Culture and corporate governance as explanation for differences in the European banking sector

Culture and corporate governance as

explanation for differences in the

European banking sector

Some indicative signals for the solvency and liquidity levels of European banks

from the 2004-2008 period

N.W. Lummen

December 2009

UNIVERSITY OF GRONINGEN

Faculty of Economics & Business

MSc Finance

Supervisor:

1

Culture and corporate governance as

explanation for differences in the

European banking sector

Some indicative signals for the solvency and liquidity levels of European banks

from the 2004-2008 period

Abstract

Key words:

Banking sector, financial crisis, panel data, solvency,

liquidity, culture, corporate governance.

No. of words:

9758

Niels Lummen

Student no. s1386263

2

Preface

After several years of study, this thesis is the last part of completing my Master in Corporate Financial Management at the university of Groningen. Coincidently or not, at the start of my Master the world was in turmoil due to the financial crisis. Some banks went bankrupt, where others faced and are still facing difficulties in solvency and liquidity. Next to this, the bonus payments within the banking sector are subject to some heavy discussions. Media often used the term “grabbing culture of banking managers” to describe the bonus policies of banks. Given these circumstances the question raised to me what the underlying reasons for these problems were. Did culture and the direct environment of directors really played a role in the performance of banks over the past years? In this thesis I try to give an answer to this question, leading up to some indicative signals of why certain European banks had better solvency and liquidity levels compared to others.

I would like to thank my supervisor prof. dr. Robert van der Meer for his supervision and advice during the completion of my thesis. Furthermore, I would like to thank my parents, who have always supported me during my study. Without their support it would have been far more difficult to accomplish this.

3

Table of contents

2.1 The European banking sector p 6

2.2 Cultural differences p 8

2.2.1 Uncertainty Avoidance p 9

2.2.2 Power Distance p 9

2.2.3 Individualism/Collectivism p 10

2.2.4 Masculinity/ Feminity p 10

2.3 Corporate governance within the banking sector p 10 2.3.1 Relative variable payment of the board of directors p 12 2.3.2 Diversification of the board of directors p 13 2.3.3 Size of the board of directors p 13 2.3.4 Independence of the board of directors p 13

2 Data p 15 3.1 Control variables p 17 3 Methodology p 17 4.1 Multicollinearity p 19 4.2 Descriptive statistics p 19 4 Results p 21 5.1 Solvency results p 21 5.2 Liquidity results p 23 5.3 Discussions p 24 5 Conclusions p 27 6.1 Limitations p 29

6.2 Recommendations for further research p 29 6.3 Implications for the banking industry p 30

References p 31

4

1. Introduction

Since the financial crisis banks and financial institutions have received a lot of critics. After more than a year many economies are still facing difficulties caused by the financial crisis. The increase in the so called subprime mortgage defaults acted as a trigger for the crisis (Brunnermeier, 2008). This increase in defaults forced rating agencies as Moody’s, Standard & Poor’s and Fitch to put a “downgrade review” on US subprime deals and other tranches. Through this series of downgrading, banks were forced to write down large amounts of money on their mortgage based assets, causing the credit market to be nervous in the summer of 2007. This nervousness combined with a widespread concern about how to value structured products and an erosion of confidence in the reliability of ratings led the market for short-term asses-backed commercial paper to begin drying up (Brunnermeier, 2008). This caused banks to fall. IKB, a small German bank, was the first European victim after unable to provide promised credit lines. It was the first of a variety of market signals which showed that money market participants had become reluctant to lend to each other and as a result the first illiquidity wave started in the beginning of August 2007. The perceived default and liquidity risks of banks rose significantly, driving up the LIBOR. As a response to this the Central Banks injected billions of euro’s and dollars into the interbank market. At first these measures appeared to be working, but later on bank write-downs continued.

In spring 2008 investment banks all over the world were the first banks that could no longer secure their funding and central banks had to take further measures in order to prevent these investment banks from falling down. However, some banks did not survive this fallout (Bear Sterns), others survived at first by making heavily use of the credit facilities provided by the central bank of their country, like Lehman Brothers. However it was just a temporary solution. As in the case of Lehman Brothers, the company did not improve their balance sheet by providing new equity and therefore the share price eroded, causing the shares to plunge. Rescue plans failed because of a lack of trust, and Lehman Brothers went bankrupt. In addition to this, AIG, a large U.S. insurer, made clear in September 2008 that it faced a serious liquidity shortage, causing it share price to fell by over 90%. To prevent AIG from falling down, the Federal Reserve of the United States bought large equity stakes of the company. However, confidence levels in the financial sector dropped to zero and stock prices all over the world plummet, the financial crisis became visible for everyone (Brunnermeier, 2008).

5

problems. Banks are believed to have too much risky assets and products in their portfolios. As managers are responsible for the governance of banks, they are getting a large part of the blame. In addition to this, the payment schedules of these managers are a topic that has been discussed a lot in the past year. Managers could multiply their fixed salary through cash bonuses based on firm performance measures. However, these bonuses were perceived as extremely high, letting the media talk about the “grabbing culture” among bank managers. Malmendier and Tate (2005) already argued that certain forms of variable payment could lead managers to become overconfident. This overconfidence could be harmful for the company, in a way that it could lead to wrong decision-making of managers. Other research showed that cultural characteristics are a cause for the differences in variable payments (Tosi and Greckhamer, 2004). Furthermore, culture is believed to influence the way organizations are organized (Hofstede, 1991; 2001; Sondergaard, 1994). If these conclusions are right, it has to be that culture also has an influence on firm performance.

Within the context of performance, the corporate governance of banks played a crucial role as well (Bozec, 2005; Adams and Mehran, 2003). Therefore, in trying to find new evidence in why certain European banks were more liquid and/or solvent than others, this study will combine the factors of culture and corporate governance. Both factors are believed to have a certain impact on the decision-making process of managers and therefore on the performance of the bank.

6

2. Theoretical background

This section will describe by which various ways managerial decision making has been/ or still is influenced in the European banking sector. In order to see what has changed in the banking environment first a summary of the developments in the European banking sector will be discussed. Hereafter the possible linkages between culture and the managerial decision making of banking directors will be explained. To conclude a section of corporate governance and the possible linkage with decision making is described. These possible linkages are explained through both traditional finance theories as well as behavioral finance theories. As both of these theories have an impact on decision-making.

