Analysis of Board of Directors and Risk Propensity

Analysis of Board of Directors and Risk Propensity Banks are all similarly confronted with particular regulations and inspections of banking supervisions. Within this topic, the board of directors plays an important role. There are different factors considering how the composition of a board might influence its performance and the decision-making process. Therefore, factors like independence, age structure, percentage of minorities and women and the size of the board will be analyzed. After determining the composition, the influence of it on the percentage of equity financing and therefore the risk propensity will be analyzed. Furthermore, the composition will also be linked to performance indicators as Return on Assets (ROA), Return on Equity (ROE), and the development of stock quotations. 3.1. Corporate Governance The term corporate governance describes processes through which an organization is controlled and directed. Those structures specify which rights and duties certain participants in a company have and how the decision-making process works. This mostly affects the board of directors, the top management team (TMT) as well as shareholders and other stakeholders (OECD, 2005). Corporate governance is concerned with the possible abuse of power of the managers and the need for certain qualities like openness, integrity and accountability during the whole decision-making process. As shown in Figure 3.1, it also examines how certain mechanisms, including incentives, can help to minimize transactions costs that arise in an organization between principals and agents as described in the agency theory below (Mathiesen, 2002). 3.2 Structure of the Board of Directors The members of the board are generally elected by the shareholders and their responsibilities vary with the nature and the complexity of the organization. However, there are two different systems regarding the boards of directors. On the one hand there is the Anglo-Saxon system in countries like the United States and Japan (12Manage: The Executive Fast track, 2008). This consists of a one-tier board structure, where executive and non-executive directors work together in the board of directors (Weimer and Pape, 1999). This single board is usually entirely appointed by the shareholders and the CEO often also holds the board chair (12Manage: The Executive Fast track, 2008). On the other hand, in countries like Germany and the Netherlands, companies have adopted a two-tier board structure. There, the board is divided into the managing board and the supervisory board to formally separate powers (12Manage: The Executive Fast track, 2008). The managing board is monitored and advised in major policies by the supervisory board (Weimer and Pape, 1999). The CEO holds the chair of the managing board, but cannot hold the chair of the supervisory board at the same time (12Manage: The Executive Fast track, 2008). Even though, the board of directors is usually elected by the shareholders, in some cases also employees elect their own representative(s) from the workforce to support their interests on the board. In state-owned banks directors are delegated to the bank by the State Council and in where the board of directors is spitted up into managing board and supervisory board, the managing directors are appointed by the supervisory board members as shown in Figure 3.2. 3.3 Tasks of the Board of Directors In general, directors represent the shareholders interests, because they provide the elementary assets for running a company. Therefore, the main role of the board of directors is to govern an organization while acting for the shareholders in order to protect their assets and to ensure a decent return on their investments (Oss, 2003; Kennon, 2008). The board of directors is the à ¢Ã¢â€š ¬Ã…“highest governing authority within the management structure at any publicly traded companyà ¢Ã¢â€š ¬? (Kennon, 2008, n.p.). For this reason, the board is in charge of defining the corporate mission, setting the companys objectives and approving the firms strategy concerning the well judged allocation of the financial resources (Oss, 2003). Even though the board holds the total authority for a companys decision making they cannot manage the companys day-to-day operations, because this is the role of the CEO and the TMT (Oss, 2003). The resulting conflict potential is discussed in the Agency Theory below. According to Oss (2003) it is the boards task to govern and the CEOs to manage. Therefore, a clarified classification of who is in charge will eliminate these conflicts. Furthermore, another stakeholder group exists, as mentioned in the Stak eholder Theory (see Figure 3.3) with additional interests and requirements for the board of directors. Regarding all players and interest groups in an organization, the responsibilities of the board are possible to be divided into a Governance Role, a Service Role and a Control Role. Beside the strategic decisions, an important task of the board members is to appoint special committees like the Audit and Risk Committee, and to select qualified managers, as well as to help and to support them with their skills and expertise. Finally, the board controls if the management meets the companys objectives concerning ethical tenets or laws (Oss, 2003). 