Banks’ ownership structure, risk, and performance
during the credit crisis
Johan Kasper
University of Groningen
Faculty of Economics and Business
MSc Business Administration
TABLE OF CHAPTERS
1 INTRODUCTION ... 3
2 LITERATURE REVIEW ... 5
2.1 Separation of ownership and control, agency theory, and corporate governance ... 5
2.2 The ownership structure of the company ... 6
2.3 Ownership structure, risk and performance ... 7
2.3.1 Effects of the degree of ownership concentration on risk taking and performance ... 8
2.3.2 Effects of the nature of the owners on risk taking and performance ... 11
2.4 Hypotheses ... 14
3 METHODOLOGY AND DATA ... 16
3.1 Methodology ... 16
3.2 Data ... ...16
4 RESULTS ... 18
5 DISCUSSION AND CONCLUSION ... 21
5.1 Discussion ... 21
5.2 Conclusion ... 23
1 INTRODUCTION
The start of the 21st century was marked by one of the world’s biggest economic downturns in modern history. During the first few months of 2007, the financial markets realized that the losses related to the subprime mortgages were far more severe than anticipated in the previous period (Ryan 2008). The subprime mortgage crisis and the credit crunch that followed in 2007 and 2008, sent shockwaves across the globe. A lot of financial institutions, including banks, incurred huge losses. Some of them were bailed out by their governments, others went bankrupt. This crisis is another example of banks’ preference for risk and the resulting unstable nature of banking (Barry et al. 2011). Academics, politicians and policy-makers are looking for answers in order to prevent such a crisis from happening in the future. However, this is not an easy task. A lot of different factors contributed to the cataclysm in 2007 (Beltratti and Stulz 2012).
One of the major factors contributing to the severity of the credit crunch was excessive risk taking by companies in the financial industry on a large scale (Barry et al. 2011). This was in turn caused by a failure of corporate governance mechanisms (Moxey and Berendt 2008; Kirkpatrick 2009; Beltratti and Stulz 2012). In banks with a poor corporate governance structure excessive risk taking was more common than in banks with a strong corporate governance structure. This made those banks riskier, which in turn resulted in larger losses during the crisis (Beltratti and Stulz 2012).
The corporate governance structure of a company consists of a set of mechanisms, which together should make a coherent system of good governance. According to Laeven and Levine (2009), the ownership structure of a company is a standard corporate governance mechanism. Beltratti and Stulz (2012) use the ownership structure of a bank as a proxy for corporate governance structure as a whole, relying on theory “that greater ownership by insiders aligns […] [their interest with] the interest of shareholders” and that those banks were better governed and as a result performed better during the crisis. Beltratti and Stulz (2012) find little evidence to support this claim. Beltratti and Stulz (2012) are to my knowledge, the first and only researchers looking into the effects of ownership structure on risk taking and performance of banks during the credit crisis.
largest shareholder in a company holds. Another dimension of ownership structure is the nature of ownership, which differentiates between different types of owners, such as the government, mutuality or private ownership (Iannotta et al. 2007). Therefore, Beltratti and Stulz (2012) used an insufficient proxy for corporate governance and produce no significant results.1
The purpose of this research is to enhance the study of Beltratti and Stulz (2012): what are the effects of ownership concentration and nature of ownership on risk taking and performance by banks, during the credit crisis? Improving their research will enhance our knowledge of the causes of the financial crisis. This will be done by using two dimensions of ownership structure instead of one dimension and see if this produces different results. This research differs from previous research for several different reasons. First, focus on only one dimension of ownership structure, specifically ownership concentration, by Beltratti and Stulz (2012) has, in the past, produced an array of contradicting results (Iannotta et al. 2007; Barry et al. 2011). Therefore this dimension alone can’t explain the effects of ownership structure on risk taking and performance by banks during the credit crisis. Second, using two different dimensions of ownership structure is a better proxy for corporate governance because it encompasses a broader base of governance types common in the banking industry across the globe. Third, using two dimensions creates an opportunity to use a larger sample of banks. This will produce more realistic results, because not only publicly traded banks, which were they only subject of research in Beltratti and Stulz (2012) study, were affected by the financial crisis.
The next section will review the relevant theoretical and empirical literature on ownership structure of the company. These fundamentals will be used to formulate hypotheses. In section 3 the research methodology and the data will be described. Section 4 presents the results of the empirical tests performed. In the last section the results will be discussed. This last section also contains some concluding remarks and possibilities for further research.
1
2 LITERATURE REVIEW
This section will describe the relevant literature, both theoretical and empirical. The review will start with a general description of underlying elements of ownership structure, namely separation of ownership and control, agency theory, and corporate governance. Next the different dimensions of ownership structure will be described. This description will be followed by theoretical and empirical research on the effects of the different ownership dimensions on risk taking and performance of banks. In the last part of this section, the views discussed in this section will be combined in hypotheses. These hypotheses will be tested in section 4.
