The Relationship Between Board Diversity, Ownership Structure and Bank Performance In Tunisian Market

This research focuses on the impact of internal governance mechanisms on the performance of listed Tunisian banks during the period 2005-2016. We try to detect the impact of banking governance, not only through the ownership structure but also through the board of director’s characteristics, on the banking performance. Our results confirm the importance of board control within the bank, but they also state that there are other mechanisms such as the ownership structure that must also be taken into account. Research shows that the ownership concentration, board size, independent and institutional administrators affect bank performance. Likewise, Duality is positively associated with the bank profitability.


Introduction
Emerging economies, characterized by economic reforms series, financial growth and integration, are still under development. Investors attracted by these growth rates can take advantage of these opportunities. The banking sectors of emerging economies often play a critical role in the success or failure of these initiatives. These sectors attract much attention, particularly because of the rich and complex environment induced by the dynamic and rapid changes in their ownership structures, but also by their governance mechanisms evolution. The banking system stability and the banks' performance development favor the smooth running of economic activities. Governance plays a crucial role in ensuring the banks stability. The European Banking Authority claims that governance weaknesses caused by insufficient accomplishment of the existing guidelines have not been a direct cause of the financial crisis, but a key factor. As a result, weak governance contributes to a decrease in bank profitability. Arie Pratama (2018) investigate corporate governance variables and whether they affect related-party transactions, uses seven proxies for corporate governance variables: ownership concentration, directors' compensation, size of the board of directors, number of independent commissioners, audit firm, size of the audit committee and debt ratio. He used age, size and profit as control variables. They conclude that not all of the proxies had a significant effect on performance. Fakhfakh and al (2009) conclude that the analysis of the economic performance should not do this only through macroeconomic variables and independent governance systems.
The corporate board enjoys a prominent place in the banks internal governance system. The corporate board is a body of control, recruitment and evaluation; it also ensures the strategy accomplishment. The corporate board must encourage the executive to maximize value creation in order to ensure consensus between the firms various partners and its manager and thus act in the shareholders' interests. The company director is one of the governance pillars; oversees the execution of the corporate board decisions and ensures the company's general management.
Extensive literature has analyzed the ownership structures effect and governance mechanisms on bank performance. These studies tend to focus primarily on the short-term effects of ownership and often ignore the long-term effects of ownership change. Moreover, despite the extensive literature on the effects of foreign banks on emerging markets, many studies did not distinguish between different ownership structures: institutional, concentrated, dispersed or public.
The recent implications of the Tunisian banks' liquidity crisis after 2011 on performance and risks lead us to investigate the internal governance mechanisms evolution to deal with these risks in banking sector. Notably because the empirical literature often provides ambiguous and country-specific evidence that shows that a particular type of property generates higher performance than others.
This article contributes to this literature by studying the effects of the banking ownership structure and the bank's board on its performance. We test the effect of internal governance mechanisms on bank performance in Tunisia. Our work is directly related to the studies of Berger et al. (2005) and Lin and Zhhang (2009).
The study's contribution to the growing body of research is four-fold. First, it extends the approach of Berger and al (2005) and Lin and Zhang (2009), who jointly analyze the static and dynamic effects of different types of ownership on banks' performance, including the effect of another internal governance mechanism, namely the characteristics of the director's board. Secondly, it distinguishes between the different properties structures: concentrated, institutional, foreign or public ownership structure.
Third, it is the first to specifically examine the effects of the ownership structures and corporate board on Tunisian banks. Our sample is made up of 11 commercial banks over the period 2005-2016. Fourth, it presents guidelines for identifying potential weaknesses in corporate governance in the Tunisian banking system.
The study completes other examining governance over a longer period. To meet these objectives, our paper begins with the presentation of the effect of the corporate board on performance, and then analyzes the impact of ownership structure, we present the data and methodology and the last section presents the empirical results.

