An empirical study of deposit banks in Turkey B ’

B

ANKING REGULATION

’

S IMPACT ON

MORAL HAZARD

An empirical study of deposit banks in

Turkey

Author

Jeroen van der Hoeven

1272012

Supervisor

Dr. C.L.M. Hermes

August 2008

UNIVERSITY OF GRONINGEN

Faculty of Economics & Management and Organization

MSc International Business and Management

Specialization International Financial Management

&

UPPSALA UNIVERSITY

B

’

An empirical study of deposit banks in Turkey

ABSTRACT

This study researches how deposit banks in Turkey were affected by the regulatory changes that took place in 1999. In the years prior to 1999 Turkey experienced some major financial crises. By looking at previous research and theory several sources of banking instability are identified, including: information asymmetry, weak regulatory and supervisory agency, deposit insurance, and moral hazard. New regulations were put into place in order to stabilize the financial sector as well as the cancelling of the full deposit insurance fund. The main aim of this research is to provide a link between the effects of these regulations on the level of moral hazard taken by deposit banks in Turkey. This study looks at banks’ asset quality; capital requirement; foreign exchange exposure; and liquidity. The statistical findings show that the deposit banking sector was affected by the regulatory changes. However against all predictions, it seemed that the moral hazard problem only increased in the years following 1999. As an implication it seems that the supervisory and regulatory infrastructure was not yet ready to fill the gap of the cancelled deposit insurance.

Key words: banking regulations, moral hazard, information asymmetry, deposit insurance, deposit banking sector, Turkey.

T

ABLE OF CONTENTS

ABSTRACT ………2

LIST OF ABBREVIATIONS………..4

LIST OF FIGURES AND TABLES……….5

1 INTRODUCTION………...6

2 LITERATURE REVIEW………...8

2.1Information problems………...…………..8

2.2Types of information asymmetry………...…………9

2.3Deposit Insurance………..………11

2.4 Regulations and its effect on Moral Hazard……….………12

3 OVERVIEW OF THE TURKISH BANKING SECTOR………...14

4 TYPES OF DEPOSIT BANKS………...18

5 METHODOLOGY AND HYPOTHESES………20

6 DATA DESCRIPTION……….25

7 EMPIRICAL RESULTS………...28

8 SUMMARY AND CONCLUSION………..34

REFERENCES………37

APPENDIX I………...42

APPENDIX II………..43

L

BRSA

Banking Regulation and Supervision Authority

FDI

Foreign Direct Investment

FXL

Foreign Exchange Risk on Loans

FXR

Foreign Exchange Risk on Liabilities

IMF

International Monetary Fund

LIQ

Liquidity Ratio

NCR

Credit Risk

SCR

Standard Capital Ratio

SDIF

Savings Deposits Insurance Funds

L

Figure 1: Rate of financial arbitrage (1992-2002)

15

Table 1: Overview of testable predictions

25

Table 2: Sample specification

27

Table 3: Results T-test mean values

28

Table 4: Results T-test mean values (continued)

29

Table 5: Results T-test median values

32

Table 6: Results T-test median values (continued)

33

Table 7: Significance T-test mean values

34

Table 8: Significance T-test median values

34

1 Introduction

This research analyses the effects of changes in banking regulation on the financial performance of Turkish deposit banks. Bank failures have been a significant problem in many countries over the world and have occurred in developing as well as developed countries. Therefore the need for effective banking regulation and supervision is a crucial factor for having sound bank activities and has been widely discussed in the literature (for example Diamond and Dybvig, 1983; Barth et al., 2004; 2007). Banking regulations are put in place for providing stability in the financial sector. One of the factors that could lead to an unstable financial sector is moral hazard. This paper focuses on how certain banking regulations affect the behaviour of deposit banks regarding moral hazard.

the full deposit insurance system on the risk taking of banks a plan for gradual reduction of the deposit insurance was put into place starting in 1999. Another problem was the existing information asymmetry between the deposit banks and the customers and between the deposit banks and the supervisory agency. The Banking Regulation and Supervision Authority (BRSA) was established to attack this problem by enforcing the transparency and disclosure of banking activities. This study researches the effect that these new changes had on the behaviour of individual deposit banks. The main research question will therefore be:

How did changes in banking regulation affect the level of moral hazard of deposit banks in Turkey?

The main objective of this research is to gain insight in the way banking regulation affects the risk taking behaviour among commercial banks. In the end, this paper may contribute to the existing knowledge of regulation and its effect on moral hazard.

2 Literature Review

In this research we focus on information asymmetry and in particular the role of moral hazard in financial markets. Information gaps exist in all layers of society and are nothing but the mere fact that some people have more information about certain subjects than others. According to Akerlof (1970), who was first to introduce the subject of information asymmetry, an economy has information asymmetry when some parties to business transactions may have an information advantage over others. For example, if someone wants to buy a car he knows less about the car than the seller. The seller may not disclose all the information, because that would fetch him a lower price for the car. As a result the price does not perfectly reflect the quality of the product. In financial markets, information asymmetries arise exactly the same way. However, due to the nature of financial markets the problem may be worse, because it usually takes longer before a transaction is fully completed. Information asymmetry exists as a bank or other lender (for example depositor) doesn’t know as much as the borrower. In real (product/service) markets, the transaction of the good or service is usually immediately or within a certain time. When transaction is completed the customer can decide whether his experience was good or bad and which again provides him information whether to purchase the same good in future or not. However, in financial markets transaction is often completed somewhere in the future. The information asymmetry is not just there for a certain time but continues till service is completed. Regarding loans and mortgages this could mean several years. A bank manager has to evaluate the loan on a continuous base till the entire loan plus interest is paid back. Financial services are often related to important activities for the customer. Especially pension savings and mortgages are important since they largely control the well being of a customers’ financial future. They have no direct information about the soundness of the provided services and can only trust the bank agent on his ethical behaviour. It is this nature of financial markets that makes information asymmetry such an important issue.