2.1 Separation of ownership and control, agency theory, and corporate governance
Most companies in the world are owned and managed by the same individual. This form of doing business is called the sole proprietorship. However, when business grows more complex, owners tend to choose a more complex form to conduct their business. In more complex forms of companies a separation of ownership and control usually occurs. The owner of the company is no longer also its manager. This separation of ownership and control creates a conflict of interest between the owning and controlling (managing) party. This conflict of interest is called the agency problem (Eisenhardt 1989).
In short, the agency problem occurs because the interest in the company of the principal (owner) differs from the interests in the company of the agent (manager).2 The major factor contributing to this agency problem is information asymmetry between the principal and the agent. The agent has direct access to all information in the company and the principal does not. This creates an opportunity for the agent to act in his own interest, ignoring the interest of the principal. The principal can try to minimize this agency problem by monitoring the agent or by using incentives to align the managers’ interest with his own interest (Jensen and Meckling 1976; Eisenhardt 1989; Shleifer and Vishny 1997). For example, the principal can reward the agent with shares, thereby trying to force the agent to think and act like a shareholder, and thus align his interests in the company with those of the principal. This principal-agent relationship and control of agency costs is one of the principles of corporate governance:
2 The agency problem not necessarily occurs in owner-manager settings, hence the definition
“Corporate governance involves a set of relationships between a company’s
management, its board, its shareholders [...] Good corporate governance should provide proper incentives for the board and management to pursue objectives that are in the interests of the company and its shareholders and should facilitate effective monitoring” (OECD 2004).
There are many different mechanisms of corporate governance, such as mechanisms like board composition and voting rules (Berger et al. 2005). One of the standard corporate governance mechanism also frequently used in theoretical and empirical literature is the ownership structure of a company (Laeven and Levine 2009), which influences other corporate governance mechanisms, such as board composition.
It is important to mention another mechanism controlling the agent besides the corporate governance structure of the company: signals from the efficient capital market. Capital markets generally make rational assessment of the value of a company, even if only imperfect information is available (Fama 1980). Thus a manager deviating from ‘market efficient’ behavior will be disciplined by the capital market, through market value of the company (Fama 1980; Altunbas et al. 2001). However, this mechanism is only available to companiess whose shares are traded publicly.
2.2 The ownership structure of the company
The ownership structure of a company can be defined in many different ways. For example, one can define the ownership structure as the ratio of debt to equity of a company.3 In this thesis however, the ownership structure of the company is a description of the actual composition of owners of a company. A company can be owned by a single person or a single entity, such as the government. However, a company can also have many owners and/or many different types of owners. These owners can again be individuals and/or entities. All these different owners have different interests and therefore also different interest in the companies’ policy.
All these different owners can also have a different stake in the company. Because of the stake difference, different owners have different voting power and can therefore influence companies’ policy in a certain way. An individual can own a few of the companies’ shares. This individual has an insignificant stake and therefore virtually no voting power. Say, another individual owns 10% of the shares. This individual does have a significant stake and voting power. He will be able to have an impact on the direction of development of the company according to his own interests, because he can influence the composition of the board and thereby the selection of managers.
If we formalize the above into a rigid ownership structure model we find that the ownership structure of a company can be defined along two main dimensions. These two dimensions are the degree of ownership concentration and the nature of the owners (Iannotta et al. 2007). Companies can differ in ownership structure because they have a different degree of ownership concentration. Ownership concentration looks into how dispersed the ownership base is. In other words, if there is a shareholder holding a majority stake (high concentration) or if the shareholdings are widely dispersed (low concentration). Companies can also differ in ownership structure because the nature of the owners differs. A company can be held by the government. A company can also be held by held in private by, for example, a family or by the public, because its shares are traded on a stock exchange. Other forms can also occur and a combination of forms is often found as well. The nature of owners can also affect the degree of ownership. In publicly held companies ownership is more likely to be dispersed as opposed to privately held companies (Barry et al. 2011). Above mentioned variations of the ownership structure of the company have an impact on the corporate governance of the company.
Because this thesis is about the effects of ownership structure on the performance of banks during the credit crisis, hereafter banks will be the vocal point of this research. The theory mentioned above also applies to banks.
2.3 Ownership structure, risk and performance
(OECD 2004). According to Jensen and Meckling (1976), corporate risk taking is influenced by the ownership structure of the company. The risk-taking incentives of banks are driven by shareholders’ behavior, who usually have incentives to take higher risks (Barry et al. 2011). In this thesis we will therefore focus on the effects of ownership structure on risk taking by banks and its effects on performance.
The ownership structure of the firm is also related to the performance of a bank. If the costs of monitoring managers are high or the agency problems are large, the bank will incur increasing cost, thereby reducing performance (Laeven 1999).