2.1 The European banking sector

Banks play an integral role in economies. They can provide the financing for commercial companies, access to payment systems and a variety of retail financial services for the economy in total. The large number of stakeholders, like employees, customers and suppliers, that depend on the well being of the banking industry is therefore large (Kern 2006). In Europe this seems to be even more the case in comparison to other economical powers like the United States and Japan (Dermine, 1996). Dermine shows this by the ratio of banking assets to gross domestic product (GDP) which is much higher in Europe than for example in the other economical powers. In contrary is the level of bond and equity market’s capitalization, which is lower for Europe compared to other economical powers. Of course the European banking sector is a wide concept, as Europe is characterized by a lot of different nations, each of them with their own laws and regulations. Furthermore the role played by banks in the national economy differs somewhat inside Europe as Mallin et al (2005) discuss in their article. They discuss the role played by banks in continental Europe and the United Kingdom and argue that continental European banks, as lenders to small en middle sized companies, traditionally have played a major role in corporate governance through equity-holdings, cross shareholdings and reciprocal board membership. In contrast, banks in the United Kingdom have not been major shareholders. Instead other financial companies like pension, insurance and mutual funds have taken this shareholder role. However, despite some of these differences, in whole of Europe the banking sector plays an important role.

7

decades to come to a single banking market. These efforts eventually lead up to the introduction of the universal banking model, which was adopted by the European Union. The universal banking model permits banks to undertake investment banking activities and leaves it to nation regulators to control financial conglomerates, the ownership structure of banks, and their relationship with the industry (Dermine, 1996). The adoption of the model not only had influence on the rules and legislation in the countries of the European Union, other non- member countries recognized the single banking market as well.

The increased monetary and financial integration initiated by the European Union were not the only factors that played a role in the deregulation processes of the European banking sector. According to Koutsomanoli- Fillipaki and Staikouras (2006) factors as: the gradual liberalization of capital flows, the rapid pace of developments in information technology, the product/service innovation in financial markets, the internationalization of banking activities, and the phenomenon of disintermediation played a role as well and are therefore prominent features of and characterizing the European banking sector. One of the major consequences of these deregulations was that it led to a large number of mergers and acquisitions among banks (Berger et al, 2005). Furthermore, these mergers and acquisitions took place in record levels compared to the past as Lepetit et al (2004) found in their study. Up to 1985, on average, there were fifteen mergers and acquisitions per year on the European market, which increased to 30-90 after 1991. The belief was that through gains of the acquired or merged banks, these could accrue through scale and scope economies, cost reduction, increased market power and reducing earning volatility. Another consequence of these mergers and acquisitions was that it led to more concentrated industries, characterized by a smaller number of larger banks that seem to have a wider set of services to clients (Cerasi et al, 2002). This continuous growing set of services made banking more complex in the past couple of decades. Furthermore, having a single market banking market for the European banking sector has some other downsides as well. For instance when facing a crisis in a single country this crisis affects the whole European banking market, as has became obvious from the current financial crisis (2008).

8

explained by cultural dimensions in the next section, followed by corporate governance features in the section thereafter.

2.2 Cultural differences

The previous section gave an insight in how the European banking sector is characterized. Despite the common characteristics of the banking sector, each bank is affected by cultural differences through the people that work for it. According to Gray (1988) not only national reporting requirements cause a variety in accounting and reporting practices across different nations, the distinctive cultural characteristics play a role as well. This section will try to explain possible linkages between bank characteristics and culture. As culture is far from an abstract concept it is difficult to define, not to mention making a statistical analysis with it. Sondergaard (1994) shows that many studies use the cultural framework of Hofstede (1991) to overcome this problem. Although the framework has been criticized on both empirical and theoretical grounds (e.g., one time, single company data; dimensions derived from factor analysis), on balance it has been largely validated and it provides a reasonable representation of national cultural attributes (Sondergaard, 1994). Given this validation this study will use the framework of Hofstede as well.

Hofstede (1980; 1991; 2001) has identified five major dimensions of national culture: uncertainty avoidance, individualism/collectivism, masculinity/ femininity, power distance and long term orientation. However, there is a note to this. The last dimension, long term orientation, was only added in 2001 and was not examined for all countries which were dealt with in his previous studies. Including this last dimension would mean that certain countries within Europe would be excluded from the sample and the sample would exist of only a few countries. In order to avoid this, the long term orientation dimension will be excluded from this study and the focus will lie on the other four dimensions. In appendix 1 one could find the selected countries and values belonging to their cultural dimensions.

9

affect managerial decision making among European banks? The next part will try to describe the possible linkage between the cultural dimensions and the decision making process of managers.

2.2.1 Uncertainty Avoidance

The concept of uncertainty avoidance concerns response to unstructured and ambiguous contexts. In high uncertainty avoidance cultures, members rely on clear procedures, well-known strategies, and well understood rules to reduce uncertainties and cope with their discomfort with unknown situations (Hofstede, 2001). In low uncertainty avoidance cultures, there is a greater tolerance for uncertainty. Members are relatively more at ease with unfamiliar situations, and presumably more tolerant of different ideas, approaches, and concepts. Based on this cultural dimension Nooteboom et al (1997) found that people and organizations in high uncertainty avoidance cultures are more sensitive to risk and perceive a higher level of risk and uncertainty in a given situation than in low uncertainty avoidance cultures. As low solvability and liquidity levels are seen as riskier it is to be expected that banks in cultures with high uncertainty avoidance have better solvability and liquidity levels. Therefore this study expects a positive relationship between high uncertainty cultures and the levels of solvability and liquidity of the banks in those cultures.

2.2.2 Power Distance

10

cash flow to meet debt repayments. Therefore a positive relation between the solvency and liquidity levels of banks and the power distance measured in the home country of the bank is expected.

2.2.3 Individualism/Collectivism

Individualism refers to the preferred level of individual freedom and opportunity. In organizational terms, individualist values have been linked to preferences for individual decision making over group consensus (Hofstede, 2001). Next to this, Hirshleifer and Thakor (1989) demonstrate that when managers care about their own performance, they would choose safer projects that have a higher probability of success. When firms cannot meet debt payments and go bankrupt, this is likely to be perceived as a failure of management. If managers in countries with a high individualistic focus are indeed more concerned with their own success, then less debt financing is expected. In respect to this study it therefore is to be expected that banks in high individualistic countries have higher solvency and liquidity levels.

2.2.4 Masculinity/ Feminity

Competitiveness, assertiveness, ambition, and the need to acquire material possessions are normally considered stereotypical masculine values. A value orientation with more emphasis on caring for others is described as more feminine (Hofstede, 2001). High masculine cultures admire the acquisition of material possessions and value aggressive attempts to acquire additional wealth or income. It could be that managers in high masculine cultures are blindfolded by acquiring additional wealth and/or income, thereby not paying attention to safeguard levels against bankruptcy. For this, this study expects a negative relationship between solvability and liquidity levels of a bank in high masculine cultures.

As mentioned before, this study will also look at what affects corporate governance characteristics could have on the decision making of managers. The next section will give an insight in these possible relationships.

2.3 Corporate governance within the banking sector

11

possible linkages between corporate governance and the decision making process will be explained.

