The effect of Basel III on value of US banks

The Effect of Basel III on Value of US Banks

Abstract

More enhanced financial regulation was introduced by the Basel III Accord. Among the stricter regulation are higher global minimum capital standards. Various literature describes the effect of capital on firm value. This paper studies the effect of the Basel III announcement on value of US banks. It further looks into the difference in the effect between high and low capitalized banks. An event study shows that there is a positive and negative effect among different time windows. Further analysis shows that there is no significant difference of the effect between low and high capitalized banks.

Bachelor Thesis

Economics and Business Finance and Organization Student: Dion Langedijk Student number: 10657843 Supervisor: J.E. Ligterink

2

Statement of Originality

This document is written by Dion Langedijk who declares to take full responsibility for the

contents of this document.

I declare that the text and the work presented in this document is original and that no

sources other than those mentioned in the text and its references have been used in

creating it.

The Faculty of Economics and Business is responsible solely for the supervision of

completion of the work, not for the contents.

3

Table of Contents

1. Introduction

2. Literature

2.1. The Basel Framework 5

2.1.1. The Basel Committee 5

2.1.2. Basel I 5

2.1.3. Basel II 6

2.1.4. Basel III 8

2.2. Capital structure and bank value 9

3. Hypotheses

4. Data

5. Research Methodology

5.1. Event study 13

5.1.1. Event of interest and timing 13

5.1.2. Specify a benchmark model 13

5.1.3. Analyzing abnormal returns 14

5.2. CAR analysis 16

6. Results

6.1. Results event studies 16

6.1.1. Event study December 2009 16

6.1.2. Event study September 2010 17

6.2. CAR analysis 18

6.2.1. CAR analysis December 2009 18

6.2.2. CAR analysis September 2010 20

7. Conclusion and Discussion

Reference list

4

1. Introduction

Capital restrictions have long been an issue in the financial banking system. For instance if they are effective (Hovakimian and Kane, 2000) and if there are costs for the rest of the economy (Van den Heuvel, 2008; Blum and Hellwig, 1995). The financial banking crisis that occurred in 2007 has been a perfect example of the insufficiency of the regulations. The banking sector was levered too much and had inadequate liquidity buffers. Banks had poor governance and risk management and

inappropriate incentive structures (Basel Committee, 2015a). Basel II was not

sufficient

On 12 September 2010 the Group of Governors and Heads of Supervision announced that higher global minimum capital standards and new liquidity requirements were to be implemented, referred to as the Basel III Accord. Basel III enhances Basel II in in a way that banks can absorb shocks arising from financial and economic stress. The Accord should ensure a more stable financial system by strengthening the regulation and risk management of the banking sector.

Basel III mainly focusses on increasing the quality and quantity of bank capital. Higher capital requirements could negatively influence the bank’s value. It has long ago been studied by Modigliani and Miller (1963) that high leverage increases a firm’s value, thus higher capital demands should decrease a bank’s value. Kraus and Litzenberger (1973) introduced the trade-off theory where optimal leverage cannot be too high due to financial distress costs associated with leverage. Mehran and Thakor (2011) mention bank value could be negatively influenced by increasing agency problems due higher capital ratios. However Berger and Bouwman (2009) show that there is a positive relation between capital and bank value by liquidity creation. Also Mehran and Thakor (2011) find a positive relationship between bank capital and value by increasing the value of loan portfolios. Blum (1999) designed a framework that increasing equity could induce the banks to take more risk to increase equity as equity becomes more valuable to the bank. Overall there are various insights in why capital should increase or reduce bank value. Therefore it should be interesting to see if banks suffered or benefited from announcing the stricter capital regulations Basel III imposes. The framework of Blum (1999) will also be taken to the test. Therefore the research question is:

What is the effect of the announcement of Basel III on the value of US banks and is there a difference in the effect between high and low capitalized banks?

An empirical analysis is on the capital structure of banks is done to help answer this question. Further an event study is conducted to measure the effect and the cumulative abnormal results are analyzed to see if there is a difference in between high and low capitalized banks.

5 The rest of the study is organized as follows. In section two the literature relevant to this topic is being discussed. In the third section the hypothesizes are developed. Section four describes the data used in this study. The research methodology is explained in section five. In section six the results of the study are provided. And at last, section seven will present the conclusion and discussion

regarding this analysis.

2. Literature

In this literature section there is a comprehensive explanation of the Basel Accords. Thereafter the relevant literature about capital structure and the effect on a banks value is described.

2.1. The Basel Framework

In this subsection the Basel Framework will be explained. Starting with an introduction to The Basel Committee. Following an elaboration on the Basel I, Basel II and Basel III Accords.

2.1.1. The Basel Committee

The collapse of Bretton Woods in 1973 caused many banks to incur massive foreign currency losses. In 1974 Bankhaus Herstatt was liquidated due to the fact that the amount of foreign exchange exposure it was exposed to amounted to three times its own capital. Foreign banks that had

unsettled trades with Herstatt therefore took big losses. The Franklin National Bank of New York also went bankrupt in the same year due to foreign exchange losses. These events indicated that there was a need to enhance the financial stability by improving the quality of banking supervision

worldwide. In response the Committee of Banking Regulation and Supervisory Practices was founded by the central bank governors of the G10 countries which was renamed to the Basel Committee on Banking Supervision later on. The membership has been expanded to 28 jurisdictions till today (Basel Committee, 2015a). Their objective is to strengthen the regulation, supervision and practices of banks worldwide with the purpose of enhancing financial stability (Basel Committee, 2015b). It has to be pointed out that the Committee only formulates supervisory standards and guidelines with the expectation the individual national authorities will implement them. It thus has no legal force. In 2012 the Committee also began monitoring the implementation of its standards to improve the resilience of the banking sector (Basel Committee, 2015a). This Committee introduced the three Basel Accords.