2.1 Information problems

speculation1 increases, hoping that future profits will be enough to pay today’s debt. According to Bernanke et al. (1999) it is the existing information asymmetry that makes this speculative behaviour possible. In their article they explain how banking problems arise due to information asymmetry resulting in moral hazard and uncertainty which could end in bank runs. When economy is doing well, loans are given with less caution and investors invest assuming/ speculating the economy would continue to grow. As there are a large number of projects, the bank should put more effort in separating good and bad borrowers. Depositors should consider banks not just only on high interest rate but also on their sound banking activities in order to know if their deposits are safe. As economy slows down, or a macro shock occurs banks and customers suddenly start to become highly risk averse. This could lead to a bank run. As Diamond and Dybvig (1983) point out a banking crisis is a loss of confidence in the banking system or fear that the bank will go bankrupt. In turn this leads to a run on banks as individuals and companies simultaneously withdraw their deposits and the bank’s reserves are not sufficient to cover the withdrawals. Since the cash reserve a bank keeps on hand is only a small fraction of its deposits, a large number of unexpected cash withdrawals will deplete available cash and force a bank to sell assets or even to close, even though it could have been fundamentally sound prior to the run. As a result other customer may lose in their bank, and start withdrawing their money as well. This could result in a chain reaction of bank runs, eventually worsening the banking sector as a whole. Thus if a single bank can’t meet its obligations it could reduce the public confidence in the whole banking system and even lead to a banking crisis. If depositors lose their confidence, damage could be done to all economic activities, as the banks are no longer able to supply capital to the rest of the economy.

2.2 Types of information asymmetries

Several forms of information asymmetries can be discerned and throughout the years many studies have been performed regarding the (adverse) effects of them (Chiyachantana et al., 2004). The main players that shape the banking environment are: the (i) government (including central bank and supervisory institution); (ii) the banks (bank agents); and (iii) the banks’ customers (borrowers as well as depositors). By making this distinction we can explain the information problems that exist between

1

those players. Asymmetric information problems can arise between regulatory/supervisory institution and banks; banks and borrowers; banks and shareholders; and between depositors and banks. The focus of this thesis will be on information asymmetries that exist between depositors and banks. Information asymmetry leads to two problems in the financial system: adverse selection and moral hazard.

Adverse Selection is an asymmetric information problem that takes place before the transaction of a service or good. It is a process where bad choices are made due to information asymmetry. For example, people who want to take on high risks are most willing to take out a loan because they are less likely to pay it back. The problem is that a lender has difficulty in distinguishing between good-risk and bad-risk investments, because borrowers tend to be better informed about their investment projects than outside investors (lenders) (Bernanke and Gertler 1989). Thus the problem is that borrowers who take out loans have much better information about potential returns and risk associated with the investment projects they plan to undertake than lenders do. The information asymmetry reduces banks' ability to monitor, creatinginefficiency as more low-quality borrowers obtain financing. As a result the ratio of ‘bad’ borrowers increases and the amount of nonperforming loans eventually rise, thereby decreasing the bank’s asset quality.

intentions and damage the lending bank. The problems due to moral hazard and adverse selection are important constraints to the efficiency of the financial system and could even unstable the banking sector. In order to minimize moral hazard and adverse selection problems more information is needed through monitoring, screening and disclosure. Governments can help by implementing regulations that enforces the transparency in the financial system but they also set capital requirements to reduce the incentives to take on too much risk. Another role of the government is to provide some security to individual depositors and stabilize the banking sector, usually by means of deposit insurance but, as will be explained hereafter, this could increase moral hazard problem.

2.3 Deposit Insurance

Recognizing the huge fiscal costs when a bank run happens, most countries have put into place financial safety nets to decrease the vulnerability of the banking sector to the contagion effects of individual bank failure (Demirguc-Kunt and Detragiache, 1998). A much used measure against bank runs is the deposit insurance system. It has been argued that such a deposit insurance might enhance the stability of the banking system (at least for a while), since it takes away the incentives for a bank run by ensuring that depositors will get their savings from the government when a bank is unable to repay the deposits. (Humphrey, 1976).

greater risks than they otherwise would. As a result moral hazard problem leads to excessive risk taking behaviour and eventually could deteriorate banks’ balance sheets (Barth et al., 2004; 2007; Boyd et al., 1998; Demirgüc-Kunt and Detragiache, 2002). McKinnon (1999) shows in another study that poorly capitalized banks are prepared to gamble on the basis of government deposit insurance. These banks might accept foreign deposits as an ongoing source of finance for loans denominated in the domestic currency, increasing the maturity mismatch unless a regulatory authority forces them to hedge. Due to the moral hazard behaviour depositors no longer have any reason to be concerned with the financial condition of the bank since in case of failure they’ll get their savings back anyway. This makes depositors less careful in their selection of their banks. Knowing this, banks will take on more risk since they know that the government, at the expense of the taxpayer, will provide the money if the bank itself can’t meet its obligations to the depositors (Antinolfi et al., 2001). Because banks know that the chance of a bank run is much smaller due to the deposit insurance, it will not need that many liquid cash. So instead of keeping a sound quantity of liquid capital, banks will provide more loans to risky projects in order to make high interest profits. But as result it puts itself financially in an unsound position, since it will have less capital to meet its debt obligations (Demirguc-Kunt and Detragiache, 1998; 2002; Barth et al., 2004).

Thus with financial safety nets, the stability of the economy could be an illusion in which underneath government guarantees to banks and risky spending could lead to fundamentally unsound practices (Krugman, 1999). Corsetti et al. (1998) describe that implicit guarantees led banks to engage in moral hazard lending. Any protection provided to banks during a banking crisis greater than they expected, could increase their risk-taking in the future. During a crisis the authorities are facing a trade-off between maintaining financial stability today, through offering protection to failing banks, and jeopardising future financial stability through increasing moral hazard later on. In other words due to the deposit insurance the chance of a bank run might be smaller, but it could result in fundamentally weaker banks.

2.4 Regulations and its effect on Moral Hazard

banks with certain incentives that adversely affects the interest of the depositors and or government. Many studies showed that deposit insurance and information asymmetry could increase moral hazard behaviour by banks (e.g. Bayir, 2001; Hellman et al., 1997; Antinolfi et al., 2001). Since information asymmetry can be one of the factors behind a banking crisis, the disclosure of information could be one way of preventing such a crisis.

4 Overview of the Turkish banking sector and regulatory changes

world, the domestic asset markets was already dependent on short-term and speculative movements of foreign capital flows. But with the pegged exchange rate, the Turkish economy offered arbitrage rates that reached 100 percent in 1996 and up to 60 percent in 1999. This financial arbitrage has been calculated by Yeldan (2002) as the end result of the conversion of initially the foreign exchange into Turkish Liras at a certain rate, and after earning the rate of interest offered in the domestic asset markets; it is reconverted back to the foreign currency at the foreign exchange rate. Figure 1 clarifies to what degree the Turkish asset market was exposed to financial arbitrage.

