Bank capital and the stock returns of the European banks

(1)

University of Amsterdam, Faculty Economics and Business

BSc Finance and Organization, Finance track

Bachelor Thesis

Bank Capital and the Stock Returns of the

European banks

Casper Kool, 10350055

June 2015

(2)

(3)

Abstract

Within this bachelor thesis the relationship between bank capital and stock returns of European banks is researched, during the Subprime mortgage crisis and the Sovereign debt crisis. For estimating this relation I do a panel data regression with time fixed effects and country dummy's. Besides taking a sample, including all European banks, I also research three other samples consisting out of banks in Northern, Southern or Eastern Europe. The results of this research show that during the pre crisis period there was a negative relation between bank capital and the stock returns in Europe. Only in Eastern Europe during the Subprime mortgage crisis and in Northern Europe during the Sovereign debt crisis there was a positive relation between capital and stock returns. During the Sovereign debt crisis only the stock returns of the Northern European banks were positively related to tier 1 capital ratio. The tier 1 capital ratio did not influence the stock returns of the Southern European countries during the Sovereign Debt Crisis.

(4)

(5)

1. Introduction

The start of the Subprime mortgage crisis was triggered by an unexpected decline in asset value in the US and the following Sovereign debt crisis in Europe. This development has made the discussion about bank capital more relevant and intensified than ever before. The Basel I and Basel II accords could not prevent these crises. Therefore, the Basel Supervision Committee developed the Basel III accords after the Subprime mortgage crisis. The discussion about the right amount of capital banks should hold and the way this influences the financial system is ongoing. In addition, the visions about this subject are not in line with each other.

A general idea is that a higher level of bank capital makes banks more stable and resistant to economic shocks (Basel Committee, 2011). Besides that, some theories state that as banks have more skin in the game they also will make more prudent investment decisions (Kim and Santomero, 1988).

On the other hand it is said that a higher regulatory level of capital increases the costs of capital for banks (Repullo, 2004). Banks will have to raise their interest rates and this will slow down economic growth (Holstrom and Tirole, 1997). Some theories also state that banks with a higher amount of bank capital take on more risk (Calomiris and Kahn, 1991).

This thesis builds on the papers of Berger and Bouwman (2013) and Demirguc-Kunt, Detragiache and Merrouche (2013) about the role that bank capital plays on bank performance. According to Berger and Bouwman (2013), banks with a higher level of bank capital have a higher probability of survival during crises times and are associated with a higher market share. Demirguc-Kunt, Detragiache and Merrouche (2013) state that banks with a higher level of bank capital also had higher stock returns during the US financial crisis.

These papers took a cross-country sample of banks in the entire world and only investigated the effect of bank capital during the financial crisis that started in the US. Within this thesis, the focus is on Europe and is investigated what effect capital had on the stock returns of European banks during the Subprime mortgage crisis and the Sovereign debt crisis that followed after. Besides the European sample, I also research the differences within Europe by dividing the sample into Northern, Southern and Eastern European banks.

One of the findings was that during the pre crisis period a negative relation existed between bank capital and the stock returns in Europe. There was a positive relation between capital and stock returns in Eastern Europe during the Subprime mortgage crisis. During the Sovereign debt crisis only the stock returns of the Northern European banks were positively related to tier 1 capital ratio.

(8)

(9)

2. Literature Review

2.1 Basel Accords

The Basel Accords have been from significant importance for the role bank capital played during the past decade. Therefore, a short description of the development of these accords is presented below.

Basel I

The first Basel accord was published in 1988 and came into existence by the end of 1992. This accord had two objectives. The first was to create an international standard capital framework that would reduce the competitive inequalities among internationally operating banks. The second was to require banks to hold enough capital to absorb losses without causing systematic problems (Basel Committee, 1988).

The assets on the balance sheet were assigned to five different categories, which all had a different risk weight. The off balance sheet assets were converted into credit risk equivalents. The total capital of the banks was divided into tier 1 capital, which consist of common stock and disclosed reserves. Tier 2 capital consists out of undisclosed reserves, revaluation reserves, hybrid debt capital instruments, subordinated debt and general provisions. The banks had to hold at least 8% of the risk-weighted assets (RWA) and least half of it had to be tier 1 capital (Basel Committee, 1988).

Although the Basel accord was easy to apply and considered a good first step, there were also critics. Jones (2000) argues that it was easy for banks to manipulate the levels of capital or risky assets because of unclear accounting rules. He also mentioned that it was possible for banks to lower their regulatory level of risky assets without really lowering their level of risk. Financial innovation and securitization helped banks to trick the regulatory measures.

Basel II

To make the capital requirements more risk sensitive the Basel committee introduced the Basel II accords in 2004. This accord introduced the three-pillar system, which is still being used. The first pillar contains the minimum capital requirements. Besides credit risk, operational risk and market risk was also captured by the way the required level of capital was calculated. The banks could choose different ways of determining the risk of their assets. They could determine their exposure to risk internally but also by using external credit rating. The second pillar contains the supervisory regulatory review. This should give the regulators better tools to control the financial institutions. The third pillar contains disclosure rules for the banks. This should improve the market discipline (Basel Committee, 2009).

As mentioned before, the main objective of the Basel II accord was to improve the risk sensitivity of the capital framework. The committee believed that the framework was more dynamic and better able to keep up with time and financial innovation.

One of the critiques on the Basel II accord was that it was too pro-cyclical. If the assets of banks were downgraded in economic hard times - and banks did not have enough capital to absorb this loss - the banks had to raise capital during times of recession to stay above the Basel II required minimum capital. Since the cost of capital was high during these times, the obligation to raise capital even further reduced lending activities and made the recession heavier (Kashyap and Stein, 2004).

Another critique was that the use of external credit ratings was dubious because of the conflict of interests. On the one hand the banks paid these credit-rating firms to determine the risk exposure of the asset, on the other hand the credit rating firms had to give a reliable risk determination (Hakenes and Schnabel, 2011). The conflict of interest is easy to determine in this case.

(10)

Basel III

As a reaction to the crisis, the Basel Committee developed the plans for the Basel III accord in 2009, which is again an improvement of the earlier accord. It was firstly introduced in 2014. The main objective is to improve the ability of the banking sector to absorb unexpected economic shocks by stricter capital requirements (Bank for International Settlements, 2011). The Basel III accord introduced higher capital requirements with a countercyclical buffer, a leverage ratio that set limits to the size of a banks’ balance sheet and liquidity requirements that should improve the liquidity over banks.

The accord increases the capital requirements to 6 percent tier 1 capital of which 4.5 percent has to be common tier 1 capital. This was 2.5 percent under Basel II. They also introduced an obligatory capital buffer of 2 percent and that can be heightened with another 2.5 percent during times of economic growth. The ratios mentioned here are only the ratios of the first year. To provide the banks with some time to implement the new requirements, the starting requirements are lower and will increase over time. The 2.5 percent of extra mandatory capital that will be build up during economic good times is called the countercyclical buffer. In economic good times the banks will hold more capital than actually needed so the banks can use this buffers in economic bad times (Bank for International Settlements, 2011).