2.1.2. Basel I

The first Basel Accord, also referred to as the 1988 Basel Accord, is the result of several years of the Committee’s work to secure international convergence of supervisory regulations governing the

6 capital adequacy of international banks (Basel Committee, 1988). In the early 1980s the Committee was concerned about the worsening capital ratios of main international banks during a time with growing international risks. The committee made a plan for greater convergence in measurement of capital adequacy to stop the decreasing capital standards. The result was an accord on a weighted approach to the measurement of risk related to the different categories of assets and off balance sheet exposure (Basel Committee, 2015a). This presents a fairer measurement for international comparisons between banking systems with different structures. There are two central objectives of this accord. Firstly, the framework should serve to strengthen the soundness and stability of the international banking system. Also there were different capital adequacy requirements among different countries, which became a competitive issue. So, secondly, the framework should be fair and have a high degree of consistency in its application to banks in different countries with a view to diminishing an existing source of competitive inequality among international banks (Basel

Committee, 1988). This first Basel Capital Accord consisted of a minimum ratio of capital to risk-weighted assets of 8% (Basel Committee, 2015a).

2.1.3. Basel II

In June 2004 the Basel Committee published a revised framework for the 1988 Basel Capital Accord. This framework has been developed to further strengthen the soundness and stability of the international banking system. The framework had to improve the way regulatory capital

requirements reflect underlying risks and to better address the financial innovation of the recent years (Basel Committee, 2015a). The Basel II accord consists of three Pillars. These Pillars are:

I. Minimum capital requirement II. Supervisory review process III. Market discipline

Pillar I: minimum capital requirement

The first Pillar specifies the minimum capital requirements for a bank’s exposure to risk. The risk exposure is based on credit risk, market risk and the new added operational risk. The Basel Committee (2004) provide the banks with three options to measure their credit risk and calculate their capital requirements. These are the standardized approach and two internal ratings based (IRB) approaches, the foundation and advanced approach. The standardized approach uses external ratings and benchmarks, provided by external rating agencies. The IRB approach allows banks to use their own internal models for risk weighting and calculating their capital requirements, where the advanced approach allows more internal influence than the standard IRB approach. The market risk

7 is described by exposure to movements in market prices, such as interest rates and foreign exchange rates. Operational risk is the risk of loss resulting from inadequate or failed internal processes, people and systems or from external events (Basel Committee, 2004). To calculate operational risk there are also three methods provided. The Basis Indicator Approach, the Standardized Approach and the Advanced Measurement approach. Banks that developed more sophisticated operational risk measurement systems and practices should use a more sophisticated approach than the Basic Indicator Approach and which is appropriate for the risk profile of the institution (Basel Committee, 2004). Calculated with these measures the capital ratio must be no lower than 8% and the Tier 2 capital is limited to 100% of Tier 1 capital.

Pillar II: Supervisory review process

The intention of the supervisory review process is firstly to ensure that banks have adequate capital to support the all the risks in their business. Secondly it should encourage banks to develop and use better risk management techniques in monitoring and managing their risks (Basel Committee, 2004). There are four key principles for the supervisory review. Cited from the Basel II Accord (2004) these are:

1. Banks should have a process for assessing their overall capital adequacy in relation to their risk profile and a strategy for maintaining their capital levels.

2. Supervisors should review and evaluate banks’ internal capital adequacy assessments and strategies, as well as their ability to monitor and ensure their compliance with regulatory capital ratios. Supervisors should take appropriate supervisory action if they are not satisfied with the result of this process.

3. Supervisors should expect banks to operate above the minimum regulatory capital ratios and should have the ability to require banks to hold capital in excess of the minimum.

4. Supervisors should seek to intervene at an early stage to prevent capital from falling below the minimum levels required to support the risk characteristics of a particular bank and should require rapid remedial action if capital is not maintained or restored.

This holds in general that the supervisors should review and evaluate the capital adequacy strategies and should intervene if the processes are unsatisfactory.

Pillar III: Market discipline

The third Pillar is aimed at the disclosure requirements of banks and is to complement the first two Pillars. Banks are required to disclose information about capital risk exposures, risk assessment processes and their capital adequacy ratio. These disclosures also should be consistent with the approaches they used for measuring risk under Pillar I. Only then it would be an effective mean of

8 informing the market about the bank’s exposure to risks and provides a consistent disclosure

framework that enhances comparability (Basel Committee, 2004).

2.1.4. Basel III

The global financial crisis in 2007-2008 was a clear demonstration that there was a need for even more enhanced financial regulation. The banking sector had too much leverage and no liquidity buffers. Banks had poor governance and risk management and inappropriate incentive structures (Basel Committee, 2015a). In 2009 the Committee proposed to strengthen the capital and liquidity regulations. In September 2010 higher global minimum capital standards were announced by the Group of Governors and Heads of Supervision. This capital and liquidity reform package is called Basel III.

Basel III strengthens the three Pillars from Basel II in order for banks to absorb shocks arising from financial and economic stress with the objective to reduce the risk of being conveyed to the rest of the economy (Basel Committee, 2010a). It focusses on the increase of quality and quantity of capital in the banking sector. It does so by increasing the minimum capital requirements and adding new capital buffers to the existing Basel II requirements in Pillar I. The Accord also introduces two new ratios. The liquidity coverage ratio and the leverage ratio.

To increase the quality of the capital, the composition of capital has been changed. Capital now consists only of Tier 1 capital and Tier 2 capital. Tier 3 capital is eliminated. The focus is now more on common equity, which is the highest quality component of a bank’s capital (Basel Committee, 2010a). The Common Equity Tier 1 ratio now must be at least 4.5% of risk-weighted assets at all times. Tier 1 Capital of risk-weighted assets 6%. Tier 1 Capital plus Tier 2 Capital should always be at least 8% of risk-weighted assets (Basel Committee, 2010a). These ratios were phased in and had to be met at 1 January 2015. Further a capital conservation buffer of 2.5% of common equity will be added in addition to the Common Equity Tier 1 ratio. The conversation buffer will be build up outside of periods of stress to counter losses during periods of financial stress. This ratio will be phased in from 1 January 2016 until 1 January 2019. In the Appendix more information can be found on these ratios and their phase in.