2.3.1 Effects of the degree of ownership concentration on risk taking and performance
Insider versus outsider ownership in theory
The degree of ownership concentration has an effect on risk-taking by banks. The diversified owner4 and the manager of a bank have a different attitude towards risk given their principal-agent relationship. The diversified owner can diversify his risk by holding shares in more than a single bank, in other words by holding a diversified portfolio of shares. The manager cannot diversify his risk, because his personal wealth is tied to his position within the bank and personal benefits stemming from this position (Jensen and Meckling, 1976; Demsetz and Lehn, 1985). Therefore, the manager may act at a certain risk level, which is lower than the risk level desired by the owners. By increasing the managers’ holding in the bank, this effect is mitigated and the manager will act more in line with the diversified owner (Saunders et al. 1990).
4
Insider versus outsider ownership tested empirically
Most of the early empirical literature focuses on this difference in ownership structure; the degree of insider ownership5 versus shareholder ownership by outsiders. This research usually produces significant results on the effects of ownership structure on risk taking, but has varying signs. Saunders et al. (1990) find a positive relationship between insider ownership and risk taking by banks. According to their hypotheses, managers increase risk in an attempt to maximize the value of their stock options. Houston and James (1995) compare compensation between banking and nonbanking companies. They find that the compensation in banking does not induce excessive risk taking. Gorton and Rosen (1995) find an inverse U-shaped relationship between insider ownership and risk-taking. They find that insider ownership induces excessive risk taking when the banking industry is performing poorly. Chen et al. (1998) find that as insider ownership increases, bank risk decreases. Anderson and Fraser (2000) find varying signs depending on the tine period of the research. During periods of deregulation, insider ownership is positively related to banks’ risk taking behavior and during periods of reregulation negatively related to banks’ risk taking behavior. They also find that systematic risk is not related to insider ownership. Sullivan and Spong (2007) find a positive relationship between insider ownership and bank risk taking. The differences between the empirical results above can partially be explained by the different time periods in which the research had taken place Bank regulation and the market environment continually change. These changes have an impact on these results as well (Chen et al. 1998; Anderson and Fraser 2000; Laeven and Levine 2009).
The theory of the powerful controlling shareholder
Contrary to the general believes above greater ownership by insiders does not necessarily align their incentives more closely with the interests of shareholders. Another way to control the managers' risk taking behavior is by concentrating shareholdings. Concentrated shareholders have enough voting power to control on the managers. The manager is controlled, because a concentrated shareholder has influence on the managers’ job security, remuneration, etc. Concentrated ownership is thus a useful tool to align the interests of the principal and the agent (Shleifer and Vishny (1997).Greater ownership by the controlling shareholder means he can use his control to his own benefit (Beltratti and Stulz 2012).
Theory states that owners prefer more risk taking by a bank than debt holders and non-shareholder managers (Laeven and Levine 2009), because increasing bank risk after collecting funds from debt holders6 benefits diversified owners. This last effect can be explained by the fact that the payoff to owners of leveraged equity resembles a call option on the banks’ assets. To generate higher profits, the owners want to increase the assets’ risk and thereby increase expected returns (Galai and Masulis 1976; Esty 1998). Increasing bank risk does not benefit managers, because, as explained above, they cannot diversify their risk (Jensen and Meckling 1976; Demsetz and Lehn 1985). Thus, banks empowering diversified owners take more risk than banks where the ownership structure induces a less active governance role for the owners (Laeven and Levine 2009). Shareholders with a larger stake in a bank have also got greater power and incentives to alter corporate behavior than diversified owners (Shleifer and Vishny 1986). These incentives are that a shareholder with a larger stake will be hurt more by managers not acting in shareholders’ interest, because their stake is larger. Therefore, shareholders with a larger stake can influence the risk-taking by banks, and thus ownership concentration influences banks’ risk-taking.
The powerful controlling shareholder tested empirically
Some empirical evidence supports this claim. Laeven and Levine (2009) find that banks with more powerful owners are riskier, using insolvency risk as a proxy. Haw et al. (2010), who also use insolvency risk, also find that concentrated ownership increases risk. Louzis et al. (2012) find that non-performing loans increase as ownership concentration increases. In line with Laeven and Levine (2009), Beltratti and Stulz (2012) find that concentrated ownership increases risk. However, Shehzad et al. (2010) find that high levels of ownership concentration (20% and 50%) significantly reduces risk, proxied by non-performing loans in banks.
Ownership concentration and performance
According to theory, concentrated ownership gives significant more power to owners and therefore increases their ability to monitor management (Sheifer and Vishny 1986). This in turn increases the performance of concentrated companies (Aghion and Tirole 1997).
Azofra and Santamaría (2011) find a positive effect between high levels of ownership concentration and profitability and efficiency in Spanish commercial banks.