Figure 1. Rate of financial arbitrage (1992-2002)

Source: Yeldan, 2002

Turkey (among others, like Basle Committee and IMF) looked for ways of ensuring the soundness of the banking sector by introducing or changing some banking regulations that focused on the information gap aspects. Following the crisis many changes took place; the most important contribution during that year was the introduction of the Banks Act that led to the establishment of the Banking Regulation and Supervision Agency (BRSA). The BRSA was formed following the enactment of the Bank Act No, 4389. This agency has five pillars on which it derives its existence. These main goals are:

• To enhance banking sector efficiency and competitiveness • To maintain confidence in the banking sector

• To minimize the potential risks to the economy from the banking sector • To enhance the soundness of the banking sector

• To protect the rights of the depositors

A more extensive overview of the main goals of the BRSA can be found in appendix I. As an independent agency, BRSA would not face the same political pressures that plagued the supervisory functions of the Treasury in the years previous to 1999 (Soral et al., 2003). By removing the involvement of the government from decisions in the area of supervision, other than the appointment of the members of the Board the BRSA became fully autonomous. This was an important contribution, because to be effective these institutions must have a considerable degree of independence from short-term political motives but at the same time, they must have broad long term political support to make a difference. However, over the past 30 years, Turkey has been governed by 22 governments (Kibritcioglu, 2005). One can imagine the effect this had on the treasury, that had constantly to follow the direction of the new government. Thus the Banks Act had to increase transparency and independence in the operation of the BRSA, strengthen regulations, and to provide all the tools needed to improve the solving of banking problems. Many regulations were imposed focusing on all different subjects in the banking sector. A list of main changes on information asymmetry and moral hazard that were imposed in 1999 can be found hereafter:

- Banks have to comply with accounting practices in accordance with BASEL I and BIS

- Capital requirements and liquidity ratios were formally set at higher level

- Banks have to keep, publish and present their annual balance sheets and profit and loss statements in accordance with principles and procedures to be laid down by the Agency

- To protect the depositors the BRSA will especially monitor the disclosure of information on the foreign exchange position of banks as well as the quality of their assets

- BRSA may take any action to prevent depositors from being misinformed including publication of such balance sheet or profit and loss statement

- In the event that the required measures are not taken the Agency can authorize to take and implement all such measures as are necessary for the secure operation of the bank and for the protection of depositors

- The coverage of the deposit insurance fund was limited in order to decrease moral hazard behaviour and to increase the soundness of banking activities

particular, it adjusted accounting rules to require consolidated accounting and proper valuation of securities. It also fully implemented capital adequacy and foreign exposure limits rules. In addition, in order to assure the strictest compliance with these regulations it introduced penalties for foreign exchange positions in excess of prudential limits. Finally, it issued regulations on internal risk management systems and amended capital adequacy rules to take into account market risks.

This research will look at those changes in banking regulation and how it affected the deposit banks in the sector. Did it have the anticipated effect? Bayir (2001) showed in his study that full coverage deposit insurance enacted in 1994 deteriorated the soundness of Turkish Banking Sector and had a major contribution in the failure of commercial banks. If deregulation and deposit insurance deteriorated the Turkish Banking Sector, would the abandoning of deposit insurance and enacting new regulations strengthen the sector? In the rest of the paper we will have a look at those questions.

4 Types of deposit banks

Although the focus of this paper is on deposit banks in the Turkish banking sector, it is also useful to make a distinction between the different deposit banks. This will provide us with more information about how certain banks differently react to changes in banking regulations. Like in many countries different types of banks with different characteristics exist in Turkey. The dominant type in Turkey is the deposit bank. They on average account for more than 90% of the total assets in the banking sector during the past decade (Banking Association of Turkey). The other 10 percent consists of investment and development banks and special finance houses2 (Ali, 2003). We focus on the deposit banks and make the further distinction by splitting these in state owned, private domestic owned, and foreign owned deposit banks.

2

Addressing the different characteristics could be important for understanding how regulations affect the banks behaviour. According to Miam (2003) the main differences between the three types of banks are based on the ownership of the cash flows and control rights of the banks. Foreign banks are privately owned and managed, but unlike private domestic banks their cash flow and control rights mainly rest with foreign shareholders. These foreign shareholders are typically part of a larger chain of multi national banks spread over the world. Because of the reputation of their (often) worldwide brand, foreign banks might have adopted more conservative banking policies. In addition foreign banks are usually being double supervised and regulated by domestic as well as foreign (home country) supervisors, resulting in more accounting disclosure and sound banking practices (Detragiache & Gupta, 2004). The many internal regulations, being subjective to newest accounting rules, their world wide reputation and being double supervised all points to the fact that these foreign banks are already subjective to higher standards of disclosure and regulations. In this paper we therefore predict that the foreign banks are less subjective to the changes in Turkey since their standards are already higher than other banks in the sector.

According to Dages et al. (2000) private domestic banks appear in emerging market economies to be more aggressive in their lending and they hold less liquid assets than foreign banks and or state owned banks. In general they tend to hold more assets in the form of loans. And because of their competitive nature these loans tend to be of less quality due to speculative allocation of capital. In line with Dages we predict that regulatory changes that focus on these matters will have a profound effect on the behaviour of private domestic banks, more than on foreign or state-owned banks.

social and political unrest, like Turkey during the nineties, government agents are easily prone to corruptive behaviour as is shown in the study of Ali (2003). This is also true for managers of state owned banks. The presence of government subsidies allows banks to operate without passing on any risk of default to their depositors. This could create a serious moral hazard problem. In fact the absence of ownership rights, and hard budget constraints may even make the bank managers corrupt or politically motivated in their lending decisions (Dinc, 2003). This paper predicts that state owned banks are less subjective to supervisors and regulations than private domestic banks and they have a certain amount of moral hazard due to the close relationship between the owner and regulator that in most cases are the same (the government).

5 Methodology

As a reaction to the problems in the banking sector, the Turkish government decided in 1999 to strengthen its banking regulations and decrease the moral hazard behaviour among banks. This led to the changes explained in chapter three. In fact it hoped that the changes that took place would affect moral hazard in three ways. By setting financial ratios, the government could directly control the moral hazard behaviour of the banks. By limiting deposit insurance system it opened to take away the adverse effects of it, namely the incentives for moral hazard behaviour. Finally, by enforcing the disclosure of financial information, information asymmetries were hoped to decrease as well as the adverse effect of it (which lead to the moral hazard problem). The objective of this paper is to find out how these changes have affected the level of moral hazard of deposit banks. The main hypothesis in this research predicts that the changes in regulations have a profound effect upon the moral hazard behaviour of deposit banks. The proposition is that the changes made to enhance transparency and decrease moral hazard correlate positively and significantly with some key balance sheet ratios. Therefore we look at 4 variables that are explicitly mentioned as being prone to moral hazard. Those variables will be discussed hereafter.