Corporate board effect on performance
The literature analyzing the corporate board role allowed us to identify several criteria associated with the control effectiveness exercised by this mechanism. It is mainly about the board size, the members' independence, the duality between firm director and board president. The corporate board plays a major role in the adoption and implementation of internal banking governance mechanisms. Board members must have sufficient experience. Special attention should be paid to its composition, qualification, appointment and member's succession. In a corporate board, the size differs from company to another; the company is generally administered by corporate board composed of at least three members and twelve members at most. The directors are in fact appointed by the shareholders and remunerated by attendance fees.
Several authors such as (Jensen, 1993) have analyzed the impact of the board size on the bank efficiency and recommend a small board size. However, Aebi et al. (2012) found that the board size has a positive impact with performance, saying that increasing the board size generates added value for banks. Cooper (2008) also found a positive relationship based on performance. However, bigger board size can affect coordination, control and flexibility in the decision-making process.
The presence of independent members is one of the crucial elements of corporate governance. It makes it possible to increase the control on the corporate board and to influence its decisions. Independents members have a clearer and neutral view of the company and they affect the firm performance positively.
Similarly, the corporate board can benefit from independent directors who objectively evaluate the manager performance. Cornett et al (2009) found a positive impact of member independence on economic performance, through improved management supervision. Yet, Aebi et al. (2012) found a negative impact on financial performance during the financial crisis, arguing that banks are pushed by management to maximize wealth before the crisis and take risks to create wealth. In this sense, the independent members exercise more effective control than dependent internal directors.
Duality occurs when a person has both the board and the executive roles. A large number of authors claim that duality appears as an obstacle to the separation of decision-making and control functions. The combination between firm director and president board affect the board independence. Few studies show that duality improves the performance of the firm (Godard and Schatt, 2000).
Duality has detrimental effects on the effective control of the director board. As soon as there is an accumulation of control and decision-making functions within the bank, the director board will be a passive organ and will not be able to assume their disciplinary role. In the case of duality, the supervisory function is diminished because of this independence lack. In consequent, duality reduces the control effectiveness mechanism and the governance structure. In addition, Dedu and Chitan (2013) find that the duality confuse the management and control exercised by the director board.
Other studies have incorporated the experience of the board members in the banking and insurance sector; they seem to find a negative influence on performance (Aebi et al., 2012) and a positive influence related to risk-taking (Minton et al., 2010). For non-executive board members, previous studies have shown that they improve performance through better monitoring and guidance from managers (Stefanescu, 2011) or that they have a negative influence, due to the associated costs, lack of company-specific knowledge (Coles et al., 2008).

The ownership structure impact on performance
Ownership structure is an internal mechanism of corporate governance. According to agency theory, the more dispersed the ownership structure is, the higher the agency costs will be (Jensen and Meckling, 1976). This means that the ownership concentration improves performance. With, majority shareholders engaging in closer monitoring activities, more effective governance structure will be achieved; together with greater value for shareholders.
The majority shareholders, considered as influenced controllers in corporate board, are actively involved in decision-making by pushing managers to increase the firm's performance. In Tunisia, ownership is highly concentrated so that the first two shareholders can have a majority in the ownership structure; this shareholding structure reduces the manager rooting. Ozili and Uadiale (2017) test the ownership concentration effect on bank profitability of developing countries; the ownership structure is classified as: concentrated, moderate and dispersed property. They find that high-concentration banks make the best profits, with higher returns and greater solvency. In addition, increasing foreign ownership should lead to better governance, increased competitiveness and easier access to finance. Stefanescu (2011) found that foreign ownership affects the firm performance. Some studies argue that the ownership concentration of private institutional investors may lead to the right dissolution of other shareholders because of the absolute priority given to maximizing profits (Laffont and Tirole, 1993). Mian (2003) studies the effect of banking ownership structure and finds that the private banks performance exceeds public banks in emerging markets. In another study Goddard et al. (2014) found that the public banks profitability is considered lower than other types of ownership. Fethi et al. (2011) find that public banks in Egypt have benefited more from liberalization policies and, in general, tend to be more productive relative to other types of banks. Naaborg and Lensink (2008) found a negative relationship between foreign ownership and bank profitability. This suggests an advantage for domestic banks. Claessens and Van Horen (2012) show that foreign banks outperform domestic banks when (i) the foreign bank comes from a high-income country; (ii) the regulation in the country is relatively weak; (iii) the origin country and the foreign purchaser has the same language and regulation; (v) the acquirer is large and has a high market share. Iannotta et al. (2007) investigated 181 banks in 15 European countries over the period 1999-2004 to evaluate the concentration ownership effect on performance. They find that mutual banks and public banks have lower profitability than private banks. They also have lower loan quality and higher insolvency risk. Boubakri et al. (2005) find that privatization in developing countries significantly improves the bank efficiency in the long run. Jiang et al. (2013) show that foreign banks have better access to capital, can better diversify risks, have technological superiority or be more innovative, for example by introducing new management skills and new production processes.
Shaban.M and James.G (2018) find that public banks tend to be less profitable and riskier than private and foreign banks. National investors tend to select the best performance for the acquisition. In addition, domestic acquisitions are associated with a decrease in the bank efficiency. Public ownership is often considered a relatively inefficient form of ownership compared to other forms of ownership. In contrast, private banks are generally less affected by political interference and seek to maximize their profits, for example by aligning objectives with higher incentive and governance systems.