3.5 Related Research regarding Board Composition From the theories and former research, it becomes clear that boards have different tasks. Therefore, an optimal structure or composition of the board is essential for fulfilling the tasks. The main tenor in the literature is that in order to work efficiently boards have to be independent with diversity in backgrounds, gender, race and age. However, a certain composition of the board might also affect how much risk the directors are willing to take. The composition of the board receives more and more attention in terms of structure and stability. If a better structured and more stable board of directors is related to a better firm performance, companies with a well-composed board should perform better than other companies. For the purpose of this paper, a well-composed board is defined as a stable and diverse board composed of a majority of independent members along with a number of women and ethnic minority directors. To build the connection to the topic terms à ¢Ã¢â€š ¬Ã¢â‚¬Å" board composition on the one side and firm performance and risk propensity on the other side à ¢Ã¢â€š ¬Ã¢â‚¬Å" it is important to focus on different behavioral patterns which are the result of variations in board compositions. By investigating the influence of the board composition on firm performance and risk propensity it is possible to get insights into how differently composed boards behave regarding specific board tasks. Differently composed boards behave differently in various situations; for example, when they decide whether to replace a poorly performing CEO or when they choose at what price the company should be sold. The boards decision is also important when the acquisition of another firm has to be approved or whe n takeover defenses have to be adopted and employed. Finally, the board plays a big role when it comes to establishing the CEO and executives compensation packages (Bhagat and Black, 1999). 3.5.4 Board Size Board size is seen as one of the most important factors when it comes to the influence on the performance of a company (Kyereboah-Coleman and Biekpe, 2005). The main view regarding board size is that large boards have a negative impact on the performance of the company. That is, because tasks like coordination, decision-making and the communication between the members are more difficult and expensive, the more directors have to be included (Belkhir, 2008). Therefore, the costs would outweigh the gains of having more expertise on the board. Belkhir (2008) cited Jensens (1993) statement that boards with more than seven or eight people are less effective and easier to control for the CEO. Earlier research of the board size supports the proposition that smaller boards are better. Yermack (1996) discovered a negative relationship between board size and firm performance measured by Tobins Q and several other accounting figures. In their sample of small Finnish firms, Eisenberg et al. (1998) also find a negative relationship between the number of directors and financial success of the company. Furthermore, Kyereboah-Coleman and Biekpe (2005) determined that large board sizes are bad for the sales and growth ratio of companies in Ghana. However, Belkhir (2008) found a non-negative relationship between the size of the board of directors and the firm performance measured by Tobins Q, as well as, by return on assets (ROA) for financial institutions. Especially savings-and-loan holding companies (SLHC) might increase the value of the company with a rising number of directors. Therefore, the next hypothesis is that: H4a: An increasing board size has no negative influence on the company performance. When it comes to taking risks, there is not as much empirical evidence. However, if one looks at the decision-making process of a board, especially when its number of directors is very high, the obvious assumption is that for risky decsions it is more difficult to get a consensus the more people have to vote for it. Furthermore, Pfeffer and Salancick (1978) and Lipton and Lorsch (1992) determined a relationship between the capital structure of a company and its board size. Additionally, a study of Abor and Biepke (2005) discovered that an increasing board size and the debt level of Ghanaian SME are negatively related. Thus, the authors assume that: H4b: The board size is negatively related to the risk propensity of the company. 3.5.5 Board Independence You can distinguish between inside directors (current officers of the company) affiliated outsiders (former company officers, and persons who have business relationships with the company) and independent directors (Bhagat and Black, 1999, p. 4). Independent board members (outside directors) are à ¢Ã¢â€š ¬Ã…“not associated with or employed by the companyà ¢Ã¢â€š ¬? (Kennon, 2008, n.p.). According to Kennon, in the United States at least fifty percent of the directors must meet the requirements of independence. A board with fifty percent of independent directors is called a majority-independent board (Bhagat and Black, 1999, p. 4). The Sarbanes-Oxley Act of 2002 places a strong emphasis on the independence of directors. Brown et al. (2004) confirmed this requirement with positive results in their study on the effects of the independence of the board members on financial firm performance data. They found that independent boards have higher return on equity (ROE) and profit margins. Furthermore, it is determined that outside directors can monitor the management more effectively than insiders (Bonn, Yoshikawa, and Phan, 2004). Therefore, the conclusion of several empirical studies is that, besides a more diverse board, a more independent board has a positive effect on the financial performance (see also Adams and Mehran, 2008). But, boards with majority-independent directors have both positive and negative effects. On the one hand, inside directors are more involved in the companys operations and might know the business better than outsiders. On the other hand, outside directors might keep cool and act in a more objective way than insiders. Besides that, several studies did not find significant evidence that a higher number of independent directors within the board is related to the quality of financial reporting, or to the likelihood of firm failure. Additionally, there is no evidence of more firm-level diversification or a connection to research and development spending (Bhagat and Black, 1999). Therefore, Bhagat and Black (1999) recommend that it might be valuable for companies to compose their boards with at least a moderate number of inside directors. This is supported by their results that there is a negative relationship between the degree of board independence and firm performance. However, different firms need different types of boards and an optimal board contains a combination of inside, affiliated and independent directors who bring different