Concluding remarks
In general, theory suggests owners desire a higher level of risk taking in banks than managers. Managers' risk taking behavior will be more in line with owners desired risk level if a manager has an equity stake in the bank. However, there is no clear empirical evidence for the exact effects on insider ownership. The signs vary significantly. Another view is the powerful controlling shareholder theory, whereby a concentrated owner can exert his power to 'force' the manager into different risk taking behavior than the manager would desire otherwise. There is empirical evidence supporting this claim. Higher levels of ownership concentration lead to high levels of risk taking
2.3.2 Effects of the nature of the owners on risk taking and performance
Banks’ nature of ownership can be defined in many different ways. For this research we distinguish between three main types of ownership within the banking industry. The three types are the government owned banks, the mutual banks and the privately owned banks7. These three types of banks are common in the banking industry. In Europe, the larger versions of these different types of banks have more or less developed into similar full-service banks, competing in the same market (Iannotta et al. 2007). However, different types of owners have different effects on the agency problem and risk-taking behavior of banks (Barry et al. 2011) and ultimately performance. This is also the case with the three main types of banks distinguished above.
2.3.2.1 Government owned banks
Risk and performance in government owned banks versus private banks in theory
The agency problem also exists in government owned banks. In government owned banks, managers may use the banks’ resources for personal benefit (Barry et al. 2011). In government owned banks the government is the sole owner of the bank and therefore the principal in the principal-agent relationship. However, when compared with privately
7
owned banks, there are less controls available to government owned banks which could align the agent with the interest of the principle. First of all, the government cannot reward his agent with shares or share options in the bank. Second, there is no efficient capital market discipline to control the agent in government owned banks (Altunbas et al. 2001; William 2004). Therefore, government owned banks are considered less efficient (Laeven 1999; Altunbas et al. 2001) and therefore less profitable (Iannotta et al. 2007) than their privately owned counterparts.
Another problem for government owned banks is that politicians, the indirect owners of such banks, have political interest and incentives, that are not in the banks best interest (Shleifer 1998). This also could hamper performance. However, government banks may be less efficient than their private counterparts on purpose. Some government banks are used to finance certain industries, sectors or regions, etc., and not focussed on profit maximization, which increases the risk of their assets (Berger et al. 2005).
Empirical results on risk and performance in government owned banks vs. private banks
The differences in risk and performance between government owned banks and privately owned banks has also been researched empirically. On risk and performance, Berger et al. (2005) find that government owned banks have riskier assets, especially in nonperforming loans, and are less profitable than their privately owned counterparts. Iannotta et al. (2007) also find that government owned banks are less profitable and more risky than privately owned banks. Government owned banks have poorer loan quality and higher insolvency risk than their private counterparts. They also seem to be less profitable than privately owned banks, which is surprisingly not due to superior cost efficiency, but due to higher net returns on assets of private banks.
2.3.2.2 Mutual banks
Risk and performance in mutual owned banks versus private banks in theory
mutual banks which could align the interest of the managers with the interest of the owners. First, because there are no such things as tradable shares in a mutual bank, the owners cannot reward the managers with incentives like share options to align the managers interest with their own interest. Second, because of the lack of tradable shares, there is also a lack of efficient capital market discipline (Altunbas et al. 2001). Third, the owners of a mutual bank usually have no power to change the management of the bank. This is due to the fact that mutual banks always have dispersed ownership (O’ Hara 1981) and the contract between owners and the bank often doesn’t grant this power to owners in the first place (Fama and Jensen 1983a; Rasmusen 1988). Because concentrated ownership in mutual banks is virtually impossible, these banks also lack the efficiency of concentrated ownership (Iannotta et al. 2007).
However, a certain mechanism of control, not found in either government owned banks or privately owned banks, is available to mutual banks. The owners of a mutual can redeem their claim in the mutual at a value predetermined in their ownership contract with the mutual and this seems like a powerful control tool (Fama and Jensen 1983a). Inefficient behavior by managers will lead to withdrawal of funds by the owners (Altunbas et al. 2001). This control has never been used on a large scale.
Because of the lack of proper controls the management of a mutual bank can decide the company policy virtually by themselves, or as Rasmusen (1988) states it: “The mutual manager does not control the firm simply because ownership is diffuse [...] what is more important is that no individual can concentrate ownership of a mutual by purchasing the diffused shares. The manager is freed not by the absence of concentration, but by the absence of the threat of concentration”.
Because the manager in a mutual bank is less controlled than the manager in a privately owned bank, the mutual bank will be run more in line with the interest of the mutual bank manager. To maintain his position, i.e. to maintain his salary and other benefits, the mutual bank manager will take less risk than his privately owned counterpart (Fraser and Zardkoohi 1996; Rasmusen 1988). Therefore, privately owned banks will hold more risky assets compared to mutual banks (Fama and Jensen 1983b).
individuals as the owners, so the owners have no incentives to increase risk of the residual claim, because it is their own investment they are risking (Valnek 1999).