Also a new minimum leverage ratio is introduced and is defined as:

𝑇𝑖𝑒𝑟 1 𝐶𝑎𝑝𝑖𝑡𝑎𝑙

𝑇𝑜𝑡𝑎𝑙 𝐸𝑥𝑝𝑜𝑠𝑢𝑟𝑒> 3%

The total exposure is measured by on-balance sheet items, repurchase agreements and securities finance, derivatives and off-balance sheet items. The ratio should always be above 3%.

9 Furthermore a liquidity coverage ratio (LCR) was introduced with the new Basel III Accord. The ratio is measured as followed:

𝐿𝐶𝑅 = 𝑆𝑡𝑜𝑐𝑘 𝑜𝑓 ℎ𝑖𝑔ℎ − 𝑞𝑢𝑎𝑙𝑖𝑡𝑦 𝑙𝑖𝑞𝑢𝑖𝑑 𝑎𝑠𝑠𝑒𝑡𝑠

𝑇𝑜𝑡𝑎𝑙 𝑛𝑒𝑡 𝑐𝑎𝑠ℎ 𝑜𝑢𝑡𝑓𝑙𝑜𝑤𝑠 𝑜𝑣𝑒𝑟 𝑡ℎ𝑒 𝑛𝑒𝑥𝑡 30 𝑐𝑎𝑙𝑒𝑛𝑑𝑎𝑟 𝑑𝑎𝑦𝑠 > 100%

This standard intends that a bank had an adequate level of high-quality liquid assets that can easily be converted into cash. This way it can meet its liquidity needs for a 30 calendar day time of severe liquidity stress. The value should be higher than the net cash outflows for those 30 days (Basel Committee, 2010b).

At last the net stable funding ratio (NSFR) was introduced. This ratio is measured as followed:

𝑁𝑆𝐹𝑅 =𝐴𝑣𝑎𝑖𝑙𝑎𝑏𝑙𝑒 𝑎𝑚𝑜𝑢𝑛𝑡 𝑜𝑓 𝑠𝑡𝑎𝑏𝑙𝑒 𝑓𝑢𝑛𝑑𝑖𝑛𝑔

𝑅𝑒𝑞𝑢𝑖𝑟𝑒𝑑 𝑎𝑚𝑜𝑢𝑛𝑡 𝑜𝑓 𝑠𝑡𝑎𝑏𝑙𝑒 𝑓𝑢𝑛𝑑𝑖𝑛𝑔> 100%

This measure complements the LCR in a way that the NSFR promotes the resilience of the bank over a more medium and long-term time period. It requires to have an available amount of stable funding that exceeds the amount of stable funding required over a one-year period of stress. This offsets incentives for banks to just fund their stock of liquid assets with funds that mature just outside the 30-day horizon (Basel Committee, 2010b).

2.2. Capital Structure and Bank Value

Modigliani and Miller (1958) introduced a very important theorem on the modern thinking on capital structure. The theorem contains two propositions. The first proposition implies that the value of a levered firm is identical to the value of a unlevered firm. Hence the value of the firm is independent of its capital structure. This proposition assumes perfect market conditions. In that case the cash flow that is paid out to the firm’s security holders is equivalent to the total cash flow generated by the firm’s assets. Thus assets and securities have the same value. Their second proposition proposes that the cost of capital cannot be reduced with debt. The cost of equity will increase when more debt is acquired caused by the more risk that is imposed on the remaining equity.

As mentioned, the theorem from Modigliani and Miller (1958) assumed perfect market conditions. They improved this theorem by including taxes (Modigliani and Miller, 1963). Because interest payments are deductible, a firm financed only by debt will pay less taxes than an equity financed firm. Therefore a firm will maximize its firm value by only being financed by debt. Thus higher minimum capital requirements should result in a lower firm value due to smaller tax

10 reductions. Furthermore, the cost of capital will now decrease when a firm acquires more debt due to the tax reductions (Modigliani and Miller, 1963). This implies that a firm should only have debt in its capital structure to maximize firm value by fully exploiting the benefits of the tax shield.

The trade-off theory proposed by Kraus and Litzenberger (1973) expanded the Modigliani and Miller theorem with financial distress costs. Because when a firm increases its leverage the chance that it cannot meet its debt obligations anymore increase. The increasing chance of bankruptcy leads to financial distress costs. The trade-off theory therefore states that the optimal capital structure is where the tax benefits of debt are offset by the financial distress costs (Kraus and Litzenberger, 1973). However high leverage may not have the same meaning for financial firms and nonfinancial firms. Whereas high leverage more likely indicates financial distress for nonfinancial firms it may not be the case among financial firms where high leverage is normal (Fama and French, 1992).

For banks the choice of capital may be different. It is widely assumed that the previously mentioned theorems do not apply to banks because financial intermediaries arise to resolve informational and other frictions (Mehran and Thakor, 2011). The standard textbook answer for the deviation on the Modigliani and Miller capital indifference theorem is given by Mishkin (2000). He states that banks only hold capital because they are required to do so by the regulatory authorities. Because of the high costs of holding capital bank managers want to hold less bank capital than is required by the regulatory authorities. Thus the amount of bank capital is determined by the bank capital requirements (Mishkin, 2000). However, Harding, Liang and Ross (2007) find that there is an interior optimal capital ratio where banks voluntarily choose to maintain capital in excess of the minimum capital requirement in order to balance the risks of insolvency against the benefits of additional debt. As long as there is a significant financial burden associated with violating capital requirements. Their model takes deposit insurance, taxes and minimum fixed capital standards into account.