In order to be efficient, good corporate governance of banks can help. Schleifer and Vishny (1997) define corporate governance for companies as dealing “with the ways that suppliers of finance to corporations assure themselves of getting a return on their investment”. However, according to Adams and Mehran (2003) corporate governance for banking firms may be different from that from unregulated, non financial firms. First, the number of parties with a stake in an institution’s activity complicates the governance of financial institutions. Non-financial firms have to deal with only investors who have a direct interest in the performance of the company. In contrast, banks have to deal with depositors and regulators as parties with a direct interest as well. Second, on a more aggregate level, regulators are concerned with the effect governance has on the performance of financial institutions, because the health of the overall economy depends upon their performance.

Additionally, Macey and O’Hara (2003) argue that the particular nature of banking causes it to be more at risk to greater moral hazard problems than non banking firms. Especially in the situation where a bank is at or near insolvency. In such a situation, the shareholders have a strong incentive to increase risk because they can allocate their losses to third parties while still receiving any gains that might result from the risky behavior. Companies outside the banking industry that are close to insolvency also have an incentive to take added risks. However, their ability to do so is limited by normal market forces and contractual obligations. Nonfinancial firms that are in financial distress usually have significant liquidity problems. Nearly insolvent banks, however, can continue to attract liquidity in the form of (government-insured) deposits. Governmental insurance eliminates the market forces that cause nonfinancial firms to run out of cash. Governments have attempted to replace these market forces with regulatory requirements such as capital requirements. These higher capital requirements force shareholders to put more of their money at risk, and this reduces moral hazard. In this way, capital requirements allow some parties with a direct interest, the regulators or deposit insurers, to impose restrictions on the shareholders. Another consequence of this moral hazard is that especially the directors of banks have to be careful in the amount of risk they will allow the bank to take.

12

have placed additional expectations on bank boards that represented their responsibilities even further. These usually take the form of laws, regulations, or guidance, and they generally reflect interest in safe and sound financial institutions. But the board of directors does not only exist because laws obligate firms to have it. The board of directors also serves as a market solution to an organizational design problem (Bozec, 2005). The board serves an essential role in resolving the contracting problem inside organizations. This problem essentially comes from the separation of ownership and control, which was first put into place by Jensen and Meckling (1976). More precisely, the dispersion of corporate ownership is supposed to give executives enough freedom and power to pursue objectives that could deviate from shareholders’ wealth maximization. Within this context, the board contributes to alleviating agency costs to the firm by monitoring and rewarding top executives to ensure wealth maximization to the shareholders.

2.3.1 Relative variable payment of the board of directors

13

2.3.2 Diversification of the board of directors

Other research in the field of overconfidence suggests that men are more prone to overconfidence than women, particular in the area of finance (Barber and Odean, 2001). This could have implications for the decision making process of the board of directors in the banking sector, as men are prone to make riskier decisions which influences the solvency and liquidity levels of banks. Next to this, other studies have argued that diversity of a board increases it effectiveness in monitoring and mitigating agency costs (Erhardt et al., 2003; Shrader et al., 1997, Collins et al, 2007). The explanation of this might be that diverse groups benefit from increased creativity and innovation and produce a variety of different perspectives that, in turn, substantially improve the quality of top level decision making (Maznevski, 1994). To the extent that effective board monitoring affects corporate performance (Fama and Jensen, 1983), more diverse boards could be contributing to the performance of an organization. A commonly used measure of board diversity is based on gender (Collins et al, 2007). Therefore this study expects a positive relationship between solvency and liquidity levels of banks and relative female presence in the board of directors.

2.3.3 Size of the board of directors

Conceptualizing the board of directors from the agency theory has not only been very useful to explain payments, it also was useful to explain the way boards are structured and how they function from this perspective, the main area of research deals with the effectiveness of the board as an internal monitoring control. Many studies have investigated the relationship between board characteristics such as size and independence and firm performance. When a board gets too big, agency problems increase, and directors are less effective in monitoring managers. Empirical results clearly support this proposition and suggest that board size and firm value are negatively correlated (Eisenberg et al., 1998). For this reason this study expects lower solvency and liquidity levels to occur in relation to the size of the board.

2.3.4 Independence of the board of directors

14

expect a positive relation between board independence, defined as the proportion of outside directors on the board, and firm performance. However, an alternative perspective would suggest a reliance on the preponderance of insiders. More precisely, inside directors possess superior information that could lead to more effective evaluation of top managers (Baysinger and Hoskisson, 1990). Furthermore, outside directors are usually part-time and may sit on a number of other boards. As a result, it may be difficult for them to understand the complexities of the bank. They may not have all the information necessary for decision-making. Considering the potential downsides of insiders and the complexness of the current banking industry as discussed in section 2 of this paper, higher board independence is expected to have a negative impact on firm performances. In extend to this study, as solvency and liquidity levels are used as measurements of firm performance, this study expects a negative impact of board independence on the solvency and liquidity levels of banks.

The assumptions that have been made in this and the previous literature section are summarized in table 2.1.

Table 2.1 Expected relations derived from the literature section.

Cultural dimensions Solvability Liquidity Literature section

Uncertainty avoidance positive positive page 9 Power distance positive positive page 9 Individualism positive positive page 10 Masculinity negative negative page 10

Board characteristics

Variable cash payment negative negative page 12 Board diversification positive positive page 13

Board size negative negative page 13

15

3. Data

The sample used for testing contains data of fifty listed European banks from seventeen different countries over the period 2004-2008. Data was obtained from the annual reports of the banks, BANKSCOPE and the cultural dimensions derived from Hofstede (2001). As the data consists both of time series and cross-sectional elements, panel data has been used (Brooks, 2008). The period 2004-2008 was chosen as those were the latest years prior to the crisis. Next to this, 2004 is for most banks the first year of using IFRS accounting standards. This universal accounting code makes it easier to compare accounting measures between companies of different countries (Choi and Meek, 2005). When one would examine the period before 2004 one has to consider the disadvantages of comparing accounting numbers of different local GAAP rules. Furthermore, an advantage of this period is that characteristics of banks were relatively stable. However, even in this relatively short time period the sample was influenced by some changes. These changes were caused by mergers or acquisitions, for example through nationalization of banks by governments or through mergers among private banks. Especially in the year 2008 these changes occurred, as some of the consequences of the financial crisis already became clear.

The choice for only listed banks is made based upon the availability of their data. Listed and non listed companies have different regulations regarding their openness and access to their financial statements. Only banks listed on the main or second stock index of their country were selected to make sure that the selected banks are important for their national economy. Some of the chosen banks are connected to each other through banking groups, as for example the UNI Credit Group. However, all banks used in the sample do have their own listings and their own annual reports.

As this study examines what variables influence managerial decision making and through that the solvability and liquidity levels among European banks, this study will have these variables as dependent. Solvability will be measured as the ratio between equity and liabilities as been used in the database of BANKSCOPE. Liquidity is defined as the ratio between liquid assets divided by total debt and borrowings, as been used in the database of BANKSCOPE as well. Variable cash payment will be measured as the annual cash bonus (excluding stock and pension oriented payments) received by the CEO of the company divided by the fixed payment received by the CEO. These data were obtained from the annual reports of the selected banks.

