national economy, namely the role of financial intermediation, the coverage of the payments needs, of liquidity, of information, of insurance of the savings products (Fama, 1985).In addition, the banks are more intensively regulated, thereby taking measures to prevent the negative externalities caused by the "systemic risk" (Flannery, 1998) and to protect the interests of the banks' depositors.
It is also well known the important role of the banks in facilitating the good governance of the companies, both in their role as creditors and shareholders (Caprio and Levine, 2002).Therefore, banks can contribute not only to ensuring the good functioning of the companies from the non-financial sector, but also to promote an efficient policy of the financial resources allocation across the economy.
The supervisory board of the bank plays an important role in achieving effective corporate governance.Thus, Caprio and Levine (2002) consider that banking corporate governance is significantly determined by the structure of their supervisory boards.However, the authors consider that neither the shareholders nor the market partners can considerably affect the banking corporate governance.
The supervisory boards' importance in terms of banking corporate governance is also emphasized by the banking regulatory authorities.For example, in the consultative paper named "Strengthening corporate governance in the banking sector" (2006), the Basel Committee on Banking Supervision (BCBS) identifies the supervisory board as an essential part of a bank's regulatory reform.Moreover, the second pillar (supervisory review process) of Basel II identifies the role of the supervisory board as an integral part of the risk management process (BCBS, 2005).
In this context, the aim of our research is very important; the literature on the relationship between the structure of the supervisory board and bank performance is not yet conclusive.This is also the first study carried out for the Romanian banking system which includes the examination of the effect of the supervisory board's gender diversity on bank performance.The literature regarding the women's presence in the supervisory boards is not yet conclusive (e.g., Adams and Ferreira, 2009

Literature review
The negative relationship between the size of the supervisory board and the companies' performance is a common finding in the literature regarding the nonfinancial sector (Hermalin and Weisbach, 2003).The relationship is driven by the agility and the cohesion between the members of smaller supervisory boards and the costs that can be generated by the communication and coordination needs in the case of larger ones (Jensen and Meckling, 1976).
The members of the large supervisory boards could also experience challenges in expressing their opinions during the limited time available during the meetings (Lipton and Lorsch, 1992).Consequently, many studies show a negative correlation between the size of the supervisory boards and the financial performance of the company (e.g., Yermack, 1996 The large supervisory boards may reflect the complexity of the organizational structures of banks.The supervisory board size could also increase through the incorporation into their structure of members from the management teams of bank subsidiaries in order to facilitate the information flow (Adams and Mehran, 2012).However, if a banking company would register a poor organizational structure, then this would be reflected into an inefficiency of large supervisory board activities.Also, the independence of board members is very important for a bank, because they can help improve the quality of processes and operations in the relationship with business partners (  , 2003).The women representation on the boards has registered a constant (but slow) increase in recent years.This trend of increasing the share of female representatives on the boards is in accordance with the appreciation that women involved in management structures can bring added value to the company.The women managers are considered to be "hard working" and have better communication skills, which help them to improve the problem solving and the ability of decision-making activities of the supervisory boards (Robinson and Dechant, 1997).
According to Eagly and Carli (2003), women can also demonstrate the additional skills that they hold to access leadership positions, which mean that they are very competent and responsible as managers.In addition, women are considered to be more aware of their responsibilities as supervisory board members and are generally better prepared and fully documented when attending the board meetings (Eagly and Carli, 2003).This means that the women work harder in achieving their responsibilities and, consequently, could help improve the effectiveness of the supervisory board regarding the decision-making process and the information flow.
However, the conclusions of the previous research on the direct effect of the gender diversity on the company performance are inconclusive.According to Carter et al. (2003) in non-banking firms there is a positive relationship between gender diversity and business performance.
Also, Adams and Ferreira (2009) demonstrate that the presence of women in supervisory boards increases the frequency of the meetings.However, they do not search for a direct effect of gender diversity on company performance.In addition, Farrell and Hersch (2005) found that women tend to hold leadership positions in the best performing companies, although they reported insignificant abnormal profits around the day of announcement of a woman's inclusion in the supervisory boards.

Study objective and research hypothesess
The primary research objective is to identify the possible connections between some characteristics of the supervisory boards from the Romanian banking system, on the one hand (namely the size, the character of independence and the gender diversity of their members) and, on the other hand, the bank financial performance, measured by the return on assets (ROA) and return on equity (ROE).
Thus, the authors aim to provide answers (justified by the results of an empirical research) to the question "Can the structure of the supervisory boards affect the financial performance recorded in the Romanian banking system?" Based on the previous research in the scientific literature, the authors formulated the following hypotheses: There is a significant association between the structure of the supervisory boards and the return on assets (ROA) of the researched Romanian banking companies; H 2 : There is a significant association between the structure of the supervisory boards and the return on equity (ROE) of the researched Romanian banking companies.