Capital adequacy:

safety and soundness of banks (Estrella et al 2000). The informal use of ratios by bank regulators and supervisors goes back well over a century (Mitchell, 1909). It determines the capacity of the bank in terms of meeting liabilities. The contractual relationship between banks and depositors gives the bank the freedom to use the deposits as they see fit, however it must have cash available whenever a customer asks for it. This means that with the exception of a small percentage that has to be kept in cash, the bank will try to use the deposits to maximise its profits, since it doesn’t expect all customers to withdraw their money at once (Campbell, 2007). Regulators and supervisors will try to ensure that banks have sufficient capital to keep them out of troubles. The capital ratio is used to protect the individual depositors and to promote the efficiency and stability of the whole banking sector. Turkish’ banks have to disclose this ratio in their financial statements and it is calculated as the amount of capital to risk weighted assets, non-cash credits and obligations3. This ratio measures whether the bank has sufficient capital against losses resulting from existing or future risks. One of the changes in regulations was focused on increasing the capital requirement of banks. These requirements are put into place to ensure that deposit banks are not participating investments that increase the risk of default and that they have enough capital to sustain operating losses while still honouring withdrawals.In this study it is predicted that the changes that took place in 1999 increased the bank’s capital adequacy significantly. For the measurement of capital adequacy the standard capital ratio (SCR) is used and is calculated as the bank’s capital divided by risk-weighted assets, non-cash credits and obligations. So the predicted hypotheses are:

H0: There is no significant difference between SCR prior to and after 1999

H1: There is a positive significant difference between SCR prior to and after 1999

Asset quality:

A common cause to bank problems is the failure to make accurate assessment of credit risk. Credit risk is the potential that a bank borrower or counterparty will fail to meet its obligations in accordance with agreed terms (Basel, 1999). The goal of adequate credit risk assessment is to maximise a bank's risk-adjusted rate of return by maintaining credit risk exposure within acceptable boundaries. However, the presence of information

3

asymmetry makes a proper assessment more difficult. In periods of economic growth this could especially be a problem as banks focuses on the expansion of the balance sheet (Minsky, 1986). Optimism about the economic future and competitive pressure to maintain market share rather than profitability puts proper credit assessment to the second place. By not assessing credit risk accordingly, banks may be tempted to deny credit to the cautious customer with less risky projects and therefore less expected return and as a result does not want to pay too much interest. By only accepting highest bidders banks might end up with a high ratio of ‘bad’ borrowers4 and therefore a higher chance of not being repaid on time or at all. This adverse selection is due to asymmetric information between bank and customer. Banking regulations, accounting and disclosure laws are there to decrease the information asymmetry. In 1999 the government of Turkey together with BRSA imposed new regulations on the disclosure of financial information. These regulations forced the banks to disclose information about the amount of non-performing loans, which gives a clear sign to the outside world of the soundness of the bank. Non-performing loan is a loan of which the principal and or interest has not been paid over three months from the due date specified in the contract (Central Bank of Turkey). As banks do not want to spread negative signals, this would lead to more accurate assessment of credit and pricing, and therefore less non-performing loans. For deposit banks loans are usually the largest of the asset items and can therefore carry the greatest amount of potential credit risk (Basel, 1999). Thus calculating the amount of non-performing loans as a percentage of total loans approaches the quality of credit assessment for deposit banks. Here it is predicted that the regulatory changes had the expected effect and led to better asset quality. For the measurement of asset quality, this paper uses credit risk (NCR) and is calculated as nonperforming loans divided by total loans. The predicted hypotheses are:

H0: There is no significant difference between NCR prior to and after the regulatory changes in 1999.

H1: The NCR prior to the regulatory changes in 1999 is significantly higher than after the regulatory changes in 1999.

Foreign exchange exposure:

4

Foreign exchange risk is the exposure of a bank’s financial strength to the potential impact of changes in foreign exchange rates. The risk is that adverse fluctuations in exchange rates may result in a reduction in measures of financial strength. If banks have open foreign exchange positions than the earnings from banking activity are prone to changes in the foreign exchange rate. This exposure should be limited to a certain degree. In time of economic growth, banks need capital to grow. However, in emerging markets this capital is not always available so banks turn to foreign investors and tend to speculate that future earnings will offset the risk in foreign exchange rate changes. To limit these losses, the BRSA monitors the open foreign position of banks and makes sure that the open foreign position is not too big. However, Mckinnon and Pill (1999) show that deposit insurance systems provides incentives to banks to increase their foreign borrowing. They show that banks that enjoy government guaranteeshave an incentive to increase foreign borrowing and incur foreign-exchange risks that are underwritten by the deposit insurance system.In the absence of capital controls, this increases the chanceof over borrowing and leaves the bank both more vulnerableto speculative attack and more exposed to the real economicconsequences of such an attack. Since loans are the most important asset for deposit banks this study uses the amount of foreign currency denominated loans to total loans as a proxy for foreign exchange exposure on the asset side. Another indicator of exposure is the amount of foreign currency denominated liabilities to total liabilities, which calculates the foreign exchange exposure on the liability side. The new regulations are predicted to have a decreasing effect on the foreign exchange risk of the deposit banks in Turkey due to decrease in moral hazard. Foreign exposure is measurement with two variables. The first variable Foreign Exchange Risk on Liabilities (FXR) is calculated as foreign exchange liabilities divided by total liabilities. The other variable Foreign Exchange Risk on Loans (FXL) is calculated as foreign exchange loans divided by total loans. The following hypotheses are predicted.

H0: There is no significant difference between FXR (and FXL) prior to and after the regulatory changes in 1999.

Liquidity:

By offering long term loans paid with short-term debts (deposits) subjects the bank to bank runs. During a bank run, depositors rush to withdraw their deposits because they expect the bank to fail in repaying their deposits. Such a sudden withdrawal can force the bank to liquidate its assets at a loss, since they only have a small percentage of the deposits kept in cash (Diamond, 2007). Besides being subject to bank runs, they also are subject to interest rate risk since deposit banks traditionally borrow short and lend long. Due to this maturity mismatch they expose themselves to interest rate risk, as the liability side of the bank balance sheet will be more sensitive to interest rate changes due to short maturities. The BRSA imposed regulations that forced banks to disclose more information about their maturity mismatches, hoping that this indirectly would force banks to decrease their maturity problems. But the BRSA also set limits to the level of mismatches, thereby directly influencing the liquidity level. By decreasing the coverage of the deposit insurance the risk from depositor runs increased. This would indirectly force the bank to keep more cash available for withdrawals. To keep track of the maturity and liquidity mismatches on bank balance sheets the ratio liquid assets to total assets is calculated. The new regulations are predicted to have an increasing effect on the liquidity of the deposit banks in Turkey due to increase in transparency, thereby decreasing the moral hazard problem. Liquidity is measurement with Liquidity Ratio (LIQ) and is calculated by dividing liquid assets by total assets. The hypotheses are as follows.