16

discussed in the literature section are picked. The cultural dimensions belonging to the country where the bank has its origin and main listing are representing the culture which the bank is subject to. However, this decision brings some difficulties with it. First, some banks might have their major listing in a specific country, but their country of origin might be different. In these cases this study looked where and for what period the headquarters of the bank was located. When the location of the headquarters of the bank was in the same country as their listing for more than ten years, the cultural values of this country counted. When the period was less than ten years, which did not occurred, the cultural values of the country of origin counted. Next to this, most large banks are internationally active and have a lot of foreign employees. This might influence the culture within a bank. However, as this would lead up to a very complex situation, these influences are not taken into account. Appendices 1 and 2 show the chosen banks, their home countries and the values of the cultural dimensions belonging to these countries.

To test whether corporate governance influences managerial decision making some board characteristics will be used. Within Europe there are different corporate governance styles, mainly through the existence of one tier ore two tier boards’ structures. However, since the number of large banks is somewhat small within Europe, this distinction will not be made in this paper. Instead, in two tier board countries this study will look at the supervisory board, whereas in one tier board countries it will look at the board of directors, as both have the monitoring function in the organization (Lederer, 2006). Furthermore, the boards of directors of companies in countries with one tier board structures have sometimes difficulties to distinguish between supervision and management, for example by the remuneration of directors. To overcome this problem committees are formed. As these committees consist solely or for the majority of outside members of the board of directors this study will not make a distinction regarding these committees. Given this, and following from the literature section, the following variables will be used as board characteristics.

-Varfix, this variable will represent the annual cash bonus received by managers and is measured as the annual cash bonus (excluding stock and pension oriented payments) received by the CEO of the company divided by the fixed payment received by the CEO.

-Female, this variable will represent board diversity and is measured as the number of females in the board divided by the size of the board.

17

-Outside, this variable will represent the board independence and is measured as the number of people in the board who are not fulltime managers or employees of the bank divided by the size of the board.

3.1 Control variables

Next to the board characteristics and culture there will be tested for two control variables, namely size and profit. Previous research has shown that size has a positive relationship with financial leverage (Panno, 2003; Cooke, 2001), thereby lowering solvency levels. Furthermore, it is believed that larger firms are more diversified, have easier access to internal and external sources of funds and are therefore more liquid (Titman and Wessels, 1988; Audretsch and Elston, 2002). The natural logarithm of total assets, retrieved from BANKSCOPE, is considered to be a proxy for size. As not all banks publish their annual reports in Euros all foreign currencies were translated to Euros using the end of the year exchange rates derived from the IMF website1.

Furthermore, the profit level could influence the solvency and liquidity levels of firms as well. Profitable firms often use their earnings to repay debt and are therefore less levered (Titman and Wessels, 1988). In addition, firms tend to issue equity following an increase in stock prices, reducing their leverage even more (Masulis and Korwar, 1986). Next to this Opler et al (1999) found evidence that firms that do well hold more liquid assets than other firms. As a reason for this they conclude that firms that do well want to be able to keep investing when cash flow is too low, relative to investment, and when outside funds are expensive. The return on assets is considered to be a proxy for profit, again this variable is retrieved from BANKSCOPE. In appendix 3 one may find all the different variables, their description and their source summarized.

4. Methodology

The hypnotized relations between culture and board characteristics at one side and the solvency and liquidity of banks at the other will be tested by a regression analysis. Data consists both of time series and cross-sectional elements, also called panel data. For conducting a regression analysis with panel data the technique of generally least squares

1

18

(GLS) is used (Brooks, 2008). The use of panel data has some advantages above only using cross-sectional or time series data. Brooks (2008) argues that with the use of panel data a broader range of issues and more complex problems can be tackled. Next to this one can examine how variables, or the relationships between them, change dynamically over time. A third advantage of using panel data is that it can reduce the effects of omitted variable bias or unobserved heterogeneity in the results. As this study both looks at the solvency and liquidity levels of banks there are two models. Model 1 will represent solvency as independent variable, whereas model 2 will have liquidity as independent variable.

Brooks (2008) identifies two models of panel data regression methods. The fixed effects model and the random effects model2. Verbeek (2004) states that the fixed effects approach is more appropriate if the units in the sample are ‘one of a kind’, such as banks, and cannot be viewed as a random draw from some underlying population. In order to see whether this assumption is right one has to perform a likelihood ratio test and a Hausman test (Brooks, 2008). The Hausman test is the statistical test to determine whether the fixed effect and random effect estimators are significantly different. This can be proven by inspecting the p-value of the chi-squared statistic (in EVIEWS 6.0). If the p-p-value is less than the 1 percent level of significance the null hypothesis (of both the random and the fixed effect estimators to be consistent, but the random effect estimator is more efficient) is rejected in favor of the alternative hypothesis (only the fixed effect estimator is consistent) (Brooks, 2008). As can be seen in appendix 4 and 5 for all models the null hypothesis of the Hausman test is rejected, therefore this study will use the fixed effect model with both time and cross-sectional items to be fixed, given they are all significant at the 1 percent significance level. Therefore equation 4.1 will be tested with both time and cross-sectional fixed effects, where solvency respectively liquidity are representing Υ_it; and where i = 1,2,3, …,N; t = 1,2,3, …, T).

(4.1) it it it it it it it it it it it it it

ROA

LOGASSETS

OUTSIDE

BOARDSIZE

FEMALE

VARFIX

MAS

IND

POWER

UNC

ε

β

β

β

β

β

β

β

β

β

β

α

+

+

+

+

+

+

+

+

+

+

+

=

Υ

See appendix 3 for a detailed description of the variables.

2

19 4.1 Multicolliniarity

When performing a regression analysis it is important to look at possible multicollinearity between the independent variables as it could influence the results (Brooks, 2008). Table 3.1 gives a correlation matrix for the independent variables.

Table 4.1 Correlation matrix of the independent variables

1 2 3 4 5 6 7 8 9 10 1 UNC 1 2 IND -0.573 1 3 MAS 0.209 -0.003 1 4 POWER 0.753 -0.194 -0.072 1 5 VARFIX 0.195 -0.122 0.347 -0.006 1 6 BOARD SIZE 0.232 -0.030 0.265 0.126 0.076 1 7 FEMALE -0.510 0.095 -0.355 -0.438 -0.227 -0.009 1 8 OUTSIDE -0.239 0.184 -0.256 -0.274 0.065 -0.627 0.153 1 9 LOGASSETS 0.052 0.205 0.092 0.112 0.147 0.403 0.019 -0.096 1 10 _ROA 0.068 -0.221 -0.015 0.069 0.091 0.024 -0.047 -0.146 -0.306 1

UNC, IND, MAS and POWER are respectively uncertainty avoidance, individualism, masculinity and power distance numbers belonging to the cultural dimensions of Hofstede (2001). VARFIX is the ratio between variable CEO cash payment and CEO fixed payment, BOARD SIZE is the number of people sitting in the board of directors, FEMALE is the percentage females in the board, OUTSIDE measures the board independence, LOGASSETS is the natural logarithm of total assets and ROA is return on assets. See appendix 3 for detailed information.