The Basel III accord also introduced the leverage ratio and liquidity requirements. The leverage ratio measures the amount of tier 1 capital divided by the total consolidated assets and should stay above 3 percent. The liquidity requirements mean that the banks should have enough liquid assets to cover a total net cash outflows over thirty days (Bank for International Settlements, 2011).

Also the net stable funding ratio was introduced by the Basel III accord and was meant to require the available amount of stable funding to exceed the required amount of stable funding over a one-year period of extended stress (Bank for International Settlements, 2011).

Since Basel III is introduced recently, the effects cannot be criticized yet. However, there already some concerns. The reaction of a lot of banks, which is also shared by for example Allen, Chan, Milne and Thomas (2012), is that the cost of capital will rise because of the heightened capital requirements and that this will harm the economy.

2.2 Development of the crises

For a good understanding of the effectiveness of the Basel accords and for the correct interpretation of the results, a summary of the Banking Industry in Europe is necessary. This will help to understand the reasons why the Basel accords are adjusted several times and why the accords could not prevent the economic crises. Finally, it helps to understand the results of this research as well.

Development of the banking industry in Europe

The European banking industry expanded rapidly and the banks became more interconnected between 1985 and 2004. Regulatory acts like the First banking directive, the Directive of capital flows and the Single European act made this possible. These acts deregulated the financial industry and stimulated international banking. In this period the banking industry became more consolidated - the number of banks fell with 40 percent - and the total amount of assets of the five biggest European banks increased with 340 percent (Goddard, Molyneux, Wilson and Tavakoli, 2007).

The deregulation increased the competition in the banking industry and the banks were looking for possibilities to increase their profitability. Internationalization was a

(11)

diversification possibility and the innovation of financial products started. Non-traditional banking returns increased from 28.3 percent in 1992 to 50 percent in 2000 (Goddard, Molyneux, Wilson and Tavakoli, 2007).

The internationalization of the banking industry and the innovation of financial products are key developments regarding the crisis in 2008 in the US. This transformed into the European Sovereign Debt Crisis in 2010.

Subprime mortgage crisis

The Subprime mortgage crisis was triggered by the unexpected decrease in the value of mortgage-backed securities and the big losses banks had to book. Mortgage-backed securities are financial products that bundled mortgages, sliced them into pieces and that were sold with different probabilities of default. These mortgage-backed securities were considered very safe and with a relatively high return. Therefore, they were a part of the asset portfolios of nearly the entire financial industry (Gorton and Metrick, 2012).

The crisis cannot be explained fully by the decline in asset value. In the years before the crisis the shadow banking industry had been developed. Shadow banks are financial entities that are designed to serve as intermediaries to channel savings into investments. This kind of entities was unregulated and had become very great in the years before the crisis. A product that had become very important for the financial market was the "Repurchase agreement", better known as repo. A repo transaction works as follows: A deposit is made at a bank for a very short term - mostly overnight - and the bank pays the repo rate on the deposit. For safety reasons, the bank provides collateral to the depositor. The depositors are very large financial institutions, for example money market funds. Because of the unexpected decline in the value of mortgage backed securities the short-term debt market in the US became disturbed (Bernanke, 2010).

When a lot of banks had to book big losses, the level of capital hold by the banks pointed out to be insufficient. These problems didn't stop in the US but also affected Europe via the short-term debt market. The problems of the Lehman brothers amplified the problems on the short-term debt market. After this large mutual fund had gone bankrupt, the trust in the financial markets was absolutely gone. Banks reduced their lending activity and started to increase their reserves. Firstly, the short-term debt market stood still and this worked through to the real economy. The demand for money was higher as the supply, the interest rates raised and the investments slowed down. The consequence was an economic recession. To rescue the financial system, a lot of banks had to be saved from bankruptcy by the US government and the European governments (Gorton and Metrick, 2012).

European Sovereign debt crisis

The short-term debt market internationally connects the banking industry. The drying up of the short-term debt market and the losses due to the necessary devaluation of assets caused that the European banks also came into financial trouble. The European governments had to support the banking industry with a lot of money to keep the European Banks from bankruptcy. This was an act of significant importance because otherwise the whole financial system would break down (Blundell-Wignall, 2012). Remarkable is that historically the number of nationalized or financially supported banks is much higher in Europe than in the US (Benink and Benston, 2005).

The balances of the European governments deteriorated because of the support provided to the banking industry. Together with the increased interest rates, the stress on the financial markets and the recession this caused the European Sovereign Debt Crisis. In Appendix 2.1 a table is presented of the development of government debt in European countries as a percentage of gross domestic product (GDP) (Blundell-Wignall, 2012).

(12)

and to lesser extent Spain, France and Italy - were seen very problematic in the international financial markets. This caused that these countries’ bond price decreased and the interest rates increased. That made it even harder for these countries to attract money and to fulfil their obligations (Blundell-Wignall, 2012). In Appendix 2.2 a chart is presented with the development of the interest rates on government bonds of those countries.

A lot of commercial banks also hold these government bonds in their asset portfolios. The decline in bond value also meant a decline in the amount of bank capital. An important fact to notice is that in 2011 the most exposure of commercial banks on the critical sovereign debt was from their home country. In case one of those countries would default, this would cause the most problems for commercial banks of that particular country (Blundell-Wignall, 2012).

In Appendix 2.3 a table is presented with the exposure of European countries to the sovereign debt of Greece, Portugal Ireland, Spain, France and Italy. Worth mentioning is the high level of government debt for the Southern European countries.

2.3 Theories about capital

The stability of the financial sector

The goal of implementing international rules that oblige banks to hold a higher level of capital is to create a more stable financial sector. Primarily, it should stabilize the financial sector by making the banks less sensitive to economic shocks (Basel Accords, 2011). During an economic downturn, the assets of banks decrease in worth and the level of capital gets under the by the Basel accords required level. Banks have to recover their capital and can do that in several ways, but in short there are two options:

Firstly, the banks can reduce their risk-weighted assets and secondly, by issuing new capital. Reducing risk-weighted assets also means reducing the amount of lending activities. If banks hold a higher level of capital the possibility that they have to reduce their risk-weighted assets is smaller. It will be less necessary to reduce lending and this is good for the economy. If it is harder for businesses to lend this will work through to the economy and have a big impact (Holstrom and Tirole, 1997).

Koehn and Santemero (1980) state that banks also prefer to higher their level of capital above changing their asset portfolios. If the asset portfolios are changed directly after the capital requirements have changed, this will make a bank directly less profitable. These assets are the sources of income for the banks.

Another positive effect of a higher level of bank capital is that banks become more resistant against economic shocks and the probability of default of the banks decreases (Repullo, 2004). Logically, banks with a higher level of bank capital can face more losses before they are having liquidity problems and eventually get into default.

Empirical evidence by Demirguc-Kunt, Detragiache and Merrouche (2013) showed that higher capitalized banks had higher stock returns during the last financial crisis. This result holds especially for large systematic banks.