H0: There is no significant difference between LIQ prior to and after the regulatory changes in 1999.

Table 1 summarizes the testable predictions and empirical variables for this study.

Variable Proxies Predicted Relationships

Standard Capital Ratio = Capital / (Risk-weighted Assets, Non-Cash Credits and Obligations) (SCR)

Capital Adequacy SCRpost>SCRpre

Asset Quality Credit Risk = Nonperforming Loans / Total Loans (NCR) NCRpost<NCRpre Liquidity Liquid Assets / Total Assets (LIQ) LIQpost>LIQpre

Foreign Echchange Risk = Foreign Exchange Liabilities /

Total Liabilities (FXR) FXRpost<FXRpre Foreign Echchange Risk = Foreign Exchange Loans / Total

Loans (FXL) Foreign Exposure

FXLpost<FXLpre

Note: The symbols POST and PRE in the predicted realtionship collumn stand respectively for the period after 1999 and the period prior to 1999

Table 1

Overview Testable Predictions

As already mentioned the Turkish deposit banking sector can be distinguished in the same way mentioned by Mian (2003). He distinguishes deposit banks in three groups: (i) state-owned banks, (ii) private domestic banks, and (iii) foreign owned banks. Each of this bank group has its own characteristics as explained in the literature part. However, it is expected that not all of the banks in the TBS are subjected in the same degree to the changes. For example, the full deposit insurance system covers the domestic deposits of all deposit banks operating in Turkey. Since state owned deposit banks are already protected due to their public nature, it is expected that changes in deposit insurance will have less influence on their behaviour. This paper will analyse for each different banking group how changes in regulations on have affected their balance sheet variables. This will provide us with information whether some banks are more subject to the changes than others.

6 Data Description

To analyse the predicted hypotheses financial data on individual banks is used. This chapter will provide information about the used data.

leads to moral hazard problem and that these effects can be reduced by adequate banking regulations (Chiyachantana et al., 2004; Hellman et al., 1997; Barth et al., 2004; 2007; Boyd et al., 1998; Demirgüc-Kunt and Detragiache, 2002). In this study we are going to research whether the changes in regulations had the expected effect on the moral hazard problem by looking at balance sheet variables. This research model is focused on the impact of changes in regulations on the financial performance of banks operating in Turkey.

The measurement of the role of regulations on the behaviour of banks is a complex task, since it is obviously impossible to filter all other factors that could have an influence. However this research adopted an approach that looks at post-balance sheet data of individual banks and compares it with the pre-data. The year 1999 is taken as the turning point since the BRSA was established that year and a gradual reduction of the full deposit insurance started from that date on5. To test the research hypotheses, the empirical indicators for each bank for an 11-year and a 7-year period are compared. This will provide information about the long-term and middle long-term. Finally the median of each variable is calculated for each bank over both time periods. The year 1999 is excluded from the analysis since the change took place that year and therefore can’t be attributed to one of the periods.

The individual bank’s balance sheet data used in this research were obtained from the Banks Association of Turkey. The sample includes privately owned banks, state owned and foreign owned banks. There are 10 years of data on 37 individual banks. The only selection criteria for the banks have been the amount of available information. This means that banks who went bankrupt (or disappeared by other means) in the period before the changes took place have not been taken into account. Of course one should notice that this could bias the selection since as a result we only used the strongest banks that survived the turmoil years. A detailed list of banks that has been used for this research will be provided in Appendix I. Table 2 details the sample. The source of data used in this study is the Banking Association of Turkey.

5

Bank Specification Number of Cases

Research Sample 37

State owned Commercial Banks 4

Privately owned Commercial Banks 28

Foreign owned Commercial Banks 5

source: Banking Association of Turkey

Sample Specification Table 2

7 Empirical results

This part of the paper discusses the empirical results. For each of the variables it will be examined whether the change in regulations had a significant effect. The empirical findings obtained from first t-test that uses the mean ratios and looks at 11 year period are summarized in Table 3.

Table 3

Results T-test mean values

Variables (-)5y (+)5y diff p Capital Adequacy All 50,9 97,3 46,4 0,165 State banks 13,6 45,7 32,1 0,160 Private banks 20,9 26,4 5,5 0,453 Foreign banks 16,4 25,2 8,8 0,432 Asset Quality All 10,7 71 60,3 0,011** State banks 3,8 37,3 33,5 0,018** Private banks 3,5 25,8 22,3 0,037** Foreign banks 3,4 7,9 4,5 0,395 FX Liabilities All 157 151,1 -5,9 0,093* State banks 30,1 29,8 -0,3 0,944 Private banks 59 55,1 -3,9 0,066* Foreign banks 67,9 66,2 -1,7 0,850 FX Loans All 142,5 121,3 -21,2 0,001*** State banks 29,7 24 -5,7 0,033** Private banks 55,7 46,6 -9,1 0,001*** Foreign banks 57,1 50,7 -6,4 0,677 Liquidity All 142,1 115 -27,1 0,005*** State banks 31,1 25 -6,1 0,309 Private banks 48,2 39,7 -8,5 0,033** Foreign banks 62,8 50,3 -12,5 0,046**

Note: * indicates statistical significance at the 10 (*), 5(**), 1(***) percent level.

Table 4

Results T-test mean values (continued) Variables (-)3y (+)3y diff p Capital Adequacy All 50,7 84,5 33,8 0,245 State banks 15,3 38,7 23,4 0,201 Private banks 20,3 23 2,7 0,628 Foreign banks 15,1 22,8 7,7 0,357 Asset Quality All 8,6 73 64,4 0,007*** State banks 3,5 43,2 39,7 0,025** Private banks 2,4 21,3 18,9 0,036** Foreign banks 2,7 8,5 5,8 0,352 FX Liabilities All 158,9 158,9 0 0,562 State banks 29,4 31,7 2,3 0,593 Private banks 60,8 59,1 -1,7 0,431 Foreign banks 68,7 68,1 -0,6 0,937 FX Loans All 144,4 130,5 -13,9 0,006*** State banks 31,6 27,5 -4,1 0,033** Private banks 56,3 48,4 -7,9 0,001*** Foreign banks 56,5 54,6 -1,9 0,677 Liquidity All 138,5 115,9 -22,6 0,01*** State banks 29,8 21,9 -7,9 0,309 Private banks 46,8 39,9 -6,9 0,033** Foreign banks 61,9 54,1 -7,8 0,046**

Note: * indicates statistical significance at the 10 (*), 5(**), 1(***) percent level.

The hypotheses to test the effect of changes in regulation on Capital Adequacy are: H0: There is no significant difference between SCR prior to and after 1999

The hypotheses to test the effect of changes in regulation on Asset Quality are:

H0: There is no significant difference between NCR prior to and after the regulatory changes in 1999.