Highly correlated variables, above 0.7, should be taken separately into the regression analysis (Brooks, 2008). In this case only power distance and uncertainty avoidance are considered to be too highly correlated. From other research it appears to be no exception to see these two to be highly correlated (Li and Harrison, 2008). Therefore power distance and uncertainty avoidance will be taken separately into the regression analysis. Giving each model two subsections where on the one hand the variable power distance is excluded and at the other the variable uncertainty avoidance.

4.2 Descriptive statistics

20

Table 4.2 Descriptive statistics

Total sample Mean per year

Mean Std. Dev. 2008 2007 2006 2005 2004 UNC 60.181 23.703 60.63 61.68 61.39 59.23 57.42 POWER 42.484 16.474 42.53 43.66 43.53 41.97 39.54 IND 70.538 12.515 69.74 70.05 70.42 71.14 71.08 MAS 46.885 23.431 48.87 47.73 46.13 47.57 47.12 VARFIX 1.123 1.048 0.22 1.30 1.64 1.33 1.27 FEMALE 0.134 0.115 0.15 0.13 0.13 0.14 0.12 BOARDSIZE 14.835 4.736 14.26 15.02 15.21 15.17 15.08 OUTSIDE 0.635 0.250 0.66 0.63 0.62 0.62 0.63 LOGASSETS 12.684 1.225 12.84 12.75 12.61 12.58 12.75 ROA 0.006 0.010 0.00 0.01 0.01 0.01 0.01 SOLVENCY 4.682 2.544 4.59 5.07 4.71 4.51 4.40 LIQUIDITY 17.457 14.797 11.79 18.73 20.82 20.96 14.09

UNC, POWER, IND and MAS are respectively uncertainty avoidance, power distance, individualism and masculinity numbers belonging to the cultural dimensions of Hofstede (2001). VARFIX is the ratio between variable CEO cash payment and CEO fixed payment, FEMALE is the percentage females in the board, BOARD SIZE is the number of people sitting in the board of directors, OUTSIDE measures the board independence, LOGASSETS is the natural logarithm of total assets and ROA is net income divided by total assets. Solvency is equity divided by liabilities and Liquidity is the liquid assets divided by total debt and borrowings. See appendix 3 for a detailed description.

21

5. Results

This section will discuss the results derived from the dataset after performing the GLS regression analysis. First the results regarding solvency of European banks as dependent variable will be presented, where after the results for liquidity will follow in section 2 of this chapter. As third the conclusions based upon these results will be shown in section 5.3.

5.1 Solvency results

Table 5.1 shows the results of the GLS regression analysis of equation 1 with the solvency of European banks as independent variable. The explanatory power of the model is quite high, given the levels of R² and the adjusted R² to be around 0.5 (Brook, 2008). Furthermore the F statistic and its probability reject the null hypothesis of all variables being zero.

Table 5.1 GLS regression analysis with the fixed effect model for solvency levels at European banks Solvency Model 1.1 Model 1.2

Coefficient T-statistic Coefficient T-statistic

C 18.2128 9.8228*** 17.8666 8.7810*** UNC 0.0135 1.4683 POWER 0.0204 1.9504* IND -0.0104 -0.8650 0.0004 0.0257 MAS 0.0215 2.9632*** 0.0172 2.4817** VARFIX -0.0169 -0.1124 -0.0281 -0.1858 FEMALE 1.8005 1.2330 1.6448 1.0703 BOARDSIZE -0.1024 -2.4218** -0.1198 -2.7324*** OUTSIDE -1.4547 -1.8668** -1.8773 -2.4391** ROA 58.5905 3.9897*** 61.2632 4.1456*** LOGASSETS -1.0115 -7.2961*** -0.9817 -7.1342*** R-squared 0.4894 0.4846 Adjusted R-squared 0.4627 0.4576 F-statistic 18.3188 17.9679 Prob(F-statistic) 0.0000 0.0000 Observations 182 182

22

As one can see there is evidence of culture influencing the solvency levels of European banks. Both the power and masculinity dimensions have significant positive relationships with the solvency levels of European banks. The power dimension shows a positive significant relationship with solvency levels at the ten percent level, which is in line with the expectations made in the literature section. Next to this, a high score on the masculinity dimension has a positive impact on the solvency levels of European banks as well. This relationship is significant at a one percent level in model 1.1 and at a five percent level in model 1.2. This positive relationship is contradicting the expectations made in the literature section, although the impact of masculinity is not very big given the numbers of the coefficients. A possible explanation for this could be that, like a high score on the power dimension, managers in high masculine cultures do not want their investment decisions to be bounded by debt covenants. Thereby preferring equity offerings instead of debt issues when attracting capital. Too much debt financing could be seen as a failure in governance of management, which would mean that managers failed in their competitiveness and ambitions. Further research should clarify this. The other cultural dimensions even have smaller coefficients than the masculinity and power dimensions, next to this they are not significant, making it hard to derive any conclusions from them.

When looking at the board characteristics one can see that the percentage of outside directors has a negative influence on the solvency levels of banks. This relationship is significant for both models at the five percent level and in line with the hypothesis. Higher board independence does have a negative influence on the solvency levels of banks. Next to the board independence, the size of the board matters as well. In line with the expectations made in the literature section, banks with larger boards have lower solvency levels compared to smaller boards. These relationships are significant at the five percent level in model 1.1 and at the one percent level in model 1.2. The positive relationship between board diversification and solvency as expected from the literature is shown in the coefficient of FEMALE, however this is not significant.

23 5.2 Liquidity

Table 5.2 shows the GLS regression analysis for liquidity levels at European banks. The R² and the adjusted R² levels are lower than in the solvency regression analysis, but for financial modeling they are still quite high (Brooks, 2008).