The agency problem between bank managers and shareholders could also be affected by capital (Mehran and Thakor, 2011). This can hurt bank value in a way that a lazy or incompetent manager may pay lower dividends to accumulate more capital and may also make poorer loans (Mehran and Thakor, 2011). Referred to previous described literature a higher equity ratio should have a lower after-tax return on equity. However, Berger (1995) showed that book values of return on equity and capital ratio are positively related for banks which increases bank value. He explains this by saying that banks that expect to perform better transmit this information through higher capital. Bank managers may have private information about future cash flows. When the

management has a stake in the value of the bank through personal ownership, it is less costly for a bank to signal high quality through a capital increase (Berger, 1995). Another explanation he gives is

11 that an increase in capital raises the expected earnings of the by reducing the expected costs of financial distress. Berger and Bouwman (2009) studied that there is a positive relation between capital and liquidity creation for large banks and negative for small banks. They also show that liquidity creation and bank value are positively correlated. The effect is positive for large banks because they are subject to greater regulatory scrutiny and market discipline than small banks, which affects their capacity to absorb risk (Berger and Bouwman, 2009). Diamond and Rajan (2001) found that greater bank capital reduces the probability of financial distress but also reduces liquidity creation which in turn reduces bank value. Banks can monitor its borrowers. This feature is missing on equity, which results in reduced liquidity creation. However Mehran and Thakor (2011) add to this that Diamond and Rajan (2001) did not examine the cross-sectional relationship between bank capital and value. If for example is assumed that bankruptcy costs are invariant across banks but riskier assets allow banks to create more liquidity, then higher bankruptcy costs are faced by banks that create more liquidity. Therefore they keep more capital and be worth more due to their higher liquidity creation (Mehran and Thakor, 2011). A study on bank capital and value in the cross section was done by Mehran and Thakor (2011). Their results show that bank capital and bank value are positively correlated. They explain this relation by two benefits from higher capital. Firstly higher capital leads to higher endogenously determined survival probability for the bank at the interim point in time. This increases the marginal benefit of monitoring loans in the first period due to the higher survival probability that increases the odds of collecting the gains from first-period

monitoring. Thus banks with higher capital monitor more. Secondly, the value of the relationship loan portfolio increases with the larger monitoring (Mehran and Thakor, 2011). Value could also decrease with the Basel III risk sensitive capital requirements because banks that could cause banks to switch from loans to less risk, lower yielding securities (Thakor, 1996). Eyssell and Arshadi (1990) conducted an event study on the announcement effect of the Basel I Accord on the value of banks. Their results showed that the announcement had a negative impact on the value of the banks.

Blum (1999) shows in a dynamic framework that increasing capital requirements may increase a bank’s riskiness. He states that if a bank has to increase its capital in the upcoming time, the bank may increase their asset risk. He provides two intuitions. Firstly, a higher capital

requirement lowers the expected profit of the bank as equity is more expensive. With lower profits, there is less to lose when the bank goes bankrupt and therefore a smaller incentive to avoid default. Therefore increasing risk is less costly for the bank the stronger the capital restriction. Secondly the marginal return on risk changes with the increase of the capital requirement. Equity becomes more valuable to the bank when they have to meet a higher capital ratio. Because when under a binding capital requirement the amount that a bank can invest in a profitable risky asset is restricted to a multiple of the value of their equity. This implies that an additional unit of equity leads to an

12 additional investment larger than one unit in the risky asset. Therefore the bank has a higher

incentive to increase equity. When its costly for a bank to raise equity on the capital market the only way to increase the amount of equity tomorrow is to increase risk today (Blum, 1999).

3. Hypotheses

The study of Modigliani and Miller (1958) gives the theory that to maximize value the bank should be completely financed by debt. So increasing capital requirements from Basel III should reduce the value of banks. Financial distress costs caused by leverage should decrease be less with higher capital requirements (Kraus and Litzenberger, 1973), which would increase the value of the bank. But this may not be the case for banks, where high leverage is normal (Fama and French, 1992). Higher capital induces larger monitoring on loans which increases the value of the bank’s loan portfolio (Mehran and Thakor, 2011). The findings of Diamond and Rajan (2001) that capital reduces liquidity creation has been countered by Mehran and Thakor (2011) that they are making too broad

assumptions. And the effect of capital on liquidity creation has proven to be positive for large banks by Berger and Bouwman (2009). Further the framework presented by Blum (1999) implies that the banks have to take more risk to increase their capital for the new regulations which should have higher expected profits and thus increase bank value. As lower capitalized banks should raise more equity this effect should be larger for low capitalized banks. Therefore I expect a positive relationship with the increasing bank capital requirements from Basel III and the value of the US banks and this effect to be larger for low capitalized banks. This leads to the following two hypotheses:

H1 : The announcement effect of Basel III has a positive effect on the value of US banks.

H2 : The positive effect is larger for low capitalized banks.

4. Data

For this study the data is collected from the CRSP/Compustat merged database. From the Bank Annual database all the financial data needed for this research is gathered for the Standard Industrial Classification (SIC) code 6020, which represents commercial banks in this database. After filtering for US banks the sample consisted of 260 US commercial banks for the event study of 12 September 2010 and 259 for 17 December 2009. For these banks the Permno’s were collected and used in Eventus to conduct a daily cross-sectional event study. For the CAR analysis the financial data of the banks from the last previous year were collected.

13

5. Research Methodology

The research methodology contains of an illustration of the event study used and the CAR analysis that is done after.

5.1. Event Study

The study of Fama, Fisher, Jensen and Roll (1969) pioneered the event study methodology. They investigated if stock prices behave differently around stock splits than in normal periods. Therefore they studied the behavior of stock prices around stock splits by comparing the holding returns on the stock around the event date and the expected return without the event happening. The difference between these returns is what they call the abnormal return. They conducted statistical tests over the aggregated abnormal returns to test the hypothesis.

Bowman (1983) made a 5 step model for conducting event studies. These 5 steps are reduced to three by de Jong (2007) and are also the three steps which are used in this paper. The first step he mentions is identifying the event of interest and in particular the timing of the event. Secondly he addresses to specify a “benchmark” model for normal stock return behavior. And last calculate and analyze abnormal returns around the event date.