Also Berger and Bouwman (2013) found empirical evidence that the size of a bank changes the role bank capital plays for banks. Bank capital is more important for small banks as it is for medium or large banks. Small banks are hit harder by financial crises and have fewer opportunities to attract new capital during those financial crises (Allen, Berger & Bouwman, 2009). For small banks a higher capital base is associated with higher profits, a higher market share and with a higher probability of survival during all times. For medium and large banks this only holds during crises times.

(13)

Risk taking

The effect of capital requirements on risk taking behavior is often researched and the outcomes are divergent and bilateral. Theories about risk taking behavior of banks - as a reaction on regulatory capital - contain important information for explaining the way the stock market investors judge the amount of capital.

A higher level of bank capital changes the risk taking incentives of banks. Banks have more “skin in the game”. This means that banks have more of their own money at risk. The expectation is that banks with more skin in the game will make more prudent investment decisions and will better monitor their investments (Kim and Santomero, 1988). Banks with a higher level of capital are expected to take on less risk and will monitor and screen their investments better than banks with less capital (Koehn and

Santomero, 1980; Mehran and Thakor, 2011; Allen, Carletti and Marquez, 2011). On the other hand, bankers say that higher capital requirements lead to higher

cost of capital and will lower their profits. If the capital requirements increase, a bank has to hold more capital for every (possibly) profit-making asset they have. This will raise the cost of capital and reduce the profits a bank can make. In order to increase their profits banks should raise their interest on loans. If the interest rate increases this will reduce the amount of investments. This will hurt the economy (Repullo, 2004).

Calomiris and Kahn (1991) state the opposite. They say that banks with a lower capital buffer will monitor better and are more reliable for depositors. If the bank has less capital the necessity of monitoring becomes bigger and will be done more accurate by the banks. A bank with a lower capital reserve is more sensitive for losses caused by bad monitoring. The banks need it more that their lenders will pay their debts than for a well-capitalized bank. Therefore, a low capital buffer is seen as a sign to the depositors that the bank has to monitor very well to stay above the minimum required level of capital. Depositors will see a high level of capital as a sign of possible risk taking by the banks and will probably take out their deposits. Which can cause a bank failure if it continues (Shrieves and Dahl, 1992).

In addition, Berger, Petersen, Rajan and Stein (2005) state that small banks better monitor their clients and have closer relationships with their lenders comparing to large banks. Because smaller banks have a smaller amount of assets, they have to pick and monitor their assets with more care than larger banks. This affects the amount of risk a bank is taking and will also influence the amount of capital small banks decide to hold. In general, smaller banks will hold a higher percentage of bank capital.

Another theory is that the risk taking profile of banks is U-shaped. This means that banks will take on more risk when they are just above the minimum required bank capital levels and when they are well above the minimum required bank capital level. The level of risk taking for banks that are just above the minimum will increase because they need to make more profits to get better capitalized. Undercapitalization can be seen as a weak sign to the depositors which will probably take of their deposits (Keeley, 1990; Calem and Rob, 1999).

Well-capitalized banks can take on more risk because the extra risk does not increase the probability of default by a lot. The bank has an extra capital buffer which reduces the impact of potential losses. There can be taken more risk by the bank without substantially increasing the risk of bank failure (Keeley, 1990; Calem and Rob, 1999).

Besides the risk taking behavior, bank capital also changes the price banks charge for a certain product. A higher regulatory level of capital also changes the price a bank asks for their products. Different assets carry different amounts of risk and the amount of capital that a bank is required to hold for each asset is different. Therefore, the cost of capital is different for every asset as well. This could even cause that products that are too risky are not sold anymore, as the prices are getting too high (Kim and Santomero, 1988).

Not only the level of bank capital influences the risk taking behavior of banks. Financial supervisors also play a role in the way banks react on capital regulation. Regarding this matter, Blum (2008) says that it is very important that supervisors have strict ex post monitoring tools to control banks. The reason for this is that supervisors

(14)

cannot check the risk taking of banks ex ante and it is costly for banks to hold capital. This creates an incentive for banks to pretend as if they hold a less risky portfolio than they actually do. This is only controllable by supervisors afterwards. This is why the power and strictness of a supervisor also influences the risk taking behavior of banks.

The presence or absence of deposit insurance also influences banks' risk taking behavior. Deposit insurance is important to create rest among financial markets and protects banks from bank-runs in moments of financial stress. Besides that, it gives banks the possibility to raise funds close to the risk free rate (Diamond and Dybvig, 1983). If there were no deposit insurances, there would be more risk to depositors. They might feel that their money is not safe at a particular bank, which causes depositors changing banks quicker. The other side of deposit insurance is that it creates an incentive for more risk taking and less screening. This is because a large amount of the money that is invested by banks is insured by the government - (Ruckes, 2004; Xu, 2010). It needs attention that besides insured deposits also a significant amount of money is not secured. So these theories will not cover the whole amount of the money banks have available to invest.

(15)

3. Data and Methodology

3.1 Sample selection

The selection of the sample of banks starts by the number of banks available in the Bankscope database in the Wharton Research Dataservices (WRDS). The condition is that there is no information on the tier 1 ratio capital ratio available.

Firstly, I delete all the non-European from this sample and secondly all the countries that weren’t in the European Union (EU) by 2005, excluding Switzerland. This decision is made as most of the available data is from the countries that were part of the EU in 2005. Besides this practical reason, the relation between capital and stock returns in Europe was also a subject for research. An economic relation has to grow over years and - due to the single European market - this relation is stronger between countries that are in the EU for a longer time. That is why I delete all the European countries that weren’t a part of the EU in 2005. The reason Switzerland is included is because it has a very large banking industry and is geographically centred in the middle of Europe. Switzerland has some differences with the countries that are in the EU but is commonly seen as a part of the modern western economy.

Secondly, all the listed banks are selected of which there is enough - this means at least one normal time period observation - stock return information available. Afterwards, all the stock returns data is collected from DataStream for the banks left in the sample. In the end, there are 108 banks out of 22 countries in the sample with information over the years 2005-2013. Not every bank is in the sample every year, so the sample is unbalanced. At last, these banks are deleted of which was less than one normal time observation available.

In this paper a separation is made between Northern, Southern and Eastern Europe. Southern Europe is defined for this thesis as the countries around the Mediterranean Sea. Eastern Europe exists of the countries that were under the influence of the Soviet-Union. Finally, Northern Europe regards the rest of Europe. In Appendix 1 a complete list of all the countries and a summary of the observations per sample group are presented.

3.2 Empirical Model

Hypotheses

In this thesis is mainly researched how bank capital is related with the stock returns of European banks during the normal time period, the Subprime mortgage crisis and the European sovereign debt crisis. The sub questions are how bank capital is related with the stock returns of Northern, Southern or Eastern Europe banks during the normal time period, the Subprime mortgage crisis and the European sovereign debt crisis.

During the normal time period is expected that the stock returns have a negative relation with the stock returns in every sample group. The explanation for this is that the amount of capital that is hold above the required minimum means that a bank is not maximizing profit because there could be invested more capital. This makes these banks less profitable as banks that hold bank capital near the minimum required level. Moreover, during normal times the expectation that banks default is lower as during crises times, therefore a lower level of bank capital is seen as sufficient.