Table 5.2 GLS regression analysis with the fixed effect model for liquidity levels at European banks Liquidity Model 1.1 Model 1.2

Coefficient T-statistic Coefficient T-statistic C -51.7815 -3.8170*** -46.8860 -3.1817*** UNC -0.0426 -0.6382 POWER 0.0050 0.0657 IND 0.1099 1.2562 0.0633 0.5643 MAS 0.0221 0.4162 0.0208 0.4133 VARFIX 2.8840 2.6460*** 2.8729 2.6420*** FEMALE 28.7290 2.5511** 24.2278 2.0770** BOARDSIZE -0.1504 -0.4897 -0.0979 -0.3082 OUTSIDE -2.1050 -0.3700 -1.7798 -0.3170 ROA 74.0291 0.6960 69.4089 0.6521 LOGASSETS 4.4241 4.3789*** 4.4934 4.4979*** R-squared 0.2249 0.2268 Adjusted R-squared 0.1834 0.1854 F-statistic 5.4177 5.4755 Prob(F-statistic) 0.0000 0.0000 Observations 178 178

The *,**,*** represent the statistically significance at respectively the 10%, 5% and 1% level. UNC, IND, MAS and POWER are respectively uncertainty avoidance, individualism, masculinity and power distance numbers belonging to the cultural dimensions of Hofstede (2001). VARFIX is the ratio between variable CEO cash payment and CEO fixed payment, FEMALE is the percentage females in the board, BOARD SIZE is the number of people sitting in the board of directors, OUTSIDE measures the board independence, LOGASSETS is the natural logarithm of total assets and ROA is net income divided by total assets. Solvency is equity divided by liabilities and Liquidity is the liquid assets divided by total debt and borrowings. See appendix 3 for a detailed description.

24

hypothesis of board diversification improving the liquidity levels of banks. The relationship between the annual cash bonus of the CEO and the liquidity levels is in contradiction to the stated hypothesis. The bonus payment has a significant positive influence on the liquidity levels of both models. A logical explanation for this would be that well performing banks, in terms of liquidity, would express this by paying bigger bonuses than less performing banks.

Again LOGASSETS, as a proxy for size, has a significant positive influence in both models at the 1 percent level. This is in line with previous research, where larger firms are believed to be more diversified and have an easier access to internal and external sources of funds and (Titman and Wessels, 1988; Audretsch and Elston, 2002). The control variable for profit, ROA, has a positive coefficient but is not significant.

5.3 Discussions

Several conclusions can be drawn from this section. From the regression analysis it follows that there are reasons to believe culture has influence on solvency levels, mainly through Hofstede’s dimensions of masculinity and power (2001). High scores on the masculinity and/or the power dimension have a positive influence on the solvency levels of European banks. Managers within cultures which score high on the power dimension apparently do not want to be bounded by debt covenants, which would undermine their power, when making a strategic investment decision.

The positive found relationship between high masculine cultures and higher solvency levels is contradicting the expectations made in the literature section. A possible explanation for this could be that, like a high score on the power dimension, managers within high masculine cultures do not want their investment decisions to be bounded by debt covenants. Too much debt financing could be seen as a failure in governance or management, which would mean that managers failed in their competitiveness and ambitions. Therefore managers within high masculine cultures could prefer equity offerings instead of debt issues when attracting capital. Further research could clarify this. The other cultural dimensions even have smaller coefficients than the masculinity and power dimensions, next to this they are not significant, making it hard to derive any conclusions from them.

25

banking industry, given the fact that the number of banks in Europe is limited in comparison to the number of firms in other industries.

Regarding board characteristics that could influence managerial behavior, the results show that board independence has a significant influence on solvency levels, whereas it does not have a significant influence on the liquidity level of European banks. As expected, board independence has a negative impact on the solvability levels of European banks. This could be a result of the complexness of the banking sector combined with high board independence. As stated in the literature section, outside board directors could have less information in the complex world of banking, therefore not realizing all the potential risks that banks are subject to. As this study only looks at outside directors as the ones who are not fulltime manager or employee of the bank there is a downside to this conclusion. From the data it does not become clear in which industries the outside board directors do have their main focus. Further research should clarify this by looking more detailed into the exact work or background of the outside directors, to see whether they indeed are not specialized enough to perform a board of directors function in such a complex world as the banking sector.

In compliance with other studies, which state that larger boards are less efficient (Eisenberg et al, 1998; Core et al, 1999), this study also found a negative impact of board size on the performance of European banks. Especially in terms of solvency, banks with larger boards were less solvent than banks with smaller boards. It could be that within larger boards too many persons had to be convinced of potential risks that banks were facing, thereby not keeping in mind solvency safeguards. This decrease in board efficiency is notable in terms of liquidity as well, however it is not significant.

26

Counter intuitive, a positive relation between variable annual payment and liquidity was found. It was expected that bank directors could be overconfident through high variable payments (Malmendier and Tate, 2005) and thereby taking too much risks which would have a downside on the liquidity levels of banks. Since the opposite is true, this explanation does not hold. There is reason to believe that well performing banks in terms of liquidity would show this by paying larger bonuses.

Next to this all, size has by far the biggest and most significant influence on both dependent variables. It shows that larger banks were less solvent than smaller banks. This is consistent with previous research of Panno (2003) and Cooke (2001), who found that size is positively related to leverage. Warner (1977) already stated that this could be due to that the ratio for bankruptcy costs to the value of the firm decreases as value increases with size. This suggests that the impact of the direct costs of bankruptcy on borrowing decisions of large firms is negligible. Furthermore, it has been argued that larger firms have easier access to capital markets and borrow at more favorable interest rates (Ferri and Jones, 1979). However, as a consequence of banks getting so large it could have been that banks have been grown too big, too fast, making them too complex to control and thereby losing sight of potential downsides of these rapid growths.

27

6. Conclusion

This study examined which variables influenced managerial decision making and through that the solvency and liquidity levels of European banks in the past five years. In order to do so, both cultural and corporate government characteristics were included and data from fifty European banks of seventeen different countries were collected over the period 2004-2008.

Given the financial crisis there is a lot of attention on the way banks are working nowadays, especially the bonus culture among banking managers has been heavily discussed. This study tried to explain in which way cultural and corporate governance characteristics could have had an impact on this. Previous research has shown that cultural characteristics have influenced the way organizations are organized (Hofstede, 2001; Sondergaard, 1994) and influenced the payments of managers (Tosi and Greckhamer, 2004). Furthermore, corporate governance also has been considered to play a crucial role within the performance of banks (Bozec, 2005; Adams and Mehran, 2003). In combining these two factors this study found new evidence why certain European banks were healthier in terms of liquidity and/ or solvency than others. Table 6.1 summarizes the findings of this study and the expected relations between the dependent and independent variables.

Table 6.1 Expected relations derived from the literature section and the results as found by this study.

Cultural dimensions Solvability Results Liquidity Results Literature section

Uncertainty avoidance positive ~ positive ~ page 9

Power distance positive * positive ~ page 9

Individualism positive ~ positive ~ page 10

Masculinity positive ***↔ negative ~ page 10

Board characteristics

Variable cash payment negative ~ negative ** ↔ page 12 Board diversification positive ~ positive ** page 13

Board size negative *** negative ~ page 13

Board independence negative ** negative ~ page 13

The *,**,*** represent the statistically significance at respectively the 10%, 5% and 1% level. The “~” indicates that this study did not find a statistically significance relation. The “_↔” indicates that the found relation is contradicting the expected relation derived from the literature section.