Sample
The sample on which our study was conducted includes 11 Tunisian banks listed on the Tunisian stock exchange for the period from 2005 to 2018. On the basis of this sample, we collected data on the governance internal mechanisms from annual reports and reference documents issued by banks. We have collected financial and accounting data from the financial statements and financial data published by the stock exchange and the financial market council.
To distinguish banks ownership, we follow La Porta et al. (1999), bank is classified as public if the government controls (directly or indirectly) at least 20% from the bank. We define a foreign institutional ownership bank when theirs share held exceeds 20%. An ownership structure is concentrated when the main shareholders detained more than 50% of bank shares.

Model
We select three dependant variables: -Return on asset (ROA): an economic performance indicator, calculated as follows: ROA= Net Profit / Total Assets Electronic copy available at: https://ssrn.com/abstract=3472174 -Market return or Market to Book MTB ratio calculated as follows: Market capitalization / equity. -Q Tobin as a stock market performance indicator calculated as follows: Q Tobin = market value / book value The independent variables used are related to ownership structure and board characteristics, two control variables are also used in our model. Indeed, the variables related to the ownership structure are the part of capital held by the majority shareholders be it institutional, foreign, or state-owned. The variables related to the board of directors are the size and composition, the proportions of the institutional, independent or external members in the corporate board and the duality between board president and firm director.
We also include in our models two control variables: the bank size and the leverage ratio. PERF i, t = α0 +α1 (MAJ) The impact of capital concentration on bank performance is theoretically complex and empirically ambiguous. In fact, Caprio et al. (2006) find that ownership concentration has a positive effect on the performance of 244 banks from 44 countries. On the other hand, Pinteris (2002) demonstrate that there is insignificant relationship between ownership concentration and bank performance in Argentina. From this overlap of previous studies, we formulate our first research hypothesis: H1 -The capital concentration has a positive influence on the bank's performance.
We consider institutional investors, banks, insurance companies, pension funds, collective investment, mutual funds and pension funds. These investors are expected to play an active role in corporate governance. Berger et al (2006) studied 695 American commercial banks during the period 1990-1995, and find that the presence of institutional investors produces very large control effects that reduce agency costs and increase performance. The role played by institutional investors in the bank performance is ambiguous. Therefore, we formulate a second research hypothesis:

H2 -The presence of institutional investors has a positive influence on bank's performance
The capital part held by foreign investors (ETR): this variable is measured by the number of shares held by foreign investors in relation to the total shares. Claessens and Laeven (2004) argue that the introduction of foreign ownership in banking, especially in developing countries, results in improved competitiveness for local banking systems and consequently improved performance. In fact, we emit a third hypothesis of research: H3 -The presence of foreign investors has a positive influence on the bank performance.
We introduce the variable ETAT to measure the impact of state participation on the Tunisian banks performance. This variable is defined as the proportion of capital held by the state. Various empirical studies have focused on the relationship between public ownership and the bank performance. Lang and so (2002) concludes that public banks are characterized by lowest performance. In consequent, we test the following hypothesis in Tunisian context: H4 -The public ownership has a negative influence on the bank performance.
Board size (TCA): measured by the number on the board of directors. Ghosh and Ansari (2018) examine the association between financial performance and the boards of banks in India. Using data on 1263 banks, the results indicate that the board size does not affect performance. A large board size negatively influences performance and creates coordination problems between board members. To do this, we present the following hypothesis: H5-Large corporate board improves bank performance.
The presence of institutional members on the board of directors (NSAI): The NSAI variable presents the ratio between the number of institutional members and the total number directors in corporate board. Institutional directors include representatives of banks, insurance companies, pension funds, collective investment, mutual funds and pension funds. Berger et al (2006) studied a sample of 659 Americans banks during the period 1990 to 1995. Their results indicate that institutional investors improve the bank performance through their control. For this purpose we expect the following hypothesis: H6-The presence of institutional directors on the board of directors has a positive influence on the bank performance.
Members Independence on the board of directors (AIND): The variable AIND is measured by the proportion of independent members on the board of directors. The independent director must not have a professional relationship with the bank. Adams and Mehran (2005) show that external directors have no impact on bank performance. So, we propose to test the following hypothesis: H7 -The presence of independent members on the board of directors has no influence on the bank performance. Duality (DUAL): Duality is considered to be the dual function between board president and bank manager. To analyze the effect of duality on the performance of banks, we consider the variable (DUAL) which is a binary variable. It takes the value of 1 when the president board is combined to bank manager and 0 otherwise. The financial literature shows that the problem called "duality" or the accumulation of the positions of the general manager and president board presents separate opinions, but the majority of these studies support the negative effect on performance. We therefore formulate our last research hypothesis: H8 -Duality has a negative and significant impact on banks performance Control variables, which are the variables that can have a significant effect on performance. In fact, we retain in this context, the bank size and leverage. Bank size (LTA): The bank size has often been considered as a variable that can have a significant effect on performance. In general, the bank size variable is expressed as the natural logarithm of the book value of its total assets. In fact, we can predict in our study that the bank size is positively associated with performance. LTA = log (asset book value). The bank's debt ratio (NDT): To take into account the debt effect on performance, we use the ratio of total debt to total assets.