H1: The NCR prior to the regulatory changes in 1999 is significantly higher than after the regulatory changes in 1999.

The asset quality was calculated as the ratio of nonperforming loans to total loans as an indicator of credit risk. Here, H0 is rejected for State owned banks and Private banks for both the long term and middle long term. According to the finding the average ratio of nonperforming loans for those banks was significantly lower before 1999. No significant difference can be found for the Foreign owned banks, so H0 cannot be rejected for them. Thus according to these findings we can state that the changes in 1999 had a certain significant impact upon the amount of non-performing loans. However, we predicted a decrease in the ratio of non-performing loans because banks would be more precautious in providing loans to bad borrowers. But as we can see in appendix III, the ratio of NPL increased significantly against the predictions. The reason for this is probably because since 1999 banks have strict rules how to measure their assets, including NPLs. The banks are also carefully monitored on the accuracy of their financial reports. Thus banks report their NPLs more accurately. If this is this reason than one can also conclude that the financial disclosure has increased and therefore asymmetric information decreased together with moral hazard problem.

To test the effect of changes in regulation on Foreign Exposure on Liabilities the following hypotheses were made:

H0: There is no significant difference between FXR prior to and after the regulatory changes in 1999.

H1: The FXR prior to the regulatory changes in 1999 is significantly higher than after the regulatory changes in 1999.

regulation/supervision those banks could act more speculative and make use of short term foreign capital inflows. Bringing those standards to a higher level, the effect would be much higher for domestic banks than for foreign banks, since they already had to comply with high standard of legislation and disclosure rules in their home country.

The hypotheses to test the effect of changes in regulation on Foreign Exposure on Loans are:

H0: There is no significant difference between FXL prior to and after the regulatory changes in 1999.

H1: The FXL prior to the regulatory changes in 1999 is significantly higher than after the regulatory changes in 1999.

The results show that the change in regulation had a significant effect upon the Private banks, since their average Foreign Loan Exposure different significant before and after 1999 on the long term as well as on the middle long term. The FXL of the State banks also different significantly, but only on the long term. No significant changes could be found for the Foreign owned banks; therefore H0 could not be rejected for them. But looking at the banking sector as a whole, we could state with 99% certainty that the changes had significant impact upon the exposure on foreign loans. Why different banks were affected as they did, can be explained in the same way as it was explained for FXR.

The hypotheses to test the effect of changes in regulation on Liquidity are:

H0: There is no significant difference between LIQ prior to and after the regulatory changes in 1999.

H1: The LIQ prior to the regulatory changes in 1999 is significantly lower than after the regulatory changes in 1999.

the banking sector with less liquid assets behind than it did in previous runs. A second reason could have been that the banks were forced to disclose financial information more accurately while using new accounting rules. This might have resulted in more accurate report of the liquidity ratio.

Table 5

Results T-test median values

Variables (-)5y (+)5y diff p

Capital Adequacy All 18,8 27,3 8,5 0,147 State banks 14,2 45,3 31,1 0,153 Private banks 19,9 25,2 5,3 0,448 Foreign banks 16 26,9 10,9 0,425 Asset Quality All 4,1 21,4 17,3 0,029** State banks 3,1 36,2 33,1 0,013** Private banks 2,3 21,1 18,8 0,05** Foreign banks 18,1 8,1 -10 0,6* FX Liabilities All 57,2 52,9 -4,3 0,021** State banks 30,2 27,8 -2,4 0,474 Private banks 60 55,2 -4,8 0,321 Foreign banks 68,2 65 -3,2 0,745 FX Loans All 52,8 44,9 -7,9 0,003*** State banks 29,3 24,3 -5 0,117 Private banks 56,2 47,3 -8,9 0,0002*** Foreign banks 56,5 52 -4,5 0,798 Liquidity All 48 38,7 -9,3 0,006*** State banks 31,8 24,2 -7,6 0,316 Private banks 47,7 39,1 -8,6 0,044** Foreign banks 63 47,9 -15,1 0,041**

Table 6

Results T-test median values (continued)

Variables (-)3y (+)3y diff p

Capital Adequacy All 18,8 25,3 6,5 0,207 State banks 16,3 39,9 23,6 0,255 Private banks 19,8 23,2 3,4 0,573 Foreign banks 14,7 22,8 8,1 0,315 Asset Quality All 2,2 21,2 19 0,011** State banks 2,7 53,4 50,7 0,039** Private banks 1,9 18,4 16,5 0,063* Foreign banks 3,7 9 5,3 0,503 FX Liabilities All 58,3 56,8 -1,5 0,485 State banks 29,4 34,1 4,7 0,379 Private banks 61,4 58,9 -2,5 0,321 Foreign banks 69,3 67,7 -1,6 0,826 FX Loans All 52,7 47,4 -5,3 0,057* State banks 29,9 27,1 -2,8 0,437 Private banks 56,1 49,7 -6,4 0,024** Foreign banks 55,9 54,7 -1,2 0,945 Liquidity All 47,1 40 -7,1 0,017** State banks 29,6 23,4 -6,2 0,282 Private banks 46,9 40,8 -6,1 0,1* Foreign banks 62,3 49,1 -13,2 0,027**

Note: * indicates statistical significance at the 10 (*), 5(**), 1(***) percent level.

Table 7

Significance T-test mean values

Group N p (7y) p (11y) p (7y) p (11y) p (7y) p (11y) p (7y) p (11y) p (7y) p (11y) All 37 n n y** y*** y* n y*** y*** y*** y*** State banks 4 n n y** y** n n y** y** n n Private banks 28 n n y** y** y* n y*** y*** y** y** Foreign banks 5 n n n n n n n n y** y** FX Loans Liquidity Empirical Results Capital Adequacy Asset Quality FX Liabilities

Table 8

Significance T-test median values

Group N p (7y) p (11y) p (7y) p (11y) p (7y) p (11y) p (7y) p (11y) p (7y) p (11y) All 37 n n y** y*** y** n y*** y* y*** y** State banks 4 n n y** y** n n n n n n Private banks 28 n n y** y* n n y*** y** y** y* Foreign banks 5 n n y** n n n n n y** y**

Empirical Results Capital Adequacy Asset Quality FX Liabilities FX Loans Liquidity