(16)

positively related with stock returns in every sample group. Banks that hold a higher level of capital can absorb more losses and the probability that a bank will survive a crisis is higher. Therefore, better-capitalized banks are seen as better investments and the stock returns will be higher than for less capitalized banks.

For Northern and Eastern Europe the same relation is expected as during the Subprime mortgage crisis during the Sovereign debt crisis. No relation at all is expected in Southern Europe during the Sovereign debt crisis. The exposure of the banks in those countries to the risk on government bonds of their homeland is higher as the required tier 1 ratio. Because the exposure on the sovereign debt is significant, the expectation is that there is no relation between the tier 1 ratio and stock return.

The empirical hypotheses of the main question and the sub questions are:

H0: β1 tier1ratioit-1 * Normal times = 0

H1: β1 tier1ratioit-1 * Normal times ≠ 0

H0: β1 tier1ratioit-1 * Subprime mortgage crisis = 0

H1: β1 Tier1ratioit-1 * Subprime mortgage crisis ≠ 0

H0: β1 tier1ratioit-1 * Sovereign debt crisis= 0

H1: β1 Tier1ratioit-1 * Sovereign debt crisis ≠

The Model

The following model is estimated:

Stock returnsit = β0+ β1 tier1ratioit-1 * normal times + γ1 control variablesnormaltimes *

normal times+ β2 tier1ratioit-1 * Subprime mortgage crisis + γ2 Control variablesnormaltimes*

Subprime mortgage crisis + β3 Tier1ratioit-1 * Sovereign debt crisis+ γ3 control

variablesnormaltimes * Sovereign debt crisis + α2D2i ...αnDni + λ2B2t ... + λnBnt + uit

Where the stock returns are yearly calculated by taking the Log of (Ri t/Ri t-1). Ri is the

Total return index calculated by DataStream. The formula DataStream uses to calculate the Total return index is:

R t = Ri t-1 * ( Pi t / Pi t-1 ) *( 1+ ( Dy t /100 ) * ( 1/N) )

Where Pi is the price index, Dy the dividend yield of the price index and n is the number of days in that financial year.

The main interest regards the coefficients of the tier 1 ratios in the different time periods since that is the measure of capital I use. The reason for choosing the tier 1 ratio as the measure of capital is that in the paper of Demirguc-Kunt, Detragiache and Merrouche (2013) the Basel ratio was significant in their full sample test. For a robustness check the same regression is performed with the total capital ratio. In this model the explanatory variables interact with the time dummies so they can change in the interested time periods. To measure only the effect of capital I keep the control variables constant during the crises times. If there are two years of normal time information the average values of

(17)

the two years are taken as the constant value for the crises years. If there is only one normal time value, this is used as value for the crises years. The last variable in the model is the disturbance term.

The normal times dummies are connected with the financial statement information of 2005 and 2006 and the stock return information of 2006 and 2007. This period is called the “normal time” because it is the period before the Subprime mortgage crisis started and since crises are unexpected, is seen as a period that could be called normal. The second period is connected with the financial statement information of 2008 and 2009 and the stock returns of 2009 and 2010. This period is called the Subprime mortgage crisis. This time period is researched independently because this crisis started in the US and there was aimed to test if it changes the way bank capital is judged by the investors in Europe.

The beginning point of the next period is the year that Greece came in to financial problems. In this paper that is the starting point of the European Sovereign debt crisis. This period hasn’t come to an end yet and therefore this last period contains financial statement information of 2010, 2011, 2012 and 2013 and is linked with the stock returns of 2011, 2012, 2013 and 2014.

In the model time fixed effect and country dummies are used to control for any omitted variables on country level. This includes economic shocks, differences in accounting principles and the way a country’s government acts a reaction on the crisis. At bank level control variables are included that control for bank specific characteristics. In order to keep the model compact, they were not mentioned above but will be reviewed in this part of the paper. The control variables in the model are:

- Liquidity ratio. This is measured as the liquid assets/total assets as provided in Bankscope.

- Deposit ratio. This is measured as the deposits/total assets as provided in Bankscope.

- Asset quality ratio. This measured as the loan loss provision/total assets as provided in Bankscope.

- The net amount of loans ratio. This measured as the net amount of loans/total assets in Bankscope.

- Bank size. This is taken into account in the model by taking the log of total assets. - Return on average equity (ROAE). This is a measure provided in Bankscope.

3.3 Summary of the statistics

A short summary of the statistics is provided in this part of the paper. All the information can be found in Appendix 3.

Table 3.1 shows that in every sample group the stock returns are positive in the normal time period and become negative in the Subprime mortgage crisis. The stock returns of Eastern Europe are the highest in every year. In the Sovereign debt crisis the stock returns in Northern and Eastern Europe become positive again and only in Southern Europe the stock returns stay negative. The stock returns in Southern Europe are much more volatile in the normal time period compared to Northern and Eastern Europe. In the Subprime mortgage crisis the stock returns in every sample group become

(18)

much more volatile. Remarkable is that the standard deviation in Northern Europe is at a much higher level - as before the Subprime mortgage crisis - in the Sovereign debt crisis and the standard deviation in Eastern Europe even becomes smaller as before the Subprime mortgage crisis.

Table 3.2 and 3.3 show the summary statistics of the tier 1 capital ratio and the total capital ratio. These two ratios follow the same patterns. The capital ratios become smaller during the Subprime mortgage crisis and increase again during the Sovereign debt crisis. The average tier 1 capital ratios are always above the 10%, that is well above the minimum required capital level. Also interesting are the differences in standard deviation between Northern, Southern and Eastern Europe. The standard deviation in Eastern Europe is structurally lower as in Northern and Southern Europe. Also remarkable is the fact that the standard deviation in Northern Europe almost halved in the Subprime mortgage crisis.

Interesting statistics of the control variables are that the liquidity of banks in every sample group decreased over time, the deposit ratio is much higher in Eastern Europe, the loan loss provisions in every sample group increased spectacular in every time period, the average amount of total assets are much higher in Northern Europe and that the ROAE is much higher in Eastern Europe in every period as in the other sample groups.

(19)

4. Results

In this chapter of the thesis the results of the regression are presented and explained. The main interest regards the regression outcomes for the tier 1 ratio since this is the chosen measure of capital. For a robustness check the same regression was performed with the total capital ratio.

In the normal time period there was a negative relation between the tier 1 ratio and stock returns. Only for the whole Europe sample this relation was significant. Also the deposit ratio significantly affected the stock returns. A higher deposit ratio was associated with higher stock returns.

An explanation for the negative relation of the tier 1 ratio with the stock returns in Europe during the normal time period is that the necessity for a higher level of capital was not seen during that time period. The Basel II accord was just introduced and it was a period of economic growth. Banks with a higher level of tier 1 ratio have a higher cost of capital and do not invest as much as banks with a lower tier 1 ratio. Therefore, there is a negative relation between stock returns and the tier 1 capital (Repullo, 2004). This is also in line with my hypotheses for this time period.