5.1.1. Event of Interest and Timing

To determine the effect of the Basel III accord on the bank’s values the event of interest is the event date the Basel III accord was first announced to be implemented. The timing of the event is the first day the information reaches the market, thus the press release date of the announcement. The official announcement of Basel III was made on 12 September 2010 by the Group of Governors and Heads of Supervision, the oversight body of the Basel Committee on Banking Supervision. They announced a substantial strengthening of existing capital requirements and the introduction of a global liquidity standard (Basel Committee, 2010c). However on 17 December 2009 the Basel Committee on Banking and Supervision had issued a package of proposals to strengthen the global capital and liquidity regulations for consultation (Basel Committee, 2009). So there also could have been an effect on the stock returns on this date already. Therefore there are two events of interest being studied in this paper.

5.1.2. Specify a Benchmark Model

The benchmark model is used for calculating the abnormal returns of stocks around the event period. Choosing the benchmark model therefore is an important step when studying the stock return behavior. The abnormal returns are calculated as follow:

14 ARit = Rit – E(Rit)

The abnormal returns for stock i at time t (ARit) is defined as the actual return of stock i at time t (Rit)

minus the expected returns from the benchmark model for stock i at time t (E(Rit)), often also

referred to as the normal returns.

In this study the market model is used as the benchmark model. This model is specified as this:

Rit = αi + βiRmarket + εit

In this equation Rit is the return of stock i on time t. The return is determined by a constant αi and βi

represents the market risk the stock is exposed to. Rmarket represents the return of the market

indexTo calculate the expected returns the market model gets estimated by:

E(Rit) = αi + βiRmarket

Where αi and βi are OLS estimates of the regression coefficients. These parameters need to be

estimated over an estimation period, [T1, T2],which is prior to the event period [t1,t2]. The event date itself, which is the announcement of Basel III is indicated by t = 0. The estimation period used is 180 days which ends 60 days prior to the announcement.

The event period starts five days before the event date and ends ten days after. There will always possible rumors going on before the official announcement and therefore a five day period before the event date is used in the event period. A period of ten days is chosen for after the period to estimate the after effect, which should be long enough. A longer period will decrease the power of significance in the event study. A longer term period increase the likelihood of other events taking place that will have effect on the stock prices. Those events will decrease the significance of the event study.

5.1.3. Analyzing Abnormal Returns

When analyzing the abnormal returns the average of the abnormal returns of all banks has to be calculated. This is due the fact that a lot of stock price movements are caused by information that is unrelated to the event of interest. Therefore analyzing them separately is not very useful and averaging them would improve the result of the analysis (de Jong, 2007). The unweighted cross-sectional average of abnormal returns is used in this paper and is calculated by:

15 𝐴𝐴𝑅𝑖= 1 𝑁 ∑ 𝐴𝑅𝑖𝑡 𝑁 𝑖=1

Large deviations of the average abnormal returns from zero indicate abnormal performance. The average of the abnormal returns should reflect the effect of that particular event, because all other information that is not related to the event should cancel out on average when all these abnormal returns are all centered around one particular event (de Jong, 2007).

Because we are interested in a longer period around the event the cumulative abnormal returns are calculated to study this performance. In this case the abnormal returns are accumulated from the start of the event period t1 up to time t2. This is done as followed:

𝐶𝐴𝑅𝑖 = 𝐴𝑅𝑖,𝑡₁+ . . + 𝐴𝑅𝑖,𝑡₂= ∑ 𝐴𝑅𝑖𝑡 𝑡2

𝑡=𝑡1

In this study the CARs for event periods [0], [0,3], [0,5], [0,10], [-5,5] and [-5,10] are calculated. To obtain the cumulative average abnormal returns, the CARs are again accumulated over the cross-section of events. This is done as followed:

𝐶𝐴𝐴𝑅 = 1

𝑁 ∑ 𝐶𝐴𝑅𝑖

𝑁

𝑖=1

After all this it is time to test if the CARs are significantly different from zero. This will be tested by a simple t test which will be conducted as followed:

𝑡 𝑠𝑡𝑎𝑡𝑖𝑠𝑡𝑖𝑐 = 𝐶𝐴𝐴𝑅 − 𝐸(𝐶𝐴𝐴𝑅𝑖𝑡) 𝑆𝐷(𝐶𝐴𝑅)/√𝑁

SD(CAR) is the standard deviation of the cumulative abnormal returns for the event window and N the sample size. E(CAARit) the expected cumulative average abnormal results which equals zero, as

the null hypothesis states:

𝐻0∶ 𝐸(𝐶𝐴𝐴𝑅𝑖𝑡) = 0

When the t statistic is higher or equal to 1.65, 1.96 or 2.58 the result is significant at the ten, five or one percent level respectively.

16

5.2. CAR Analysis

𝐶𝐴𝑅𝑖,(𝑡−𝑛,𝑡+𝑛) = 𝛼 + 𝛽1𝐻𝐶 + 𝛽2𝑅𝑂𝐸 + 𝛽3𝑆𝐼𝑍𝐸 + 𝛽4𝑀𝐵 + 𝜀

The CAR is the dependent variable and the main independent variable of interest is the dummy variable HC. HC equals zero if the risk adjusted capital tier 1 ratio is lower than the average of the sample and one if higher. Thus if β1<0 and significant the hypotheses will be correct that low capitalized banks have higher abnormal returns due to the announcement than high capitalized banks. Furthermore there is a constant and three control variables added. The first control variable ROE is return on equity which is a proxy for the profitability of the bank. A high profitable bank should have higher returns. The second control variable is the size of the bank (SIZE). Fama and French (1993) showed that size has a significant effect on the stock returns. The size of the bank is measured by the logarithm of the bank’s total assets. At last the market to book ratio (MB) is added, which represents the market value of the bank over its book value. Chen & Zhao (2006) state that firms with higher market to book ratios are more profitable as the ratio is a measure of growth.