skills and knowledge to the board (Bhagat and Black, 1999, pp. 32-33). Along with the companys objectives and shareholder interests, boards of banks additionally bear micro- and macro-economic responsibilities, which can be positively influenced by the optimal composition of the board. For board members of financial institutions, a cooperative board-CEO relationship is elementary. Only when the board gets the complete information about the operating business processes from the CEO, can they make the right decisions for the company. For this reason, it is important to know if inside or outside directors can deal better with the CEO or TMT and generate a higher performance. The writers therefore hypothesize for the banking sector, that: H5a: A higher number of outside directors does not influence firm performance. Pfeffer and Salancik (1978) developed the Resource Dependency Theory and determined that a number of outside directors upgrade a companys ability to protect itself against outside influences and reduce the uncertainty level. Furthermore, they stated that outsiders might help the company retain a certain status and raise funds. Thus, a higher number of outside directors on the board should increase the debt level of the company. On top of that, independent directors might act more in the shareholders interests than inside directors (Bonn, Yoshikawa, and Phan, 2004), and for this reason we hypothesize: H5b: A higher number of outside directors will be positively connected to the risk propensity of the company. 3.6 Summary of the Hypotheses In table 3.1. below, all hypotheses are presented at one glance. These propositions have been derived from past studies and behavioral theories as presented above. Table 3.1 Hypotheses Composition Company Performance Risk Propensity Gender Diversity H1a: A higher percentage of women on the board of direc-tors has a positive influence on firm performance. H1b: A higher percentage of women on the board is nega-tively connected to the risk pro-pensity of the company. Average Age of Directors H2a: The average age of the board of directors is nega-tively connected with firm performance. H2b: The average age of the board is positively connected to risk avoidance of the company. Ethnic Diversity H3a: A higher number of minority directors on the board is positively related to company performance. H3b: Minority directors do not affect the risk propensity of the company. Board Size H4a: Increasing board size has no negative influence on the company performance. H4b: The board size is nega-tively related to the risk pro-pensity of the company. Board indepen-dence H5a: A higher number of outside directors does not in-fluence firm performance. H5b: A higher number of outside directors is positively related to the risk propensity of the company. 4. Empirical Study 4.1 Research Methodology 4.1.1 Sample In this empirical research the top 50 banks in the world according to Bankersalmanac.com4 were investigated. The banks were ranked according to their total assets as of June 30, 2008. For the data collection, there was a time span of three years, from 2005 to 2007. This particular group of banks has been chosen, because of their size and international branches. The reasons for choosing the largest banks from all around the world were to have a comparable size of international business when comparing them. If the banks had only been from one or two countries the differences in size would have been significant and the developments on the financial market would have probably only affected the bigger banks. This might have had an effect on the performance. Thuse, for 2007 the results could have possibly been very inconsistent. Regarding the chosen sample, it can be assured that the international situation has affected them all à ¢Ã¢â€š ¬Ã¢â‚¬Å" some banks more than others, depending on how risky their business operations were. That results in a possible interpretation on how each bank, with managers and board directors, has dealt with the problems and obstacles. 4.1.2 Data Collection Method The necessary data for this study were collected from the annual reports of the 50 financial institutions. The data about the board of directors were found in the corporate governance section of the reports. Data about the company performance were gathered from the consolidated income statements and balance sheets of the banks. The data were usually dated the 31st December of the year. However, some bans adopted a fiscal year ending on March 31, or September 30 of the year. Then, we considered the Annual Reports from March 31, 2008 as belonging to 2007 as well as the Annual Reports from September 30, 2007. This way, it could be assured that the figures were all derived during the similar time period. Furthermore, the main capital ratios, necessary for the risk propensity were taken from the section risk management. However, there are no strict regulations on how companies have to structure their annual reports. Therefore, the relevant data of some financial institutions was found in different parts of the annual reports or on the websites of the companies. 4.2 Operationalization The research data were collected in an Excel sheet for further calculations and preparation purposes with regard to the statistical analysis using the statistics program SPSS. During the research process the researchers also used a complementary list to record secondary and supportive information needed to calculate the total numbers for the primary Excel list. 4.2.1 Board Composition Data The five board composition variables were selected by the authors and the corresponding information about those data were collected as presented in the following paragraphs. 4.2.1.1 Board Size The board size was recorded by counting the members and listing their names. This was done for all three years to find out if there were any changes in the board composition from one year to another. The changes were recorded in the complementary lists. The total number of board members for each bank and each of the three years were transferred to the primary Excel sheet. 