Empirical results on risk and performance in mutual owned banks versus private banks
Several researchers have done empirical research into the effects of mutual bank ownership on risk taking behavior and performance. O’Hara (1981) finds that mutual banks in the savings and loan industry are less cost efficient and less profitable than their private counterparts. Fraser and Zardkoohi (1996) as well as Esty (1997) find that privately owned banks in the savings and loan industry are more risky than their mutual counterparts. Also, mutual bank converting to privately owned banks in this industry increase their risk (Esty 1997). Valnek (1999) finds that mutual building societies in some cases outperform private retail banks. He argues that the mutual specific advantages, namely the merger of owner and depositor, give these banks a low risk profile, and thereby attracting a large group of risk averse depositors. Iannotta et al. (2007) find that private banks seem to be more profitable than mutual banks.
2.4 Hypotheses
This study tries to enhance the research conducted by Beltratti and Stulz (2012) and therefore looks into the effects of ownership structure on risk and performance in the banking industry during the credit crisis, using two dimensions of ownership structure as opposed to one dimension used by Beltratti and Stulz (2012). Combining the theory above leads to the following hypotheses.
Hypothesis 1a
Risk taking behavior in private banks at the start of the credit crisis is higher where there is a powerful owner proxied by an ownership concentrations of 10% and above.
Hypothesis 1b
In short, as pointed out by Laeven and Levine (2009), powerful diversified owners have incentives to increase risk taking in banks. Since excessive risk taking was one of the major causes of the credit crisis, an increase in risk will mean a decrease in performance.
Hypothesis 2
Government owned banks are riskier than their privately owned counterparts and therefore had worse performance during the credit crisis.
In short, government owned banks usually serve specially financing needs in certain industries, regions, etc. (Berger et al. 2005). As a consequence their assets are usually more risky and their nonperforming loan ratio is usually rather high when compared with privately owned banks. Therefore government owned banks are considered more risky. As mentioned above, risky banks are expected to perform worse during the credit crisis.
Hypothesis 3
Mutual banks are less risky than their privately owned counterparts and therefore had better performance during the credit crisis.
3 METHODOLOGY AND DATA
This section will give a description of the methodology and data used to empirically test the hypotheses formulated in the previous section.
3.1 Methodology
To examine the influence of ownership structure on risk and performance during the credit crisis this research will conduct independent samples t-tests to compare the different groups of banks. Both ownership concentration and nature of ownership are tested separately, because there is only 100% ownership concentration in both government owned banks and mutual banks. The independent samples t-test will tell us if the different groups of banks are comparable.
Bank risk
A common measurement for risk used in corporate finance is the beta of a company. However, the database used in this thesis did not contain any information on the beta's of banks. Therefore risk had to be proxied by other data available in the database. Like Iannotta et al. (2007) I use the ratio of loan loss reservation to gross loans as a proxy for risk. This ratio is a standard measurement of risk (Barry et al. 2011). The higher this ratio is, the higher the amount of low quality assets a bank possesses. The poorer the asset quality the more riskier the bank becomes.
Bank performance
Simple income statement information is not sufficient to describe the performance of banks, because a lot of factors can easily influence this form of data. To measure performance I used the return on average assets (RoAA). This ratio is commonly used as an indicator to assess the profitability of the assets and thereby the performance of banks (Heffernan and Fu 2010).
3.2 Data
order to be able to enhance the study of Beltratti and Stulz (2012) it is necessary to try and construct a similar sample of banks. Like Beltratti and Stulz (2012) the focus is on the largest banks in the world. Like their sample, only banks with assets in excess of $10 billion at the end of 2006, the year before the start of the credit crisis, were selected. However, Beltratti and Stulz (2012) sample only consists of publicly traded banks. This thesis also considers privately owned banks, which are not publicly traded, and also mutual banks and governement owned banks. Therefore, the sample differs in this respect. These criteria gives us a sample of 1.282 banks. Banks from 18 countries8 across the globe were selected, removing 735 banks from the sample.
For the research on the effects of ownership concentration we apply the following extra criteria to arrive at a sample of 370 banks. First, banks need to have shares which can be owned, thus excluding mutual and government owned banks, removing 85 banks from the initial sample. Second, banks that are a owned by a parent company already in the sample are removed from the sample. Then the sample is split in two, one group of banks has an ownership concentration of >10% and the other group has an ownership concentration of <10%. Several really extreme outliers were deleted, because the information seemed incorrect. These banks had a ratio of loan loss reservation to gross loans of several thousands. Also some banks had incomplete data, removing 18 banks from the sample. For the research on the nature of ownership we apply different criteria and arrive at a sample of 304 banks from the initial sample of 1.282 banks. This sample is split into three groups. The first group includes all banks where the ultimate owner is not a mutual owned banks nor where the government has any stake. The second group includes only banks where the government is the ultimate owner of the majority of the shares. The third group includes only mutual owned banks. Again, several really extreme outliers were deleted, because the information seemed incorrect. These banks had a ratio of loan loss reservation to gross loans of several thousands and again some banks had incomplete data.