8 Summary and Conclusions

Turkey, after a decade of being hit by some serious banking crises, introduced in 1999 new regulations to strengthen the banking sector. One of the existing problems was the high level of moral hazard behaviour among banks. This was said to come from information asymmetries and the full deposit insurance. Understanding the risk of these two the government implemented regulations that forced the banks to disclose more information and limited the deposit insurance. The main aim of this research was to provide a link between the effects of these regulations on the level of moral hazard taken by deposit banks in Turkey. This paper investigated whether the regulatory reforms, which have been implemented in Turkey in 1999, have resulted in the reduction of moral hazard in the deposit banking sector. The outcomes of the analysis have shown that the evidence does not support the hypothesis that the regulatory reforms have contributed to reducing moral hazard in the deposit banking sector is. Instead, it might even be argued that the changes led to even more moral hazard.

banks. In the research, t-test statistics could only confirm the hypothesis that the deposit banks in Turkey have become less subject to moral hazard due to the implemented regulatory changes in 1999 regarding exposure on foreign loans. These findings do match with the conclusion of previous studies. Mckinnon and Pill (1999) showed, for example, that the by removing deposit insurance system the incentives to banks to increase their foreign borrowing are also removed. Consistent with this we found that the banks became significantly less exposed to foreign exchange exposure regarding their foreign loans. However, there was hardly any significant evidence that the exposure in foreign liabilities changed. Other theoretical and empirical studies conclude that information asymmetries in the banking sector as well as deposit insurance systems lead to moral hazard problem resulting in excessive risk taking by deposit banks which could result in high ratio of nonperforming loans (Barth et al., 2004; 2007; Boyd et al., 1998; Demirgüc-Kunt and Detragiache, 2002). This is absolutely not consistent with our findings. Our findings show that the ratio of non-performing loans only increased. The reason for this is probably because the changes in regulations forced the banks to report there NPL more accurately. In addition, McKinnon (1999) shows in his study that sound banks without moral hazard would limit the risk of maturity- and foreign exposure. Our findings do not support these outcomes. Our findings show that banks became significantly more subject to liquidity problems after the regulations were implemented. This is probably due to several bank runs in the years after 1999, which resulted in less liquidity in the banking sector. Finally, in line with previous studies we predicted banks’ capital adequacy being positively influenced by the implemented regulatory changes. The evidence showed a positive relation, however the difference is not significant, which means that this hypothesis is fully rejected.

hazard. Altogether, this could mean that by decreasing the deposit insurance, banks became more prone to bank runs. In the case of Turkey this means that instead of decreasing the moral hazard problem, the imposed changes resulted in more moral hazard due to the lack of supervisory and regulatory support by the BRSA. The agency might not yet have the ability to support adequately because it has only just been established and it might have lacked the experience and strength at that moment to enforce properly. As a result in 2000 and 2001 some bank runs occurred putting the Turkish banking sector again in a banking crisis.

However, some limitations are in place. By not conducting a multivariate analysis one should be rather conservative in extracting any conclusions, since the cause might be in an additional, not yet analysed field. We limited this problem by conducting two sorts of t-tests, a mean test and a median test. The results of these two do support each other for a great deal, thereby strengthening our final conclusion. In this study we also made a distinction in banking groups. The results give us the idea that in line with the literature private banks are easier affected by regulatory changes. But due to small sample size of foreign and state-owned banks it is difficult to put a lot of weight on our findings. However, for the banking sector as a whole, the findings are stronger.

References

Akerlof, G. (1970) “The Market for Lemons: Quality uncertainty and the Market Mechanism”, The Quarterly Journal of Economics, Vol. 84, No. 3, pp. 488-500

Akerlof, G. and Romer, P. (1994) “Looting: the economic underworld of bankruptcy for profit,” NBER Working Paper, No. R1869

Ali, S. (2003) “Financial distress and bank failure: Lessons from closure of Ihlas Finans in Turkey”, Islamic Research and Training Institute, Islamic Development Bank, Jeddah

Alper, C. and Onis, Z. (2002) “Emerging market crises and the IMF: rethinking the role of the IMF in the light of Turkey’s 2000-2001 financial crises”, METU International Conference VI September, Ankara

Antinolfi, G., Huybens, E. & Keister, T. (2001) “Monetary stability and liquidity crises: the role of the lender of last resort”, Journal of Economic Theory, Vol. 99, No 1-2, pp. 187-248

Barth, J., Caprio, G. & Levine, R. (2004) “Bank regulation and Supervision: What works best?”, Journal of Financial Intermediation, Vol. 13, No 2, pp. 205-248

Barth, J., Caprio, G. & Levine, R. (2007) “Bank regulations are changing: But for better or worse?”, Worldbank, Working Paper

Basel Committee on banking Supervision (2000) “Principles for the Management of Credit Risk”, Consultative Paper issued by the Risk management Group of Basel Committee of Banking Supervision, Basel

Bell, J. and Pain, D. (2000) Leading indicator model of banking crises –a critical review”, Financial Stability review, Vol. 9, pp. 113-129

Bernanke, B., Laubach, L., Mishkin, F., & Posen, S. (1999) “Inflation targeting: Lessons from the international experience”, Princeton, NJ: Princeton University Press.

Chiyachantana, C., Jiang, C., Taechapiroontong, N. & Wood, R. (2004) “The impact of regulation fair disclosure on information asymmetry and trading: an intraday analysis”, The Financial Review, Vol. 39, No 4, pp. 549-577

Corsetti, G., Pesenti, P. & Roubini, N. (1998) “Paper tigers? A model of the Asian crisis”, NBER, Working Paper No 6783

Dages, B., Goldberg, K. & Kinney, D. (2000) “Foreign and Domestic bank participation in Emerging Markets: Lessons from Mexico and Argentina”, Economic Policy Review, Vol. 9, pp. 17-36

Demirgüc-Kunt, A. and Detragiache, E. (1998) “The determinants of banking crises in developing and developed countries,” IMF Working Paper, Vol. 45, No 1, pp. 81-109

Demirgüc-Kunt, A. and Detragiache, E. (2002) “Does deposit insurance increase banking system stability? An empirical investigation”, Journal of Monetary Economics, Vol. 49, No 7, pp. 1373-1406

Demirgüc-Kunt, A. and Kane, E. (2000) “Deposit insurance around the Globe: where does it work?” Journal of Economic Perspectives, Vol. 16, No 2, pp. 175-195

Detragiache, E. and Gupta, P. (2004) “Foreign banks in emerging market crises: Evidence from Malaysia”, IMF Working Paper, WP/04/129

Diamond, D. (2007) “Banks and Liquidity creation: a simple exposition of the Diamond-Dybvig model”, Economic Quarterly, Vol. 93, No. 2, pp. 189-200

Dinc, S. (2003) “Banks and Politicians: Political Influences on State-owned Banks in Emerging Countries”, Journal of Financial Economics, Vol. 77, No 2, pp. 453-479

Estrella, A., Park, S. & Peristiani, S. (2000) “Capital ratios as predictors of bank failure, Economic Policy Review, Vol. 6, No 2, pp. 34-52