The research metodology
The research methodology used is mainly quantitative, based on a statistical deductive analysis, testing and identifying the type causeeffect connections, and also appreciating the degree of their significance.This represents the main objective of this research.In order to achieve it, the authors used the SPSS software for Windows (the correlation tests and the regression analysis).
At the end of the first semester of 2015, there were 40 credit institutions into the Romanian banking system; out of these, 38 were privately owned banks.9 are the subsidiaries of foreign banks.
The Table 1 (BNR, 2015) shows the evolution of the number of banks during the last 7 years.Out of the total number of credit institutions in the Romanian banking system, the research sample comprises a number of 28 credit institutions.
According to the information collected from their respective websites as of December 31 st 2014, all these are Romanian legal entities.The 9 subsidiaries of foreign banks were not included, since they are not required to make financial disclosures (annual reports) in Romania (such information are disclosed by the group, in the parent company's country of origin).There were also excluded from the research a number of three credit institutions, which have not made publicly available the information on their official websites.The banks within the Romanian banking system are administered either through a unitary administration system, the supervision and management being exercised by the board of directors or through a two-tier system of administration, in which the executive management falls to the executive committee or directorate which operate under the supervision of a supervisory board, according to the statute/constitutive act of each credit institution and the banking legislation in force.
The supervisory board sets the strategies and the development guidelines and supervises their implementation, while the executive committee or the directorate provide the operative management and prevent the operational and the compliance risk.
In order to achieve the main research objective there were defined two sets of variables, independent and dependent, which underlie the study and help to identify the potential correlations.
The details related with the independent variables in the study, their definition and measurement techniques are presented in In order to test the possible correlations between the independent variables and the dependent variable, there was calculated the Pearson coefficient, which is often used to appreciate the level of linear dependence intensity that occurs between two variables.Thus, when the Pearson correlation coefficient has a value of +1 it reflects a perfect direct relationship between the two variables; when it has a value of -1, the Pearson correlation coefficient indicates a perfect inverse relationship.This analysis of possible correlations between the variables provides indications of the meaning and significance of potential interdependencies of these variables, allowing us to accept or reject the research hypotheses.

Data processing and analysis
Based on the regression analysis (we used the multiple regression model in SPSS, the Backward method) for the purpose of accepting or rejecting the hypotheses made, the potential influence of each independent variable has been processed and analyzed.
The results are detailed in the Tables 4a and 4b "The partial correlation matrix".In these tables, we present the correlation coefficients (Pearson Correlation), the significance (Sig.) for each correlation coefficient and the number of cases considered in this study (N).
In order to test the H 1 hypothesis: There is a significant association between the structure of the supervisory boards and the return on assets (ROA) of the researched Romanian banking companies, the authors analysed the impact of each independent variable on the bank's profitability.
Thus, the research results regarding the impact of the independent variables on ROA are shown on the Table 4a "The partial correlation matrix".Accordingly, it can be noticed that the diagonal coefficient equals 1, because each variable is perfectly correlated with itself.It is noted that the most significant relationship is between the banks return on assets (ROA) and the independence of the supervisory board members.
There is a positive significant relation between the dependent variable ROA and the independent variable Independence of members.The Pearson coefficient determinated is 0.545 and Sig.less than 0.05 (0.036).
Thus, we can accept the H1 hypothesis: There is a significant association between the structure of the supervisory boards (Independence of members) and the return on assets (ROA) of the researched Romanian banking companies.
In order to test the H2 hypothesis: There is a significant association between the structure of the supervisory boards and the return on equity (ROE) of the researched Romanian banking companies, we analysed the impact of each independend variable on the Return on equity (ROE) of the banks included in the sample.
In the Table 4b "The partial correlation matrix", the research results related to the influence of the independent variables on the ROE are presented.It is noted that there is no significant association between the dependent variable ROE and the independent variables, because the Pearson correlation coefficient is between 0.132 and 0.382, and the recorded Sig. is greater than 0.05 (0.073, 0.252 and 0.193).
Finally, the H2 tested hypothesis is rejected, based on the research results.We consider that there is no significant association between the structure of the supervisory board and the banking return on equity (ROE).

Conclusions
The primary research objective was to identify potential interdependencies between the structure of the supervisory boards in the Romanian banking system, on the one hand and, on the other hand, the banks' financial performance, measured by the Return on assets (ROA) and Return on equity (ROE), for the banks included in the research sample.
Testing the potential influences of the structure of the supervisory boards on company value was performed, over time, in a large number of studies.In contrast to that situation, the current research covers a different approach to this issue in a specific field of activity, namely the banking sector.The segment was not sufficiently explored so far from this perspective.Also, the hypotheses and the relationship between independent and dependent variables included in the study offer a touch of originality to the research and, by default, bring added value to the paper.
At the same time, this research has some limitations, mainly caused by the size of the sample and by the fact that the analysis was conducted using the information in only one financial year (2014).Despite these limitations, the authors consider that this study can represent a useful source of reflection and information for the practitioners from the banking sector, and it can be a challenge for the future research in this field.

Table 2 . Tabel 2. The independent variables Independent variable Definition Symbol Measurement
Source: Authors' conception.

Table 4b . The correlation matrix ROE Size Gender diversity Independence
Source: Conception of the authors using SPSS.