That banks with a higher deposit ratio also had higher stock returns during the normal time period can be explained by the profit making function deposits can have. Banks with a higher deposit rate have more money that they can invest. The expected profitability of these banks is higher and the stock market returns as well.

During the Subprime mortgage crisis the relation between the tier 1 ratio and stock returns changed and became positive but this is only significant for Eastern Europe. There also appeared to be a significant negative relation between the normal times asset quality and the stock returns in Southern Europe. The outcomes of the regression are different than expected in the hypotheses, but can be explained as follows:

The problem of the Subprime mortgage crisis was that banks were highly connected with each other and that the financial markets came to a stop because of the problems in the short term debt market. Big banks had capital problems and there was a lot of financial distress on the US and European financial markets. When the trust in the financial system is very low and investors expect big losses the trust in the tier 1 ratio also decreases. The meaning of a high tier 1 ratio becomes less relevant. This explains why there isn't a relation between the tier 1 ratio and the stock returns for the whole Europe sample, the Northern Europe sample and the Southern Europe sample.

During the Subprime mortgage crisis period there is only a significant positive relation for the Eastern Europe sample, which can be explained by the fact that Eastern Europe countries recently joined the European Union. Therefore, these countries were less connected with the US and other European banks. The Eastern European banks were less exposed to the mortgage backed securities and less connected with the toxic banks. That is why the tier 1 ratio still was significant and positive related with stock returns in Eastern Europe during the Subprime mortgage crisis.

During the Sovereign debt crisis period the tier 1 ratio had a significant positive effect on the stock returns for the whole Europe sample. That the tier 1 capital ratio in the Europe and Northern Europe sample groups are positively significant while they are not in the Southern and Eastern can be explained by the fact that the whole Europe sample is dominated by the banks out Northern Europe. The outcomes are partly in line

(20)

with the hypotheses and partly not. The expectation was that there would not be a relation in Southern Europe and that there was a positive relation between capital and stock returns in Europe. However, this relation was expected in Eastern Europe as well.

Only the Northern European banks had a positive relation between the tier 1 ratio and stock returns during the Sovereign debt crisis. This can be explained by the fact that investors searched for the safest investment opportunities in Europe. The banks in Northern Europe are bigger in size and these banks are less sensitive to economic shocks and can attract new capital easier as smaller banks during an economic crisis (Berger, 2009). Besides that the Northern European banks are bigger, they are also part of the Western Economies for a longer time and therefore better known as the Eastern European banks. This can cause that the Northern European banks are considered a safe haven. This can also explain that - even while the average tier 1 ratio was not lower in Eastern Europe -investors better rewarded the Northern European banks. In addition, this can also be an explanation for the fact that in the normal times bank size is positive and significant related with stock returns for this period

The high level of exposure on the government bonds of their homeland countries can explain that there was no significant relation between the tier 1 ratio and the stock market returns in Southern Europe. These government bonds were decreasing in value and the level of exposure on this government bonds was higher than the required tier 1 capital ratio. This can explain why the tier 1 capital was not considered a good measure of survival probability and the relation bet. A high level of tier 1 capital was not judged as a good predictor of future profitability anymore and therefore, less rewarded on the stock markets.

Also some control variables show significant results for the Sovereign debt crisis period. The normal times deposit ratio is significantly related with higher stock returns for Northern Europe and a higher normal time asset quality - which is measured by the loan loss provision. This is also significantly and negatively related with stock returns for Europe.

An explanation for the positive relation between the normal time deposit ratio and the higher stock returns during the Sovereign debt crisis is that banks that rely more on deposits in their way of funding also are less active in non traditional banking activities. In times of recession, this can be considered more reliable and more profitable. It might explains the positive relation between the normal time deposit rate and the stock returns during the Sovereign debt crisis. The negative relation between the normal times asset quality and the stock returns during the Sovereign debt crisis can also be clarified. Banks that had higher loan loss provisions during the normal times are probably more conservative and therefore more stable during crisis time.

The regression results of the robustness check - with the total capital ratio instead of the tier 1 ratio - shows similar results as the regression with the tier 1 ratio. Please refer to Appendix 4 for the full results of the regression with the total capital ratio instead of the tier 1 ratio. However, there are some differences. There was no negative significant relation between the total capital ratio and stock returns for the normal time period. The coefficients for the Southern and Eastern Europe sample groups were not negative either. In addition, there is no significant positive relation between the normal times deposit ratio and the stock returns for the normal time period.

(21)

This table shows the results of the regression with the tier 1 ratio as measure of capital.

Regression results Europe Northern Europe Southern Europe Eastern Europe

Tier 1 ratio * normal times -0,0088* -0,0035 -0,0090 -0,0158

(0,0052) (0,0066) (0,0146) (0,0433)

Liquidity * normal times -0,0832 -0,4175 -0,2955 0,8650

(0,3012) (0,4056) (0,6783) (1,9542)

Deposits * normal times 0,4782** 0,5849** 0,4676 -1,1341

(0,2316) (0,2854) (0,5472) (2,1118)

Asset quality * normal times 9,25 -13,20 11,26 58,87

(10,48) (14,75) (22,93) (80,10)

Net amount of loans * normal times -0,1349 -0,3752 -0,2980 -0,6074

(0,2853) (0,3528) (0,7520) (2,4418)

Bank size * normal times -0,0220 -0,0180 -0,0100 -0,2031

(0,0193) (0,0250) (0,0443) (0,2698)

ROAE * normal times 0,0028 0,0049 0,0036 0,0173

(0,0046) (0,0070) (0,0090) (0,0172)

Tier 1 ratio * Subprime mortgage crisis 0,0066 0,0032 0,0028 0,0785*

(0,0068) (0,0116) (0,0116) (0,0365)

Liquidity * Subprime mortgage crisis 0,2572 0,1522 0,1519 2,4615

(0,3117) (0,4058) (0,6516) (1,8845)

Deposits * Subprime mortgage crisis 0,1458 0,2157 -0,1602 -0,1127

(0,2304) (0,2831) (0,5673) (2,0271)

Asset quality * Subprime mortgage crisis 0,74 -8,89 1,54** 99,14

(11,89) (16,70) (31,70) (119,02)

Net amount of loans * Subprime mortgage crisis 0,3193 0,3467 0,0305 -0,4723

(0,2945) (0,3763) (0,6847) (1,9266)

Bank size * Subprime mortgage crisis 0,0035 0,0084 0,0054 -0,2561

(0,0192) (0,0265) (0,0395) (0,3326)

ROAE * Subprime mortgage crisis -0,0022 -0,0056 0,0027 0,0052

(0,0048) (0,0065) (0,0114) (0,0218)

Tier 1 ratio * Sovereign debt crisis 0,0078** 0,0231*** 0,0047 -0,0088

(0,0032) (0,0065) (0,0052) (0,0329)

Liquidity * Sovereign debt crisis -0,1482 -0,4644* 0,1446 0,0462

(0,2083) (0,2707) (0,4065) (1,4207)