6. Results

In this chapter the results are provided. In the first part the event studies that are conducted are presented. Secondly the results of the CAR analyses are demonstrated.

6.1. Results Event Studies

In this section two event studies are conducted. One on the first announcement that took place on 17 December 2009, where the first package of proposals was issued by the Basel Committee on Banking Supervision. Secondly an event study is done on the second announcement on 12 September 2010. On this date the proposals were agreed upon and the official requirements published.

6.1.1. Event Study December 2009

The first event study is conducted on the event date of 17 December 2009 and shows the effect of the first Basel III announcement where the first proposals were issued. This study is done over six different event windows which the results for are shown in table 1. As seen in the table the CAAR on the event date [0,0] itself is slightly positive with a value of 0.7596 percent. This means the US banks have an abnormal return of 0.7596 percent on average. This result is also highly significant at the 1% level with a t statistic of 3.67017. Furthermore the CAARs over all periods are positive, which means that the announcement made on 17 December 2009 had a positive influence on the value of US

17 banks. The CAARs of period [0,3] and [0,5] are 1.6407 and 1.846 percent respectively and are both highly statistical significant at the one percent level. The CAAR of period [0,10] has the highest value with a value 2.5889% and is also highly significant at the one percent level. The periods 5,5] and [-5,10], where the days before the event date are included, the CAARs do show positive values, but are both not significant at the ten percent level. These results confirm the hypothesis that the value of US banks will be higher after the announcement of Basel III and that increasing capital will increase the value.

Table 1: Results event study December 2009

17 December 2009 Event window [0,0] [0,3] [0,5] [0,10] [-5,5] [-5,10] CAAR 0.007596 0.016407 0.01846 0.025889 0.002807 0.010237 St. Dev. (0.033374) (0.065691) (0.077843) (0.086471) (0.097265) (0.104439) T-test 3.67017*** 4.02722*** 3.82386*** 4.82768*** 0.46542 1.580477 Observations 259 259 259 259 259 259 Significant levels: 1% ***, 5% **, 10% *

6.1.2. Event Study September 2010

The second event study is done on the official announcement of Basel III, where the definite capital requirements and liquidity standards were published. This event study is done over the same six event windows as in the previous event study. In table 2is seen that the also here the CAAR on the event date itself,[0,0], is slightly positive with a value of 0.9575 percent. Also this value is highly significant at the one percent level. The time frame of [0,5] is the only other period which shows a positive CAAR, which is very little and also not significant at the ten percent level. All of the other periods show negative CAARs of which are [0,10] and [-5,10] significant at the one percent level with values of -1.6 and -2.356 percent respectively. The periods [0,3] and [-5,5] are not significant at the ten percent level. The value on the event date is in line with the hypotheses and is also significant. The negative values could be explained by the fact that a higher capital ratio indeed results in lower bank value as described in (Thakor, 1996; Modigliani and Miller, 1958).

Table 2: Results event study September 2010

12 September 2010 Event window [0,0] [0,3] [0,5] [0,10] [-5,5] [-5,10] CAAR 0.009575 -0.00394 0.001689 -0.016 -0.00586 -0.02356 St. Dev. 0.031148 0.048731 0.081631 0.096754 0.108582 0.123839 T-test 4.95659*** -1.30332 0.333535 -2.66709*** -0.87069 -3.06707*** Observations 260 260 260 260 260 260 Significant levels: 1% ***, 5% **, 10% *

18

6.2. CAR Analysis

In this section the analyses on the CARs of the US banks are discussed. This analysis is done by performing a regression as mentioned in the research method. First the analysis of 17 December 2009 is treated and secondly the analysis of 12 September 2010.

6.2.1. CAR Analysis December 2009

In this analysis the hypotheses is tested if low capitalized banks have higher abnormal returns than high capitalized banks during the event window of the first announcement on 17 December 2009. Therefore the coefficient of the dummy variable High Capital ratio should be negative in order for this to be true. In table 3 these results are shown.

As seen in this table the variable for a high capital ratio is positive in all event windows. This indicates that if banks have a high capital ratio, it will have a positive impact on their abnormal returns. For example on the event date itself [0,0] the abnormal returns should be 0,1217 percent higher if the bank is high capitalized. This is against the hypotheses. The positive influence could be explained by the assumption that high capitalized banks would have less changes to make to reach the new capital requirements. These changes could be costly for low capitalized banks, which ultimately would have a negative influence on their abnormal returns. However none of the dummy variables are statistically significant at at least the ten percent level. So there is no statistical evidence that a high capital ratio indeed does have a positive influence on the value of the banks. But this also means that there is no statistical evidence entirely that the capital ratio does have an effect on the abnormal returns of banks at all. Thus the hypotheses that lower capitalized banks should have higher abnormal returns than high capitalized banks will be rejected.

Furthermore the only variable that seems to have an significant effect on the abnormal returns is size. This variable is slightly positive in the event windows [0,0] and [0,3] and significant at the one percent level.

The fact that almost none of the variables are significant can be due to the R-squared being really low for every event window. The R-squared indicates the proportion of the variance in the dependent variable that is predicted by the independent variables. In this case the R-squared is the highest on the event day itself, which is only 4.91 percent. Thus the variables only explain 4.91 percent of the variance of the cumulative abnormal returns. This is a very small percentage and could be the cause of the insignificancy of the variables.

In the development of the hypotheses was assumed that the banks had to increase their equity in order to reach the new capital requirements. Low capitalized banks therefore had to increase more equity than higher capitalized banks. According to Blum’s (1999) framework this should induce more risk taking behavior for low capitalized banks, which in turn should increase their

19 expected returns and in turn the bank’s value. However, most of the banks in the sample used for this study already had a capital ratio above the required capital ratio imposed by Basel III. The banks did not have to make an adjustment to their ratio, which could be cause of the biased and

insignificant results.