Results
Table (1) presents descriptive statistics relative to different variables of our model.
The average ROA for our sample is 0.821%. The minimum and maximum ROA for the sample is -8.13% and 2.91%. This proves the weak performance of Tunisian banks compared to international banks. The average MTB is (1,5), the minimum is (-1,34) and the maximum value is (3,86). This result shows that the majority of Tunisian banks are overvalued in the market. The majority shareholders held in average (39%), the minimum is (6%) and the maximum value is (75%). The result proves that Tunisian banks are characterized by concentration ownership structure. The institutional shareholders held on average (50%), the result prove that Tunisian banks are detained by institutional ownership structure. The state and foreign shareholders held in average respectively 11% and 29%.
The average part of independent directors in corporate Board is 8.14%, which remains much lower than that of international banks.
Our correlation matrix in Table (2) shows a significant correlation between ETR and MAJ and between AINS and ETAT. After analyzing the correlation between the independent variables, we performed a homogeneity test in the panel model. To determine the model specificity, we carried out a Hausman test which made it possible to distinguish fixed and random effects.     Barth et al. (2005) results, that demonstrate the negative impact of public ownership on bank performance and that public banks have lower performance than private banks.
Although the public Tunisian banks occupy the last three places in a decreasing order of average ROA during the period 2005-2015, the results do not prove the negative effect of public ownership on the banking performance, this results can be explained, among other things, by the competence and experience of the manager's banks.
The board size isn't significant related to banks' financial performance. The board size doesn't affect performance. The large number of board members increases conflict and therefore has greater potential for disagreement and lack of coordination in management decisions. Thus, the board composition seems to have various effects on banking performance. Indeed, the independent board members have a negative and significant impact on the bank performance. This can be explained by the high ownership concentration of Tunisian banks, which can replace the control exercised by the board of directors. Thus, the role played by institutional administrators seems weak and limited due to the high ownership concentration. Thus, we confirms the results found by Pathan and Faff (2013) that show the negative relationship between director independence and performance, and the results obtained by Ben Bouheni et al (2016) who claim that board independence is statistically significant and negatively related to the bank profitability during the period 2005-2011.
In addition, the presence of institutional directors on the board of directors isn't correlated with the financial performance of listed Tunisian banks. For duality, we find that it is positively associated with financial performance (Qtobin and MTB). However, we note that the combination of functions has a greater positive and significant impact on the stock market performance. This can be explained by the fact that the functions combination improves the manager's competence and consequently the bank performance. So we will conclude that duality improves the bank performance. The bank size has a positive and statistically significant impact on performance. So, the larger banks are able to improve performance, measured by the ROA. In addition, the leverage has a significant and positive impact on the Tunisian banks performance for the ROA.

Conclusion
In this paper, we test the impact of internal governance mechanisms on bank performance. This subject is fueling a rich debate and empirical results remain controversial. Therefore, we have tried to contribute, through this paper, to enrich the debate on this subject using empirical study of the Tunisian banks listed on the stock market during the period of 2005-2016. Our results reveal that ownership concentration isn't related to the bank performance. Thus, the institutional, foreign or public investors didn't affect the financial and economic bank performance. On the other hand, we didn't found a relationship between board size and financial performance. In addition, the independent directors presence tends to reduce the bank performance. The duality between board president and bank manager has a positive impact on the financial and stock market performance. The bank size has a positive influence on the bank's performance.