Deposits * Sovereign debt crisis 0,1925 0,5362*** -0,2920 -0,8715

(0,1615) (0,1878) (0,3847) (1,5718)

Asset quality * Sovereign debt crisis -33,51*** -7,90 -58,10 13,65

(8,79) (11,76) (23,52) (108,50)

Net amount of loans * Sovereign debt crisis 0,0688 0,0829 0,4163 -0,6224

(0,1977) (0,2415) (0,4479) (1,9836)

Bank Size * Sovereign debt crisis 0,0166 0,0476*** -0,0101 -0,1697

(0,0121) (0,0163) (0,0255) (0,3224)

ROAE * Sovereign debt crisis 0,0007 -0,0008 -0,0086 0,0261

(0,0033) (0,0042) (0,0081) (0,0172)

Fixed effects dummies Yes Yes Yes Yes

Number of observations 931 515 323 93

(22)

(23)

5. Conclusion

The conclusion of this thesis is that the level of tier 1 capital only has a marginal effect on the stock returns of the European banks. During the normal time period - the financial statement information of 2005 and 2006 and the stock return information for 2006 and 2007 - there was a negative relation between the tier 1 capital ratio and the stock returns of banks in Europe a year later. A possible explanation is that the necessity of bank capital was not considered as important as after the crisis. It was a period of economic growth and the Basel II accord was just implemented. Banks that hold less capital had lower cost of capital and invested more money; this was considered more profitable and rewarded on the stock market. Therefore, there was a negative relation between the tier 1 ratio and stock market returns.

During the Subprime mortgage crisis - the financial statement information of 2007 and 2008 and the stock return information for 2008 and 2009 - there was only a positive significant relation between the tier 1 capital ratio and the stock returns of banks in Eastern Europe. This could be explained by the fact that during the Subprime mortgage crisis the trust in the financial industry - and the Basel accords measures of risk - was small. This caused that the tier 1 capital ratio was not considered a good measure of risk and future profitability in Northern and Southern Europe. The fact that the Eastern European countries entered the European Union in 2004 could explain why there was a relation between the tier 1 ratio and stock market returns during the Subprime mortgage crisis. These banks had less exposure to the mortgage backed securities and were less connected to the international short term debt market. Therefore, the tier 1 capital ratio had a significant influence in Eastern Europe during the US financial crisis period.

The findings during the Sovereign debt crisis show that only the stock returns of the Northern European banks were positively related to tier 1 capital ratio. Investors searched for the safest investment opportunities in Europe. The banks in Northern Europe are bigger and they are also part of the Western Economies for a longer as the Eastern European banks. This caused that Northern European banks are considered a safe haven and explains that - even while the average tier 1 ratio wasn't lower in Eastern Europe -investors better rewarded the Northern European banks.

The fact that there is no relation between the tier 1 capital ratio and the stock returns of the Southern European countries can be explained by the fact that the banks in these countries had a big exposure to their own countries' sovereign debt. The exposure to this risk was higher than the required tier 1 capital ratio and therefore the meaning of this ratio decreased.

A possible follow-up on this thesis could be one which considers a longer time period that includes more banks in the Eastern Europe sample group. The available amount of information was not very extensive in the databases used. This could have influenced the outcomes of this research.

(24)

References

Allen, F., Carletti, E. & Marquez, R. (2011). Credit market competition and capital regulation. Review of Financial Studies, 24(4), 983-1018.

Allen, B., Chan, K., Milne, A. & Thomas, S. (2012). Basel III: Is the curse worse than the disease? International Review of Financial Analysis 25, 159-166.

Basel Committee on Banking Supervision. (1988). International Convergence of Capital Measurement and Capital Standards. Bank for International Settlements, July.

Basel Committee on Banking Supervision. (2009). History of the Basel Committee and its Membership. BCBS Report, August.

Basel Committee on Banking Supervision. (2011). A global regulatory framework for more resilient banks. From: http://www.bis.org/publ/bcbs189.htm.

Bennink, H. & Benston, G. (2005). The future of banking regulation in developed countries: lessons from and for Europe. Financial Markets, Institutions and Instruments 14 (5), 289-328.

Allen N., Berger, A. & Bouwman, C. (2009). Bank liquidity creation. Review of Financial

Studies 22(9), 3779-3837.

Berger, A. & Bouwman, C. (2013). How does capital affect bank performance during financial crises? Journal of Financial Economics, 109, 146-176.

Berger, A., Miller, N., Petersen, M., Rajan, R. & Stein, J. (2005). Does function follow organizational form? Evidence from the lending practices of large and small banks.

Journal of Financial Economics, 76, 237-269.

Bernanke, B. (2010). Causes of the Recent Financial and Economic Crisis. From: http://www.federalreserve.gov/newsevents/testimony/bernanke20100902a.htm.

Blum, J. (2008). Why Basel II may need a leverage ratio restriction. Journal of Banking &

Finance, 32, 1699–1707.

Blundell-Wignall, A. (2012). Solving the Financial and Sovereign Debt Crisis in Europe.

OECD Journal: Financial Market Trends.

Calem, P. & Rob, R. (1999). The impact of capital-based regulation on bank risk-taking.

Journal of Financial Intermediation 8, 317–352.

Calomiris, C. & Kahn, C. (1991). The Role of Demandable Debt in Structuring Optimal Banking Arrangements. The American Economic Review 81(3), 497-513.

Dahl, D. & Shrieves, R. (1992). The relationship between risk and capital in commercial banks, Journal of banking & Finance 16 (2), 439–457

(25)

Demirguc-Kunt, A., Detragiache, E. & Merrouche, O. (2013). Bank Capital: The lessons from the financial crisis. Journal of Money, Credit and Banking 45 (6).

Diamond, D. & Dybvig, P. (1983). Bank runs, deposit insurance and liquidity. Journal of

Political Economy 91 (3), 401-419

Eurostat (2015), General government gross debt - annual data, from: http://ec.europa.eu/eurostat/tgm/table.do?tab=table&init=1&language=en&pcode=teina 225&plugin=1.

Goddard, J., Molyneux, P., Wilson, J. & Tavakoli, M. (2007). European Banking: An overview. Journal of Banking and Finance 31, 1911-1935

Gorton, G. & Metrick, A. (2012). Getting up to speed on the Financial Crisis: A One Weekend-Readers Guide.

Jones, D. (2000), Emerging problems with the Basel Capital Accord: Regulatory capital arbitrage and related issues. Journal of banking & finance 24, 35-58.

Hakenes, H. & Schnabel, I. (2011). Bank size and risk-taking under Basel II. Journal of

Banking and Finance 35, 1436-1449.

Holstrom, B. & Tirole, J. (1997). Financial intermediation, loanable funds and the real sector. Quarterly Journal of Economics 112, 663-691.

Kashyap, A. & Stein, J. (2004), Cyclical implications of the Basel II capital standards’.

Federal Reserve Bank of Chicago Economic Perspectives 1, 18-31.