Table 3: Results CAR analysis December 2009

A correlation matrix is added in table 4 If there are high correlations between the independent variables there could be a problem of multicollinearity which results in biased estimates. In the case of multicollinearity two variables are almost perfect linear combinations of another. The correlation is calculated by the following formula:

𝜌𝑥,𝑦=

𝑐𝑜𝑣(𝑥, 𝑦) 𝜎𝑥𝜎𝑦

Cov(x,y) is the covariance between variable X and Y. σx is the standard deviation of X and σy the standard deviation of Y.

In table 4 is seen that there are no high correlations between the variables and thus should not have influence on the results of the regression.

17 December 2009

Event window [0,0] [0,3] [0,5] [0,10] [-5,5] [-5,10]

High Capital ratio 0.001217 0.013208 0.002136 0.000328 0.007859 0.006051 St dev (0.004328) (0.008547) (0.008664) (0.009888) (0.011668) (0.012815) T statistic 0.28 1.55 0.25 0.03 0.67 0.47 Return on equity 0.009948 0.001211 0.006355 0.009 0.016508 0.019154 St Dev (0.014185) (0.020219) (0.027121) (0.026118) (0.027115) (0.024624) T statistic 0.7 0.06 0.23 0.34 0.61 0.78 Size 0.003601 0.007592 -0.000202 -0.00501 0.001989 -0.00282 St Dev (0.001211) (0.002547) (0.003087) (0.003081) (0.003822) (0.003917) T statistic 2.97*** 2.98*** -0.07 -1.63 0.52 -0.72 Market to Book -0.0006 0.001134 0.0086 -0.00247 0.017191 0.006123 St Dev (0.003413) (0.008064) (0.01091) (0.01179) (0.010966) (0.012326) T statistic -0.18 0.14 0.79 -0.21 1.57 0.5 Constant -0.02 -0.05009 0.012179 0.067963 -0.02988 0.025907 St Dev (0.011733) (0.021218) (0.025918) (0.028597) (0.034615) (0.037439) T statistic -1.7* -2.36** 0.47 2.38** -0.86 0.69 Observations 259 259 259 259 259 259 R-squared 0.0491 0.042 0.0074 0.0103 0.0275 0.0124 Significant levels: 1%***, 5%**, 10%*

20

Table 4: Correlation matrix December 2009

6.2.2. CAR analysis September 2010

In this second CAR analysis again the hypotheses if low capitalized banks have higher abnormal returns than high capitalized banks, but for the event date 12 September 2010. These results are shown in table 5. The same situation here is that the dummy variable High Capital ratio should be negative in order for this to be true.

As the table shows the variable is negative in the two event windows [0,10] and [-5,10]. This is in line with the hypotheses. However both these variables are not significant either. For the other four event windows the variables are again positive, but again not significant. Therefore the

hypotheses for this event date should again be rejected.

In this analysis size and the market to book ratio only have a significant effect on the abnormal returns. For size this is in the event windows [0,0], [0,10] and [-5,10]. These effects are negative in the [0,10] and [-5,10] event windows and positive on the event date itself. However these are again only small influences that do not get higher than one percent. The market to book ratio seems to have a negative influence on the abnormal returns of 1.129 percent in the event window [0,3] and is significant at the five percent level.

The same case holds here that the R-squared is very small for the regressions which does not reach higher than 2.68 percent in any of the event windows. Thus again this could be the case of the lack of significances.

Also in this year almost all of the banks in the used sample had an capital ratio above the required ratio that was imposed by Basel III. So this also could have been the cause of the insignificant results in this event study.

HC ROE Size MB ratio

HC 1

ROE 0.1702 1

Size -0.1318 0.0859 1

21

Table 5: Results CAR analysis September 2010

12 September 2010

Event window [0,0] [0,3] [0,5] [0,10] [-5,5] [-5,10]

High Capital ratio 0.005878 0.002465 0.005261 -0.00096 0.005579 -0.00065 St dev (0.004086) (0.006098) (0.009041) (0.010658) (0.011215) (0.013152) T statistic 1.44 0.4 0.58 -0.09 0.5 -0.05 Return on equity -0.00014 -0.00061 -0.013387 -0.00432 -0.01691 -0.00784 St Dev (0.004319) (0.009009) (0.009031) (0.015593) (0.017276) (0.024014) T statistic -0.03 -0.07 -1.48 -0.28 -0.98 -0.33 Size 0.002198 -0.00193 -0.002379 -0.00879 -0.00244 -0.00885 St Dev (0.000993) (0.001679) (0.002354) (0.00317) (0.003369) (0.004345) T statistic 2.21** -1.15 -1.01 -2.77*** -0.72 -2.04** Market to Book -0.00459 -0.01129 -0.000494 -0.00726 0.008578 0.001811 St Dev (0.003344) (0.005334) (0.008103) (0.00893) (0.010168) (0.010594) T statistic -1.37 -2.12** -0.06 -0.81 0.84 0.17 Constant -0.00639 0.019796 0.017454 0.059363 0.00218 0.044089 St Dev (0.010352) (0.016479) (0.022884) (0.030056) (0.032652) (0.04121) T statistic -0.62 1.2 0.76 1.98** 0.07 1.07 Observations 260 260 260 260 260 260 R-squared 0.018 0.0232 0.0085 0.0268 0.0064 0.0142 Significant levels: 1%***, 5%**, 10%*

The correlation matrix of this regression is added in table 6. Also in this matrix is seen that there are no high correlations between the independent variables and thus should not affect the results of the regression.