Gavin, M. and Hausmann, R. (1996) “The roots of banking crises: the macroeconomic context”, Banking crises in Latin America, John Hopkins University Press, pp. 27-63

Goldstein, M. and Turner, P. (1996) “Banking crises in emerging economies: origins and policy options”, BIS Economic Papers, No. 46

Hellman, T., Murdock, K. & Stiglitz, J. (1997) “Liberalization, Moral Hazard in Banking, and Prudential Regulation: Are Capital Requirements Enough?”, Stanford Graduate School of Business, Working Paper

Humphrey, D. (1976) “100% Deposit insurance: What would it cost?” Journal of Bank Research, Vol. 7, No 3, pp. 192-198

Kibritcioglu, A. (2005) “Banking sector crises and related new regulations in Turkey”, Economia Exterior, Vol. 32, No 1, pp 141-148

Knutsen, S. (2001) “Financial fragility or information asymmetry? The inter-war banking crisis in Norway”, EBHA Conference 2001: Business and Knowledge, Norwegian School of Management

Krugman, P. (1999) “Balance Sheets, the transfer problem, and financial crises”, Festschrift Volume, Working Paper

Mian, A. (2003) “Foreign, Private Domestic, and Government Banks: New evidence from Emerging Markets”, Graduate School of Business, Working Paper, University of Chicago

Minsky, H. (1986) “Stabilizing an unstable economy”, New Haven, Connecticut, Yale University Press

Mishkin, F. (1997) “The causes and propagation of financial instability: lessons for policy makers”, Proceedings, Federal Reserve Bank of Kansas City, pp. 55-96,

Mitchell, W. (1909) “The Decline in the Ratio of Banking Capital to Liabilities” The Quarterly Journal of Economics, Vol. 23, No 4, pp. 697-713

McKinnon, R. and Pill, H. (1999) “Exchange rate regimes for emerging markets: moral hazard and international over borrowing,” Oxford Review of Economic Policy, Vol. 15, No 3, pp. 19-38

McKinnon, R. (2002) “Limiting moral hazard and reducing risk in international capital flows: the choice of an exchange rate regime”, The Annals of the American Academy of Political and Social Science, Vol. 579, No 1, pp. 200-207

Müslümov, A. (2005) “Full deposit insurance and the moral hazard problem: the case of the Turkish Banking System”, Bogazici Journal, Review of Social, Economic and Administrative Studies, Vol. 19, No 1-2, pp. 1-22

Özatay, F. and Sak, G. (2002) “Financial Liberalization in Turkey: Why was the impact on growth limited?”, Emerging Markets Finance and Trade, Vol. 38, No 5, pp. 6-22

Rochet, J. (2008) “Why are there so many banking crises?”, Princeton University Press, New Jersey, USA

Steinherr, A., Tukel, K. & Ucer, M. (2004) “The Turkish banking sector: challenges and outlook in transition to EU membership”, Centre for European Policy Studies, Working Paper, No 4

Togan, S. (2004) “Turkey: Toward EU Accession”, The World Economy, Vol. 27, No. 7, pp. 1013-1045

World Bank (2000) Turkey: Country Economic Memorandum -Structural Reforms for Sustainable Growth”, Report No. 20657-TU

Yeldan, E. (2002) “Behind the 2000/2001 Turkish Crisis: Stability, Credibility, and Governance, for Whom?” Bilkent University, Working Paper

Websites:

www.nationsencyclopedia.com www.worldbank.org

www.imf.org

www.tcmb.gov.tr (Central Bank of Turkey)

Appendix I

MAIN GOALS OF BRSA

To enhance banking sector efficiency and competitiveness- elimination of distortions created by the state banks; strengthening of the banks' capital base; reduction of the banks' intermediation costs; minimization of group banking and non-financial activities. To maintain confidence in the banking sector - in accordance with market discipline and "self responsibility" principle, to design the proper regulation for public awareness; making adequate, understandable and accurate information accessible to the markets in a timely manner; promoting international best standards in accounting and reporting systems; providing a transparent environment in which information on risks is clear and accessible for all parties.

To minimize the potential risks to the economy from the banking sector - prevention of all kinds of transactions and practices that can jeopardize the smooth and safe operation of the banks; developing early warning and prompt correction systems to prevent individual problems from causing systemic risk.

To enhance the soundness of the banking sector - enhancing the flexibility of the sector against risks; giving importance to the improvement of corporate governance;

developing internal control and risk management systems; taking market risk into account in calculation of capital adequacy; improving the BRSA's capacity for risk-focused and consolidated supervision and control.

Appendix II

List of banks used for the research of this paper

Privately-owned Banks State-owned Banks

Adabank A.Ş. Türkiye Cumhuriyeti Ziraat Bankası A.Ş. Akbank T.A.Ş. Türkiye Emlak Bankası A.Ş.

Alternatif Bank A.Ş. Türkiye Halk Bankası A.Ş. Anadolubank A.Ş. Türkiye Vakıflar Bankası T.A.O. Bayındırbank A.Ş.

Birleşik Türk Körfez Bankası A.Ş. Foreign Bank Founded in Turkey Denizbank A.Ş. Arap Türk Bankası A.Ş.

Ege Giyim Sanayicileri Bankası A.Ş. Citibank A.Ş. Fiba Bank A.Ş. Denizbank A.Ş. Finans Bank A.Ş. Deutsche Bank A.Ş. İktisat Bankası T.A.Ş. Osmanlı Bankası A.Ş. Kentbank A.Ş.

Koçbank A.Ş.

Milli Aydın Bankası T.A.Ş. MNG Bank A.Ş.

Oyak Bank A.Ş. Pamukbank T.A.Ş. Sitebank A.Ş. Şekerbank T.A.Ş. Tekstil Bankası A.Ş. Toprakbank A.Ş. Turkish Bank A.Ş.

Appendix III

Overview of “time-lines” of the financial ratios

Foreign Exchange Position Liabilities (%)

0,0 10,0 20,0 30,0 40,0 50,0 60,0 70,0 80,0 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994

State owned banks Private banks Foreign banks All

Liquidity (%) 0,0 10,0 20,0 30,0 40,0 50,0 60,0 70,0 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994

State owned banks Private banks Foreign banks All

Foreign Exchange Position Loans (%)

0,0 10,0 20,0 30,0 40,0 50,0 60,0 70,0 80,0 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994

Appendix III (Continued)

Standard Capital Ratio (%)

0,0 10,0 20,0 30,0 40,0 50,0 60,0 70,0 80,0 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994

State owned banks Private banks Foreign banks All

Credit Risk (%) 0,0 10,0 20,0 30,0 40,0 50,0 60,0 70,0 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994