8 The countries used in this study are: Australia, Austria, Belgium, Canada, Denmark, Finland, France,
4 RESULTS
This section contains the results of the independent sample t-test performed. First, the results on ownership concentration will be describe, followed by the results on nature of ownership. First, descriptive statistics will be described, followed by the test statistics. The output tables can be viewed in the appendix.
Ownership concentration
In general the average ratio of loan loss reservation to gross loans is larger when banks have an ownership concentration of 10% and above as opposed to an ownership concentration of 0% to 10%. Comparing these averages, without paying attention to the independent sample t-test would suggest that banks with an ownership concentration of 10% and above were more risky before the start of the crisis.
In general the average return on average assets was much lower in 2008 compared to 2007 for both groups of ownership concentration. The average return on average assets was higher for banks with an ownership concentration between 0% to 10% in 2007 and lower in 2008 for the same banks. Comparing these averages, without paying attention to the independent sample t-test would suggest that banks with an ownership concentration between 0% to 10% were more profitable during crisis year 2007 and less profitable in crisis year 2008.
For the ratio of loan loss reservation to gross loans the equal variances assumed hypothesis is rejected. Therefore the Welch t-test results have to be used to analyze if there is an effect of different ownership concentration levels on risk taking behavior in banks before the credit crisis. This test however, produces no significant results, leading to the conclusion that there is no relationship between different ownership concentration levels and risk taking behavior before the credit crisis.
Nature of ownership, comparing privately owned banks with government owned banks
In general the average ratio of loan loss reservation to gross loans is larger for privately owned banks compared to government owned banks. Comparing these averages, without paying attention to the independent sample t-test would suggest that privately owned banks were more risky before the start of the crisis.
In general the average return on average assets was much lower in 2008 compared to 2007 for both government owned banks and privately owned banks. The average return on average assets was only slightly higher for privately owned banks in 2007 and quite a bit higher in 2008. Government owned banks have a negative average return on average assets in 2008. Comparing these averages, without paying attention to the independent sample t-test would suggest that privately owned banks were more profitable during the crisis. For the ratio of loan loss reservation to gross loans the equal variances assumed hypothesis is not rejected. Therefore the pooled t-test results have to be used to analyze if there is an effect of nature of ownership, comparing government owned banks and privately owned banks, on risk taking behavior in banks before the credit crisis. This test however, produces no significant results, leading to the conclusion that there is no relationship between the nature of ownership, comparing government owned banks and privately owned banks, and risk taking behavior before the credit crisis.
For the return on average assets the equal variance assumed hypothesis is not rejected either. Therefore the pooled t-test results have to be analyzed if there is an effect of effect of nature of ownership, comparing government owned banks and privately owned banks, on profitability of banks after the credit crisis. However, these test also produce no significant results, leading to the conclusion that there is no relationship between the nature of ownership, comparing government owned banks and privately owned banks, and profitability during the credit crisis.
Nature of ownership, comparing privately owned banks with mutual banks
In general the average return on average assets was much lower in 2008 compared to 2007 for both mutual banks and privately owned banks. The average return on average assets was also higher for mutual banks compared to privately owned banks in both 2007 and 2008. Comparing these averages, without paying attention to the independent sample t-test would suggest that mutual banks were more profitable during the crisis compared to their privately owned banks counterparts.
For the ratio of loan loss reservation to gross loans the equal variances assumed hypothesis is not rejected. Therefore the pooled t-test results have to be used to analyze if there is an effect of nature of ownership, comparing mutual banks and privately owned banks, on risk taking behavior in banks before the credit crisis. This test however, produces no significant results, leading to the conclusion that there is no relationship between the nature of ownership, comparing mutual banks and privately owned banks, and risk taking behavior before the credit crisis.
5 DISCUSSION AND CONCLUSION
This section will discuss the results presented and described in the previous section, in the light of the theoretical framework developed in section two. The section will end with some concluding remarks and suggestion for further research.
5.1 Discussion
Ownership concentration
According to the powerful controlling shareholder theory a concentrated owner can exert his power to 'force' the manager into different risk taking behavior than the manager would desire otherwise. This theory has been tested empirically before and leads to the conclusion that higher levels of ownership concentration lead to high levels of risk taking. In line with this theory I hypothesized that risk taking behavior in private banks at the start of the credit crisis would higher if there was a powerful owner, proxied by an ownership concentration of 10% and above. This hypothesis was tested empirically.
The results contradict the prediction made based on theory. According to the results there is no relationship between different ownership concentration levels and risk taking behavior before the credit crisis. So the risk taking behavior is not dependent on levels of ownership concentration.