Kim, D. & Santomero, A. (1988). Risk in Banking and Capital Regulation. The Journal of

Finance 43, 1219–1233.

Keeley, M. (1990). Deposit insurance, risk and market power in banking. American

Economic Review 80, 1183-1200.

Koehn, M. & Santomero, A. (1980). Regulation of bank capital and portfolio risk. The

Journal of Finance 35 (5).

Mehran, H. & Thakor, A. (2011). Bank capital and value in the cross-section. The Review

of Financial Studies 24 (4).

Repullo, R. (2004). Capital requirements, market power and risk taking in banking.

Journal of financial intermediation 13, 156-182.

Ruckes, M. (2004). Bank competition and credit standards. Review of financial studies 17, 1073-1102.

(26)

Trading economies (2015), Switzerland/Greece Government Debt to GDP, from: http://www.tradingeconomics.com/switzerland/government-debt-to-gdp.

Xu, Y. (2010). Equity Capital, Market Power and Bank Liquidity Creation. Working Paper

(27)

Appendix 1: Specification of countries in each

sample group

Table 1.1 shows a specification of the countries in each sample group.

List of countries in each sample

Countries Northern Europe Southern Europe Eastern Europe

Austria X Belgium X Cyprus X Czech Republic X Denmark X France X Germany X Greece X Hungary Ireland X Italy X Lithuania X Luxembourg X Malta X Netherlands X Poland X Portugal X Slovakia X Spain X Sweden X Switzerland X United Kingdom X

(28)

Appendix 2: Sovereign debt and government

bond's interest rates

Table 2.1 shows the government debt of the countries in this research as a percentage of GDP. The Northern European countries are highlighted in blue, the Southern European countries in red and the Eastern European countries in grey.

Government Debt as percentage of GDP

Country 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Austria 68.3 67.0 64.8 68.5 79.7 82.4 82.1 81.5 80.9 84.5 Belgium 94.7 90.7 86.8 92.2 99.2 99.5 102.0 103.8 104.4 106.5 Czech Republic 28.0 27.9 27.8 28.7 34.1 38.2 39.9 44.6 45.0 42.6 Cyprus 63.4 59.3 54.1 45.3 54.1 56.5 66.0 79.5 102.2 107.5 Denmark 37.4 31.5 27.3 33.4 40.4 42.9 46.4 45.6 45.0 45.2 France 67.2 64.4 64.4 68.1 79.0 81.7 85.2 89.6 92.3 95.0 Germany 67.1 66.5 63.7 65.1 72.6 80.5 77.9 79.3 77.1 74.7 Greece 100 106.1 105.4 112.9 129.7 146 171.3 156.9 175.0 177.1 Hungary 60.8 65.0 65.9 71.9 78.2 80.9 81.0 78.5 77.3 76.9 Ireland 26.2 23.8 24.0 42.6 62.3 87.4 111.2 121.7 123.2 109.7 Italy 101.9 102.5 99.7 102.3 112.5 115.3 116.4 123.1 128.5 132.1 Luxembourg 6.3 7.0 7.2 14.4 15.5 19.6 19.1 21.9 24.0 23.6 Malta 70.1 64.6 62.4 62.7 67.8 67.6 69.7 67.4 69.2 68.0 Netherlands 49.4 44.9 42.7 54.8 56.5 59.0 61.3 66.5 68.6 68.8 Poland 46.7 47.1 44.2 46.6 49.8 53.6 54.8 54.4 55.7 50.1 Portugal 67.4 69.2 68.4 71.7 83.6 96.2 111.1 125.8 129.7 130.2 Slovakia 33.8 30.7 29.8 28.2 36.0 40.9 43.4 52.1 54.6 53.6 Spain 42.3 38.9 35.5 39.4 52.7 60.1 69.2 84.4 92.1 97.7 Sweden 48.2 43.1 38.2 36.8 40.3 36.8 36.2 36.6 38.7 43.9 Switzerland 48 42.9 39.4 37.2 35.6 34.3 33.6 34.6 34.9 34.2 United Kingdom 41.6 42.5 43.6 51.8 65.8 76.4 81.8 85.8 87.3 89.4 (Eurostat, 2015; Tradingeconomies, 2015)

(29)

Table 2.2 shows the development of the interest rates on government bonds of France, Italy, Spain, Ireland, Portugal and Greece.

(Blundel-Wignall, 2012).

Table 2.3a shows the exposure of banks on the sovereign debt of Greece, Ireland, Portugal, Spain, Italy and France.

(30)

Table 2.3b shows the exposure of banks on the sovereign debt of Greece, Ireland, Portugal, Spain, Italy and France.

(31)

Appendix 3: Summary statistics tables

Table 3.1 shows the summary statistics of the stock returns.

Europe Num.

Observations

Mean Std.

Deviation

Min Max

Stock returns normal times 203 1.14 0.34 0.21 3.38

Stock returns Subprime mortgage crisis 214 0.88 0.55 0.06 3.28

Stock returns Sovereign debt crisis 514 0.98 0.43 0.09 4.15

Northern Europe

Southern Europe

Eastern Europe

Stock returns normal times 21.00 1.26 0.30 0.72 1.97

Stock returns Subprime mortgage crisis 22.00 0.95 0.51 0.25 1.82

Stock returns Sovereign debt crisis 50.00 1.11 0.26 0.58 1.65

Table 3.2 shows the summary statistics of the tier 1 ratios.

Europe Num.

Observations

Mean Std.

Deviation

Min Max

Tier 1 ratio normal times 203 11.03 7.27 5.40 84.19

Tier 1 ratio Subprime mortgage crisis 214 10.41 5.42 3.40 45.89

Tier 1 ratio Sovereign debt crisis 514 13.65 6.58 3.40 67.64

Northern Europe

Southern Europe

Eastern Europe

(32)

Table 3.3 shows the summary statistics of the total capital ratios. Europe Num. Observati ons Mean Std. Deviation Min Max

Total capital ratio normal times 203 13.42 6.78 8.10 84.56

Total capital ratio Subprime mortgage crisis

214 12.93 5.68 4.50 64.25

Total capital ratio Sovereign debt crisis 514 16.22 8.36 4.80 95.15

Northern Europe

114 13.03 3.72 8.60 32.10

Southern Europe

76 12.97 8.28 4.50 64.25

Eastern Europe

22 12.21 1.96 9.32 16.09

Table 3.4 shows the summary statistics of the liquid assets ratios.

Europe Num.

Observations

Mean Std.

Deviation

Min Max

Liquidity Normal Time 203 0.25 0.16 0.04 0.87

Liquidity US Financial Crisis Time 214 0.21 0.14 0.02 0.84

Liquidity Greece Financial Crisis Time 514 0.19 0.14 0.02 0.79

Northern Europe

Southern Europe

Eastern Europe

(33)

Table 3.5 shows the summary statistics of the deposit ratios.

Europe Num.

Observations Mean Std. Deviation Min Max

Deposits normal time 203 0.67 0.16 0.22 0.93

Deposits Subprime mortgage crisis 214 0.67 0.17 0.21 0.94

Deposits Sovereign debt crisis 514 0.68 0.15 0.27 0.99

Northern Europe

Southern Europe

Eastern Europe

Table 3.6 shows the summary statistics of the asset quality ratios.