4.2.1.3 Independent Directors Next, the authors looked at the percentage of independent directors. The financial institutions usually indicated in the annual reports or on their homepage which members of the board were independent. However, sometimes it was not explicitly alluded neither in the annual report nor on the companys website. Then, the researchers decided if a director was independent or not using an own definition described above. The authors examined if s/he has any other connection to the company beyond the board activities; for instance if him/her is or was employed by the bank in the last years or bears executive tasks. If there was no connection (excluding shareholding) at all, s/he was defined as independent director. The share ownership of directors was excluded, because at some banks each director is obliged to hold at least a small number of shares. The sum of all independent directors of each bank was copied to the list and divided by the total number of members. 4.2.2 Company Performance Indicators To investigate the influence of the board composition on the firm performance the authors chose four performance measures divided into two categories: operating performance and shareholder payout. The three most important financial indicators are the performance measures Return on Assets (ROA), Return on Equity (ROE), and the Efficiency Rate (CIR). Besides this, the researchers also looked at the share performances compared to previous years. These are all common indicators, which are important for shareholders. Furthermore, they are well comparable to the results previous studies in other industry branches showed using the same indicators. To control for the possibility that the performance indictors will be connected to the size of the bank, total assets were also recorded and will be included in the correlation tables and regression models. 4.2.2.1 Return on Assets The accounting measure of a companys profitability, Return on Assets (ROA), indicates net income from all of the banks operations relative to the average book value of all assets (Carter, DSouza, Simkins, Simpson, 2007, p. 15). It shows how beneficial assets are used by management to create earnings for the company. This means that it is possible to see how much profit was derived from invested assets (Investopedia, Definitions, 2008). It is calculated as: ROA= Net Income/ Avg. Total Assets 4.3 Credibility of the Research Data 4.3.1 Reliability Reliability is concerned with the question if the data that were collected by the researchers would be consistent with the findings other researchers would have using the same sources (Saunders et al., 2007). The most data were collected by the authors in a quantitative but diligent manner from the published and certified annual reports of the banks. Due to this fact, the research data cannot be interpreted wrong by the researchers and therefore have a high reliability. However, when it comes to the board composition data about women, minorities and independent directors, the authors had to interpret by using pictures or curriculum vitae of the directors. The gender of the board members is usually recognizable when using pictures. Thus, it is clear and should not be inconsistent when other researchers collect these data. A little more difficult is the question about minorities. For that part, the biographies have to be considered, especially, when it comes to questions about backgrou nds and where the people grew up. This fact can lead to different interpretations depending on who collects the data. However, the definitions on ethnic minorities were made clear in the theoretical part and thus, the results should be very consistent. When deciding about the independence of the directors, there are two factors to consider. First, when the banks noted if the directors were independent, this was just copied for the research. These data are very reliable, because there is no space for interpretation. However, if it was not indicted and the biographies of the directors were read and the decisions about the independence were basically made after reading the professional background. Therefore, other researchers could have a different opinion about directors independence. To summarize, most of the data are very reliable, because they are published and just have to be copied. Only for factors, that the authors had to interpret, it could come to inconsistencies, which should be very limited though, because the data collection was done very diligently and clearly set definitions have been used. 4.3.2 Validity Validity of data is concerned with the question if the findings are what they appear to be. The researcher has to find out if the variables really have a causal relationship (Saunders et al., 2007, p.150). In this study, it was made clear through the theoretical background that board composition and company performance, as well as, risk propensity influence each other. This is mainly secured by the fact that the board of directors makes decisions which are intended to influence the financial results. However, to make sure that the relations between the board and firm performance and risk propensity are not accidental, four performance measures and two risk measures that were studied. A problem could arise, if the results are inconsistent. If that was the case, the contradicting results have to be interpreted and explained. 4.3.3 Generalisability The aim of this research study was to be able to generalize the results, which means to be able to apply the results to other settings (Saunders et al., 2007). Other settings could be for example a different group of banks or maybe other companies located in the same countries and also have international operations. For this reason, the quantitative research method was applied. To get reliable and valid results the authors collected almost 150 data sets by investigating 50 banks over three years. The number of banks was not selected by the authors but provided by a public resource which registered the 50 largest banks as measured by their total assets as of June 30, 2008. This amount of research data and the fact, that the sample includes banks situated on three continents in many different countries, allows the authors to generalize the findings. 