These results are hard to explain. A lot of theory and empirical evidence suggest that there is a relationship between ownership concentration and risk. However, Iannotta et al. (2007) also find contradicting results, which they fail to explain. They however do find a relationship, but contrary to the theory. They suggest further empirical research should focus to explain these contradicting results.
The results contradict the hypothesis. The results suggest that there is no relationship between different ownership concentration levels (and thus assumed risk levels) and profitability during the credit crisis.
So powerful owners are not necessarily better in monitoring management. Maybe more powerful owners interfere with management against all odds and thereby create a negative impact on performance. It might also be the case that the unusual nature of the performance and the worldwide economic downturn could explain the fact that no relationship between ownership concentration and performance was found. Since this relationship has not been tested empirically often further research might shine some more light on this relationship.
Nature of ownership, comparing privately owned banks with government owned banks
Theory states that government owned banks are more risky than their privately owned counterparts, because they finance industries, sectors and regions, which cannot obtain finance from private banks (Berger et al. 2005). Empirical research confirms the theory that government owned banks are more risky than privately owned banks. As a result of this higher risk, government owned banks are expected to perform worse during the credit crisis. In line with this theory the following hypothesis was formulated and tested empirically: government owned banks are riskier than their privately owned counterparts and therefore had worse performance during the credit crisis.
Nature of ownership, comparing privately owned banks with mutual banks
According to theory mutual banks are less risky than the privately owned counterparts because the separation of ownership and control is much larger in mutual banks when compared to privately owned banks (Fraser and Zardkoohi 1996). There are also less controls available in mutual banks when compared to privately owned banks (O' Hara 1981; Fama and Jensen 1983a; Rasmusen 1988; Altunbas et al. 2001; Iannotta et al. 2007), which gives management, who are generally risk averse according to agency theory, control of the bank. Mutual banks are therefore less risky than their privately owned banks. In general, empirical research find, in line with theoretical research, that mutual banks are less risky than privately owned banks. Banks with less risk before the start of the credit crisis are expected have superior performance during the credit crisis. The theory led to the following empirically tested hypothesis: mutual banks are less risky than their privately owned counterparts and therefore had better performance during the credit crisis.
The results suggest that the hypothesis should be rejected, leading to the conclusion that there is no relationship between the nature of ownership, comparing mutual banks and privately owned banks, and risk taking behavior before the credit crisis and profitability during the credit crisis. These results are not in line with theory and earlier empirical research. The fact that there is no relationship could be explained by the argument made by Iannotta et al. (2007), that mutual banks, like government owned banks, and private banks have converged into the same full-service banking models. Therefore these banks have the similar asset portfolio's and similar risk levels.
In all the results suggest, against prevailing theoretical en empirical research, that there is no relationship between ownership structure, risk taking and performance before and during the credit crisis. The only explanation for this fact can be found in the very unusual nature of the financial environment right during the credit crisis.
5.2 Conclusion
dimensions of ownership structure instead of one enhances our understanding the impact of different ownership structures on risk in banks during the financial crisis.
In order to assess the effects of ownership structure on risk and performance in banks, several hypotheses were formulated. The first states that banks with a large ownership concentration would be more risky than banks that had a dispersed ownership concentration. Two other hypotheses researched the effect of nature of ownership on risk. The nature of ownership was covered by three banking forms: the government owned bank, the mutual bank and the privately owned bank. In general, it was also hypothesized that banks that were more risky before the credit crisis, would perform worse during the crisis. These hypothesis were tested and using independent sample t-tests. All results indicate that there is no relationship between ownership structure, risk and performance. This is contrary to the general believes in theoretical and empirical research in this field. The only explanation for this fact can be found in the very unusual nature of the financial environment right during the credit crisis
Further research
6 REFERENCES
Aghion P., Tirole J., 1997. Formal and Real Authority in Organizations, Journal of
Political Economy, 105, 1, 1-29.
Altunbas Y., Evans L., Molyneux P., 2001. Ownership and Efficiency, Journal of Money,
Credit and Banking, 33, 4, 926-954.
Anderson R.C., Fraser D.R., 2000. Corporate control, bank risk taking, and the health of the banking industry, Journal of Banking & Finance, 24, 1383-1398.
Azofra V., Santamaría M,. 2011. Ownership, control, and pyramids in Spanish commercial banks, Journal of Banking & Finance, 35, 1464-1476.
Barry T.A., Lepetit L., Tarazi A., 2011. Ownership structure and risk in publicly held and privately owned banks, Journal of Banking & Finance, 35, 1327-1340.
Berger A.N., Clarke G.R.G., Cull R., Klapper L., Udell G.F., 2005. Corporate governance and bank performance: A joint analysis of the static, selection, and dynamic effects of domestic, foreign, and state ownership, Journal of Banking & Finance, 29, 2005, 2179-2221.
Beltratti A., Stulz R.M., 2012. The credit crisis around the globe: why did some banks perform better?, Journal of Financial Economics, 105, 1, 1-17.