Europe Num.

Asset quality normal time 203 0.0020 0.0032 -0.0091 0.0207

Asset quality Subprime mortgage crisis 214 0.0031 0.0036 -0.0043 0.0224

Asset quality Sovereign debt crisis 514 0.0086 0.0117 -0.0009 0.1482

Northern Europe

Asset quality Sovereign debt crisis 290 0.0074 0.0096 -0.0009 0.0654

Southern Europe

Asset quality Subprime mortgage crisis 76 0.0042 0.0035 0.0000 0.0219

Asset quality Sovereign debt crisis 174 0.0111 0.0150 0.0000 0.1482

Eastern Europe

(34)

Table 3.7 shows the summary statistics of the net amount of loans ratios.

Europe Num.

Net amount of loans Normal Time 203 0.53 0.19 0.04 0.88

Net amount of loans US Financial Crisis

Time 214 0.57 0.19 0.04 0.87

Net amount of loans Greece Financial

Crisis Time 514 0.56 0.18 0.00 0.84

Northern Europe

Net amount of loans Normal Time 109 0.52 0.19 0.04 1

Time 116 0.54 0.20 0.04 0.87

Crisis Time 290 0.52 0.18 0.00 0.84

Southern Europe

Time 76 0.61 0.18 0.06 0.83

Crisis Time 174 0.61 0.16 0.09 0.83

Eastern Europe

Time 22 0.57 0.14 0.31 0.78

Crisis Time 50 0.62 0.12 0.33 0.81

Table 3.8 shows the summary statistics of the total amount of assets.

Europe Num.

Bank size normal time 203 253507 495813 118 2739361

Bank size Subprime mortgage crisis 214 328444 647254 181 3543974

Bank size Sovereign debt crisis 514 332000 621944 207 3484949

Northern Europe

Southern Europe

Eastern Europe

(35)

Table 3.9 shows the summary statistics of the return on average equity ratio.

Europe Num.

ROAE Normal Time 203 14.62 7.32 -24.34 33.11

ROAE US Financial Crisis Time 214 9.38 14.58 -86.68 44.82

ROAE Greece Financial Crisis Time 514 -1.74 34.18 -469.52 185.71

Northern Europe

ROAE Normal Time 109 15.14 6.82 -24.34 32.4

Southern Europe

ROAE Normal Time 73 12.82 7.85 -22.40 33.11

Eastern Europe

ROAE Normal Time 21 18.24 6.44 6.75 32.26

ROAE US Financial Crisis Time 22 17.33 7.24 6.75 32.26

(36)

Appendix 4: Results of the regression with the

total capital ratio

Europe Northern Europe Southern Europe Eastern Europe

Total capital ratio * normal times -0,0060 -0,0033 0,0024 0,0446

(0,0052) (0,0072) (0,0128) (0,0493)

Liquidity * normal times -0,1269 -0,5123 -0,1101 2,1711

(0,3082) (0,4270) (0,6408) (1,5099)

Deposits * normal times 0,4857 0,5677*** 0,6049 -2,4684

(0,2335) (0,2889) (0,5297) (1,9145)

Asset quality * normal times 11,78 -10,97 13,87 46,87

(10,52) (15,02) (22,80) (79,41)

Net amount of loans * normal times -0,1001 -0,3937 0,0708 1,1867

(0,2838) (0,3565) (0,7018) (1,9762)

Bank size * normal times -0,0141 -0,0142 0,0007 -0,1238

(0,0184) (0,0251) (0,0394) (0,2731)

ROAE * normal times 0,0034 0,0048 0,0044 -0,0005

(0,0047) (0,0072) (0,0090) (0,0190)

Total capital ratio * Subprime mortgage crisis 0,0020 0,0008 -0,0008 0,0891*

(0,0058) (0,0132) (0,0089) (0,0489)

Liquidity * Subprime mortgage crisis 0,2334 0,1085 0,0685 0,7766

(0,3183) (0,4270) (0,6394) (1,6615)

Deposits * Subprime mortgage crisis 0,1320 0,1965 -0,1285 0,1627

(0,2317) (0,2862) (0,5633) (1,9023)

Asset quality * Subprime mortgage crisis -0,17 -8,44 2,43 18,55

(11,83) (16,66) (31,36) (113,99)

Net amount of loans * Subprime mortgage crisis 0,2280 0,3055 -0,1189 -0,4299

(0,2886) (0,3770) (0,6633) (1,7275)

Bank size * Subprime mortgage crisis -0,0031 0,0067 0,0016 -0,0644

(0,0179) (0,0261) (0,0378) (0,3261)

ROAE * Subprime mortgage crisis -0,0023 -0,0058 0,0023 0,0077

(0,0048) (0,0066) (0,0113) (0,0207)

Total capital ratio * Sovereign debt crisis 0,0045* 0,0205*** 0,0032 -0,0097

(0,0024) (0,0071) (0,0036) (0,0370)

Liquidity * Sovereign debt crisis -0,1375 -0,4595 0,1372 -0,1605

(0,2119) (0,2837) (0,4078) (1,1619)

Deposits * Sovereign debt crisis 0,2052 0,5122*** -0,2187 -1,6451

(0,1632) (0,1898) (0,3883) (1,6624)

Asset quality * Sovereign debt crisis -34,66*** -11,46 -59,97** -33,54

(8,85) (11,87) (23,54) (102,74)

Net amount of loans * Sovereign debt crisis 0,0148 0,0391 0,4021 0,3178

(0,1988) (0,2494) (0,4385) (1,6175)

Bank Size * Sovereign debt crisis 0,0132 0,0413** -0,0122 -0,0197

(0,0119) (0,0164) (0,0259) (0,3200)

ROAE * Sovereign debt crisis 0,0009 -0,0001 -0,0096 0,0252

(0,0033) (0,0043) (0,0082) (0,0164)

Fixed effects dummies Yes Yes Yes Yes

Number of observations 917 503 321 93

Bank capital and the stock returns of the European banks

University of Amsterdam, Faculty Economics and Business

BSc Finance and Organization, Finance track

Bachelor Thesis

Bank Capital and the Stock Returns of the

European banks

Casper Kool, 10350055

June 2015

Abstract

Table of Contents

1. Introduction

2. Literature Review

2.1 Basel Accords

Basel I

Basel II

Basel III

2.2 Development of the crises

Development of the banking industry in Europe

Subprime mortgage crisis

European Sovereign debt crisis

2.3 Theories about capital

The stability of the financial sector

Risk taking

3. Data and Methodology

3.1 Sample selection

3.2 Empirical Model

Hypotheses

The Model

3.3 Summary of the statistics

4. Results

5. Conclusion

References

Appendix 1: Specification of countries in each

sample group

Appendix 2: Sovereign debt and government

bond's interest rates

Appendix 3: Summary statistics tables

Appendix 4: Results of the regression with the

total capital ratio