28

power distance cultures (Hofstede, 2001) do not want to be bounded in their decision making by debt covenants. As this would undermine their power and authority (power dimension), or it can be seen as bad performance of management, thereby showing that management is less competitive and could not fulfill ambitions (masculinity dimension). The other cultural dimensions, uncertainty avoidance and individualism, of Hofstede (2001) did not appear to have a significant influence on solvency and liquidity levels.

Regarding board characteristics, which could influence management’s decision making, this study shows that higher board independency, measured as relative more people who are not full time employee or manager of the bank, the less solvent a bank is. Outside directors often have less information about the company and given the complexness of the modern banking sector it could be that outside directors were not realizing all the potential risks that banks were and are subject to. Additionally, the more diverse a board of directors is, the more liquid a bank is. More diverse groups benefit from increased creativity and innovation and produce a variety of different perspectives which leads to improve the quality of decision making at the top level (Maznevski, 1994). This could also have been the case in the European banking sector over the past five years, where more diverse boards wanted better safeguards against possible risks and/ or more diverse boards saw more potential risks. Next to this, board size has a significant negative influence on the solvency levels of European banks. Other research found that board size is negatively related to firm performance due to a decrease of board efficiency (Eisenberg et al, 1998; Core et al, 1999). Given the found relation between board size and the solvency levels of European banks, it could be that a decrease of board efficiency played a role in recognizing potential risks of having lower solvency levels at European banks as well. Furthermore, the ratio of bonus payment to the fixed payment of the CEO was positively related with the liquidity levels of banks. These findings are not in line with the expectations made in the literature section. A possible explanation for this could be that liquidity levels are not influenced by the relative bonus amount of the CEO, but that more liquid banks could easier pay bonuses and/or pay relatively higher bonuses as they have the resources to do so.

29

couple of decades, making them a lot more complex. This complexness could cause them to be too difficult to control for managers, as it was too hard to see all potential risks.

6.1 Limitations

Some limitations on this study have to be considered. As the European banking sector was subject to deregulations processes and a lot of mergers have taken place in the last couple of decades, the number of banks in Europe has decreased severely. Making it hard to get a large sample size with comparable banks. Furthermore, not all banks gave sufficient information about their policies regarding variable payment, decreasing the sample size even more. Expanding the research by including other geographical areas could solve this, but one then has to consider regulations. These are already different within Europe, not to mention the differences in regulations between banks of different continents.

There is a considerable limitation concerning culture, given the seventeen cultures included in the sample. The diversity among the different national cultures may be too small to notice and therefore making it hard to make any conclusion with it. Especially in the case of the individualism dimension, as almost all of the European countries which are represented in the sample, show a high individualism number.

Regarding the board characteristics there are some limitations as well, mainly through to different regulations. In some countries a two tier board is obligated, whereas in other countries single tier boards are. This could have influenced the results, but given the sample size it was not an option to make a distinction between these.

6.2 Recommendations for further research

This study is one of the first to combine cultural and corporate governance factors to explain differences among banks. The results show that these factors can explain some differences and therefore making it interesting for further research. As cultures in Europe are still somewhat close to each other, including other cultures could give better and more diversified results. Next to this, given the relatively small number of banks in Europe it would be interesting to see whether the same results hold in other industries, with more companies representing each culture. Furthermore, it could be that in other industries other cultural dimensions play bigger roles to explain possible linkages.

30

study. In addition to this, a closer look to outside directors in a board is recommended. As the banking sector is a complex one, it could make a difference whether outside directors have a background in this sector or in complete other ones.

6.3 Implications for the banking industry

After the nationalization of the ABN AMRO bank there has been a lot of discussion in the Dutch media about the formation of the new supervisory board and the appointment of Mr. Zalm as CEO of the bank. Since this study looked at circumstances that could have influenced the solvency and liquidity levels of European banks from a cultural and corporate governance perspective, the results of this study could add some interesting features to this discussion. Stating that high solvency and liquidity levels are perceived as better and given the results derived from this study, the ideal supervisory board of ABN AMRO would consist of persons from a masculine culture in combination with a high score on the power distance dimension (Hofstede, 2001). Next to this, the board has to be a relatively small, diversified one, consisting of mainly directors who are managers or fulltime employees of the bank.

31

References

Adams, R. and H. Mehran, 2003, “Is Corporate Governance Different for Bank Holding Companies?”, Economic Policy Review, Vol. 9, pp 123–142

Audretsch, D.B., and J.A. Elston, “Does firm size matter? Evidence on the impact of

liquidity constraints on firm investment behavior in Germany”, International Journal of Industrial Organization, Vol. 20, pp 1-17.

Baysinger, B.D. and R.R. Hoskisson (1990), “The Composition of Boards of Directors and Strategic Control: Effects on Corporate Strategy”, Academy of Management Review, Vol. 15, pp 72–87

Barber, B. and T. Odean, 2001, “Boys will be boys: gender, overconfidence and common stock investment”, The Quarterly Journal of Economics, Vol. 2, pp 261- 292

Barro, J., and R. Barro, 1990. “Pay, performance and turnover of bank CEOs”, Journal of Labor Economics, Vol. 8, pp 448–481

Berger, A., Clark, G., Cull, R., Klapper, L. and G. Udell, 2005, “Corporate governance and bank performance: A joint analysis of the static, selection, and dynamic effects of domestic, foreign, and state ownership”, Elsevier, Vol. 29, pp 2179-22

Bozec, R., 2005,”Boards of Directors, Market Discipline and Firm Performance”, Journal of Business Finance & Accounting, Vol. 32, pp 1921-1959

Brooks, C.,2008, “Introductory Econometrics for Finance”, Second edition, New York: Cambridge University Press.

Brunnermeier M., 2008, “Deciphering the 2007-08 Liquidity and Credit Crunch,” Journal of Economic Perspectives, Vol. 23, No. 1, Winter 2009, pp 77-100

32

Choi, F.D.S. and G.K. Meek, 2005, “International accounting”, Fifth edition, Pearson education international

Chui, A., Lloyd, A. and C. Kwok, 2002, “The Determination of Capital Structure: Is National Culture a Missing Pieceto the Puzzle?” , Journal of international business studies, Vol. 33, No. 1, pp 99-127

Collins, M., Hagendorff, J. and K. Keasy, 2007, “Bank Governance and Acquisition

Performance” Corporate Governance, Blackwell publishing ltd., Vol. 15, No. 5, pp 957-968

Cooke, T.E., 1991, ‘An assessment of voluntary disclosure in the annual report of Japanese corporations’, The International Journal of Accounting, Vol. 26, No. 3, pp. 174–89.

Core, J., R. Holthausen and D. Larcker, 1999, ‘Corporate Board Effectiveness, Chief Executive Officer Compensation, and Firm Performance’, Journal of Accounting and Economics, Vol. 51, pp. 371–406.