Table 6: Correlation matrix September 2010

HC ROE Size MB ratio

HC 1

ROE 0.1689 1

Size -0.1277 0.0845 1

MB ratio 0.3136 0.3647 0.1646 1

7. Conclusion and Discussion

The event studies show that the Basel III announcements did have a significant effect on the stocks of the US banks. For the announcement in December 2009 these were all positive and significant at the one percent level for the event windows [0,0], [0,3], [0,5] and [0,10], which is in line with the

literature (Berger and Bouwman, 2009; Berger, 1995; Mehran and Thakor, 2011; Blum, 1999) and confirms the first hypotheses. The announcement made in September 2010 had different and mixed

22 results. On the event date itself it had a highly significant positive effect of 0.9575%, which is also in line with the literature (Berger and Bouwman, 2009; Berger, 1995; Mehran and Thakor, 2011; Blum, 1999). But there are negative significant effects for the event windows [0,10] and [-5,10]. This could be explained by investors anticipating on the result that banks will switch from loans to less risk, lower yielding securities due to the risk sensitive capital requirements from Basel III that Thakor (1996) describes. The first hypotheses is confirmed for the event date itself but rejected for the other event windows.

Further the CAR analyses show that there are is not any significant evidence that there is a difference in the effect on the value of the banks between high and low capitalized banks. The second hypotheses therefore is rejected. An explanation for this could be that investors already anticipated the information before the announcements were made. This is quite likely because as soon as the Basel Committee will meet, there will be rumors going around. And after the financial crisis investors could partly expect this new regulations would contain stricter capital requirements. Another explanation could be that the dummy variable used in the regression is based on capital ratios of the US banks which most of them were already above the Basel III capital

requirement. Therefore a low capital ratio would imply that the bank does not have to raise a lot of capital or any capital at all and could not verify Blum’s (1999) framework. This may have led to biased results and therefore it may not be fit to answer the research question.

Also the R-squared of the regressions are really low. This might imply that there are a lot more factors that influenced the stock returns of banks in the studied time periods. This results in a omitted variable bias, which could disturb the estimates of the used variables.

Further the Basel III Accord is as described, much more than just a minimum capital

requirement. The research could be improved by using more sophisticated variables of capital ratios in the regression that give a better reflection of the complexity that matches the Basel III Accord. This data is unfortunately really hard to acquire.

23

Reference list

Basel Committee. (2015a). A brief history of the Basel Committee. Basel Committee on Banking Supervision, Basel.

Basel Committee. (2015b). About the Basel Committee. Accessed at 15 June 2016, from http://www.bis.org/bcbs/about.htm?m=3%7C14%7C573

Basel Committee. (2010a). Basel III: A global regulatory framework for more resilient banks and banking systems. Basel Committee on Banking Supervision, Basel.

Basel Committee. (2010b). Basel III: International framework for liquidity risk measurement, standards and monitoring. Basel Committee on Banking Supervision, Basel.

Basel Committee. (2010c). Group of Governors and Heads of Supervision announces higher global minimum capital standards. Accessed at 10 June 2016, from

http://www.bis.org/publ/p100912.htm

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

Basel Committee. (2004). International Convergence of Capital Measurement and Capital Standards: A Revised Framework. Basel Committee on Banking Supervision, Basel.

Basel Committee. (2009). Strengthening the resilience of the banking sector. Basel Committee on Banking Supervision, Basel

Berger, A. N. (1995). The Relationship between Capital and Earnings in Banking. Journal of Money, Credit and Banking, 27, pp. 432-456.

Berger, A. N., & Bouwman C. H. S. (2009). Bank Liquidity Creation. Review of Financial Studies, 22, pp. 3779–3837.

Blum, J. (1999). Do capital adequacy requirements reduce risks in banking? Journal of Banking and Finance, 23, pp. 755-771.

Blum, J., & Hellwig, M. (1995). The Macroeconomic Implications of Capital Adequacy Requirements for Banks. European Economic Review, 39, pp. 739–749.

Bowman, R. G. (1983). Understanding and Conducting Event Studies. Journal of Business Finance and Accounting, 10, pp. 561-584.

Chen, S. (2011). Capital ratios and the cross-section of bank stock returns: Evidence from Japan. Journal of Asian Economics, 22, pp. 99-114.

Diamond, D. & Rajan, R. (2000). A theory of bank capital. Journal of Finance, 55, pp. 2431-2465. Eyssell, T., & Arshadi, N. (1990). The wealth effects of the risk-based capital requirement in banking:

24 Fama, E. F. & French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of

Financial Economics, 33, pp. 3-56.

Fama, E.F., Fisher, L., Jensen, M.C., Roll, R. (1969). The adjustment of stock prices to new information. International Economic Review, 10, pp. 1-21.

Fama, E. F. & French, K. R. (1992). The cross-section of expected stock returns. Journal of Finance, 47, pp. 427-465.

Harding, J. P., Liang, X., & Ross, S. L. (2007). The Optimal Capital Structure of Banks: Balancing Deposit Insurance, Capital Requirements and Tax-Advantaged Debt. University of Connecticut, Economics Working Papers. Paper 200729.

Hovakimian, A., & Kane, E. (2000). Effectiveness of capital regulation at U.S. commercial banks, 1985 to 1994. Journal of Finance, 55, pp. 451-468.

Jong, F.de. (2007). Event Studies Methodology. Tilburg University.

Kraus, A., & Litzenberger, R. H. (1973). A state-preference model of optimal financial leverage. The Journal of Finance, 28, pp. 911-922.

Mehran, H., & Thakor, A. (2011). Bank Capital and Value in the Cross-Section. The Review of Financial Studies, 24, pp. 1019-1067.

Mishkin, F. (2000). The Economics of Money, Banking and Financial Markets. New York: Addison Wesley.

Modigliani, F. & Miller, M. (1958). The Cost of Capital, Corporation Finance, and the Theory of Investment. American Economic Review, 48, pp. 261-297.

Modigliani, F. & Miller, M. (1963). Corporate Income, Taxes and the Cost of Capital: A Correction. American Economic Review, 53, pp. 433-443.

Thakor, A. V. (1996). Capital Requirements, Monetary Policy, and Aggregate Bank Lending: Theory and Empirical Evidence. Journal of Finance, 51, pp. 279–324.

Van den Heuvel, S. (2008). The Welfare Cost of Bank Capital Requirements. Journal of Monetary Economics, 55, pp. 298–320.

25

Appendix

Table 1

Basel III phase-in arrangements