5. Analysis 5.1 General Findings The sample of fifty banks consists of the largest financial institutions from North America, Europe and Asia. The biggest group are the European banks. One bank from the sample had to be excluded because its structure differed too much from the other banks and did not fit to the research questions. It was a state-owned bank which was controlled by politicians to a large percentage. Furthermore, for the year 2005, one more bank had to be left out, because it was just created in 2006 by a merger of two smaller banks. To be able to compare the banks, all performance indicators that were stated in different currencies have been converted into Euros with the currency rate of December 31 of each year.6 In table 5.1 general statistical values of the variables are listed. A value that was controlled for in the research was total assets. This was important for detecting if the board size or any of the other independent variables changed with the size of the bank. However, there is no significant relationship between the total assets and the size of the board (see table 5.2). Therefore, it is possible to say, that banks do not decide about the number of directors based on their size measured by total assets. It is rather noticeable that banks with a two-tier board system have larger boards than the other banks. The maximum number of directors came up to 48, when adding up the number of directors in the supervisory and the managing board compared to a minimum of seven board members in a one-tier system bo ard of directors. Table 5.1 Descriptive Statistics years 2005-2007 Samples Minimum Maximum Mean Median Std. Deviation Total Assets (Mio. à ¢Ã¢â‚¬Å¡Ã‚ ¬) 146 168,119 2,579,194 732,994 557,269 415,293 Board Size 146 7 48 18.36 17.00 6.92 Board Age (years) 142 49.1 64.70 57.32 57.74 3.44 Women (%) 146 0.00 42.90 10.02 6.8 9.32 Minorities (%) 146 0.00 41.20 3.73 0.00 7.79 Independent Directors (%) 146 0.00 94.10 49.15 50.00 28.80 Share Development cp. to Previous Year (%) 100 -44.75 106.67 14.61 16.90 27.52 ROE (%) 146 -37.90 37.50 14.56 15.35 8.60 ROA (%) 146 -0.30 1.75 0.65 0.59 0.40 Cost/Income Ratio (%) 125 34.70 114.00 58.89 56.00 13.07 Debt-Ratio (%) 146 87.73 98.54 95.31 95.95 2.19 B.I.S Capital (%) 135 8.5 19.70 11.79 11.60 1.72 5.2 Interdependency between the Independent Variables Before checking for the influence of the board composition on performance and risk propensity, the interdependency of the independent variables was evaluated in table 5.2. It is noticeable that many of the factors correlate with each other within the one percent significance level. The strongest correlation exists between the variables board age and percentage of independent directors. It shows that the higher the average age of the directors is the more independent directors are on the board. This leads to the conclusion that outside board members are usually older than executive directors. Another strong significant relationship is shown between the variables board age and board size. This correlation is negative and implies that the average age of the directors decreases when the number of board members increases. The reason for this link could be that the younger board members are introduced into the tasks before the older directors retire. One more noticeable factor is that the percentage of women on the board is positively correlated with the percentage of minorities on the board. This supports the results of Carter et al. in 2002. Furthermore, independent directors correlate significantly positively with women and minorities, which supports the conclusion that female and minority directors usually seem to be outsiders to the bank. An interesting fact is also that minority directors usually seem to be of more importance in smaller boards. The correlation between the board size and the percentage of minorities is slightly negative, which indicates that smaller boards have a higher percentage of ethnic minority directors. Furthermore, it is interesting that the boards of directors of banks do not significantly correlate with their total assets, as mentioned before. However, the board age has a very significant positive correlation with the assets. This implies that larger banks usually have an older board of directors. Besides, those banks also seem to have a slightly higher percentage of independent outside directors as shown by the positive correlation between these two factors. With the high interdependencies between the independent variables, it could come to multicollinearity problems in the regression analyses for the dependent performance and risk indicators. Fortunately, this is not the case as shown by the VIF-values, which are lower than 2.5, in the regression models below. Table 5.2 Correlations between Independent Variables 1 2 3 4 5 6 1 Total Assets a 1 2 Board Size a -.113 1 3 Board Age a .308*** -.403*** 1 4 Women (%) a -.052 .037 -.026 1 5 Minorities (%) a .098 -.176** .254*** .267*** 1 6 Indep. Directors (%) a .208** -.114 .465*** .369*** .308*** 1 *** Correlation is significant at the 0.01 level (2-tailed) ** Correlation is significant at the 0.05 level (2-tailed) a Pearson correlation coefficient 5.3 Influence of Board Composition on Performance Data At the beginning, the correlations of the data from all three years were evaluated together to get a general overview over the connections made between the independent and dependent variables. 5.3.1 Return on Assets When regarding the influence of the board composition on the first performance indicator it is very obvious that ROA is connected to all variables, except for total assets (see table 5.3). The strongest positive correlation exists between the variables return on assets and percentage of independent directors, followed by minorities. A little weaker connection is shown with the percentage of female directors and the board age. The linkages indicate that outsiders or a more divers and experienced board (concerning average