Dermine, J., 1996, “Professional forum European banking integration, ten years after”, European Financial Management, Vol. 2, No. 3, pp 331-353

Eisenberg, T., S. Sundgren and M. Wells, 1998, “Larger Board Size and Decreasing Firm Value in Small Firms”, Journal of Financial Economics, Vol. 48, pp. 35–54.

Erhardt, N. L., Werbel, J. D. and Shrader, C. , 2003, “Board of Director Diversity and Firm Financial Performance”, Corporate Governance: An International Review, Vol. 11, 102–111

Fama, E.F. and M.C. Jensen,1983, “Separation of Ownership and Control”, Journal of Law and Economics, Vol. 26, pp 301–49

Ferri, M. G. and Jones. W. H., 1979, Determinant of financial structure: a new methodological approach, Journal of Finance, Vol. 34, pp 631- 644.

33

systems internationally’’, Abacus, Vol. 24, No. 1, pp. 1-15

Grossman, S. and O. Hart, 1983, “ Implicit contracts under asymmetric information”, Quarterly Journal of Economics,Vol.98, pp 123–157

Harrison, R. and J. Li, 2008, “National Culture and the Composition and Leadership Structure of Boards of Directors”, Corporate governance, Vol. 16, No. 5, pp 375-385

Hirshleifer, D. and Thakor, A., 1989, Managerial reputation, project choice and debt. Working paper No.14, Anderson Graduate School of Management at UCLA

Hodgetts, R. M. and Luthans, F. , 1993, ” U.S. multinationals’ compensation strategies for local management: Cross-cultural implications” , Compensation and Benefits Review, Vol. 25, pp 42–48.

Hofstede, G., 1980, Culture’s Consequences: International Differences in Work-related Values, Sage, Newbury Park, CA.

Hofstede, G., 1991, Culture and Organizations: Software of the Mind, McGraw-Hill, London.

Hofstede, G., 2001, Culture’s consequences: Comparing Values, Behaviors, Institutions, and

Organizations across nations., second edition, Sage publications, Thousand Oaks, CA.

Holmstrom, B., 1979, “Moral hazard and observability”, The Bell Journal of Economics, Vol. 10, pp 74–91.

Houston, J. and C. James, 1995, ”CEO compensation and bankrisk- is compensation in banking structured to promote risk taking? ”, Journal of Monetary Economics, Vol. 36, pp 405–432

34

Kern, A., 2006, “Corporate governance and banks: The role of regulation in reducing the principal-agent problem.” Journal of Banking regulation, Vol. 7, pp 17-40

Koutsomanoli-Fillipaki, A. and C. Staikouras, 2006, “Competition and concentration in the new European banking landscape”, European Financial Management, Vol. 12, No. 3, pp 443– 482

Lederer, P., 2006, “A comparative analysis of the liability of non-executive directors in the UK and of members of the supervisory board in Germany”, European business law review, Vol. 17, Issue 6, pp 1575- 1613

Lepetit, P., Patry, P. and P. Rous, 2004, “Diversification versus specialization: an

event study of M&As in the European banking industry”, Applied Financial Economics, 2004, Vol. 14, pp 663–669

Li, J. and R. Harrison, 2008,”National Culture and the Composition and Leadership Structure of Boards of Directors”, Corporate Governance: An International Review, Sep2008, Vol. 16, Issue 5, p375-385

Macey, J. and M. O’Hara, 2003, “The Corporate Governance of Banks”, FRBNY Economic Policy Review, pp 91-107

Mallin, C., Mullineux, A. and C. Whilborg, 2005, “The Financial Sector and Corporate

Governance: the UK case”. Corporate Governance, Blackwell publishing ltd., Vol. 13, No. 4, pp 532-541

Malmendier, U. and G. Tate, 2005, “Does Overconfidence Affect Corporate Investment? CEO overconfidence Measures Revisited”, European Financial Management, Vol. 11, No. 5, pp 649–659

35

Masulis, R., and A. Korwar., 1986, "Seasoned Equity Offerings: An Empirical Investigation." Joumal of Financial Economics, Vol. 15 , pp 91-118.

Nooteboom, B., Berger, H. and Nooderhaven, N. G., 1997, “Effects of trust and governance on relational risk”, Academy of Management Journal, Vol. 40, No. 2, pp 308–338.

Opler, T., Pinkowitz, L, R. Stulz and R. Williamson, 1999, “The determinants and implications of corporate cash holdings”, Journal of Financial Economics, Vol. 52, pp 3-46

Panno, A., 2003, An empirical investigation on the determinants of capital structure: the UK and Italian experience, Applied Financial Economics, Vol. 13, pp 97–112.

Shleifer, A., and R. Vishny. 1997. “A Survey of Corporate Governance.” Journal of Finance, Vol. 52, No. 2, pp 737-84

Shrader, C, Blackburn, V. and Iles, P., 1997, “Women in Management and Firm Financial Performance: An Exploratory Study, Journal of Managerial Issues, Vol. 9, pp 355–372

Sondergaard, M., 1994, “Hofstede’s consequences: A study of reviews, citations and replications”, Organization Studies, Vol. 15, pp 447– 56

Titman. S. and Wessels, R., 1988, “The determinants of capital structure choice”, The Journal of Finance, 43. pp 1-19.

Tosi, H. L. and Greckhamer, T., 2004, “Culture and CEO compensation”, Organization Science, Vol. 15, No. 6, pp 657–671.

Verbeek, M., 2004, A Guide to Modern Econometrics Second Edition (Revised and Updated). Chichester etc., Wiley. 2004. pp. 340-60.

36

Appendices

Appendix 1

Selected banks and their home country

Bank Country Bank Country

ABN AMRO Netherlands Friesland Bank Netherlands

Allied Irish banks Ireland HBOS United Kingdom

Alpha Bank Greece HSBC United Kingdom

Anglo Irish bank Ireland Hypo real estate Bank Germany

Attica Greece ING Groep Netherlands

Banco comercial Portuguese Portugal Julius Bar Switserland Banco Monte Pachni Sienna Italy KBC bank NV Belgium Banco Popular Espanol Spain Komercni banka Czech Republic Banco Portugues de Negocios Portugal Lloyds Banking Group United Kingdom

Banco Santander Spain Natixis France

Banesto Spain NBG Greece

Bank Austria Creditanstalt Austria Nordea Bank Sweden

Bank Pekao Poland OTP bank Hungary

Barclays United Kingdom Piraeus Bank Greece

BBVA Spain PKO bank Poland

BNP Paribas France RBG Austria

Caisse d'epargne France Royal bank of Scotland United Kingdom

Commerzbank Germany Saxo Bank Denmark

Credit agricole France Skandinaviska Enskilda Banken Sweden Credit Suisse Group Switserland SNS Bank Netherlands

Danske Bank Denmark Societe Generale France

Deutsche Bank Germany Standard Chartered United Kingdom

Dexia Belgium Svenska Handelsbanken Sweden

Erste Bank Austria Swedbank Sweden