Chen C.R., Steiner T.L., Whyte A.M., 1998. Risk-taking behavior and management ownership in depository institutions, The Journal of Financial Research, 21, 1, 1-16.
Demsetz H., Lehn K., 1985. The Structure of Corporate Ownership: Causes and Consequences, Journal of Political Economy, 93, 6, 1155-1177.
Eisenhardt K.M., 1989. Agency Theory: An Assessment and Review, Academy of
Management Review, 14, 1, 57-74.
Esty B.C., 1997. Organizational form and risk taking in the savings and loan industry,
Journal of Financial Economics, 44, 25-55.
Esty B.C., 1998. The impact of contigent liability on commercial bank risk taking, Journal
of Financial Economics, 47, 189-218.
Fama E.F., 1980. Agency Problems and the Theory of the Firm, The Journal of Political
Economy, 88, 2, 288-307.
Fama E.F., Jensen M.C., 1983a. Separation of Ownership and Control, Journal of Law and
Fama E.F., Jensen M.C., 1983b. Agency Problems and Residual Claims, Journal of Law
and Economics, 26, 2, 327-349.
Fraser D.R., Zardkoohi A., 1996. Ownership Structure, Deregulation, and Risk in the Savings and Loan Industry, Journal of Business Research, 37, 63-69.
Forssbæck J., 2011. Ownership structure, market discipline, and banks’ risk-taking incentives under deposit insurance, Journal of Banking & Finance, 35, 2666-2678. Galai D., Masulis R.W., 1976. The option pricing model and the risk factor of stock,
Journal of Financial Economics, 3, 53-81.
Gorton G., Rosen R., 1995. Corporate Control, Portfolio Choice, and the Decline of Banks,
Journal of Finance, 50, 5, 1377-1420.
Haw I.-M., Ho S.M.S., Hu B., Wu D., 2010. Concentrated control, institutions, and banking sector: An international study, Journal of Banking & Finance, 34, 485-497.
Heffernan, S.A., Fu, X., 2010. Determinants of financial performance in Chinese banking,
Applied Financial Economics, 20, 1585-1600.
Houston J.F., James C., 1995. CEO compensation and bank risk. Is compensation in banking structured to promote risk taking?, Journal of Monetary Economics, 36, 405-431.
Iannotta G., Nocera G., Sironi G., 2007. Ownership structure, risk and performance in the European banking industry, Journal of Banking & Finance, 31, 2127-2149.
Jensen M.C., Meckling W.H., 1976. Theory of the firm: managerial behavior, agency costs and ownership structure, Journal of Financial Economics, 3, 305-360.
Kirkpatrick G., 2009. The Corporate Governance Lessons from the Financial Crisis, OECD
Journal: Financial Market Trends, 96, 1, 1-30.
Laeven L., 1999. Risk and Efficiency in East Asian Banks, World Bank Discussion Paper, 2255.
Laeven L., Levine R., 2009. Bank governance, regulation and risk taking, Journal of
Financial Economics, 93, 259-275.
Merton R., 1977. An analytical derivation of the cost of deposit insurance and loan guarantees: an application of modern option pricing theory. Journal of Banking &
Finance, 1, 3-11.
Moxey P., Berendt A., 2008. Corporate Governance and the Credit Crunch, ACCA
Discussion Paper.
O’Hara M., 1981. Property Rights and the Financial Firm, Journal of Law and Economic, 24, 2, 317-332.
OECD, 2004. OECD Principles of Corporate Governance, Paris.
Rasmusen E., 1988. Stock Banks and Mutual Banks, Journal of Law and Economics, 31, 395-422.
Ryan S.G., 2008. Accounting in and for the Subprime Crisis, The Accounting Review, 83, 1605–1638.
Saunders A., Strock E., Travlos N.G., 1990. Ownership Structure, Deregulation, and Bank Risk Taking, The Journal of Finance, 45, 2, 643-654.
Shehzad C.T., de Haan J., Scholtens B., 2010. The impact of bank ownership concentration on impaired loans and capital adequacy, 34, 399-408.
Shleifer A., 1998. State versus Private Ownership, The Journal of Economic Perspectives, 12, 4, 133-150.
Shleifer A., Vishny R.M., 1986. Large Shareholders and Corporate Control, The Journal of
Political Economy, 94, 3, 1, 461-488.
Shleifer A., Vishny R.M., 1997. A survey of corporate governance, Journal of Finance, 52, 2, 737-783.
Sullivan R.J., Spong K.R., 2007. Manager wealth concentration, ownership structure, and risk in commercial banks, Journal of Financial Intermediation, 16, 229-248.
Valnek T., 1999. The comparative performance of mutual building societies and stock retail banks, Journal of Banking & Finance, 23, 925-938.
William J., 2004. Determining Management Behavior in European Banking, Journal of
