流動性創造是否會導致更多的流動性風險?

全文

(1)國立高雄大學經營管理研究所碩士論文. Will Liquidity Creation Cause Higher Liquidity Risk? 流動性創造是否會導致更多的流動性風險？. 研究生：鍾欣潔撰指導教授：陳怡凱. 中華民國九十九年六月.

(2) 致謝詞時光荏苒，轉眼又到了離別的時刻了！研究所兩年的時光真是酸甜交雜，如今回首真的覺得精采萬分。其中最感謝我的指導老師. 陳怡凱博士在這對時間對. 我細心的指導，是我在學海迷途的一盞明燈指引我學習的方向！每次的伊媚兒回復得如此神速，讓我感的是悲喜交加。而每次咪聽對我來都是挖寶的過程，老師很貼心的給大家的都是一把鏟子……感謝口委王澤世老師、李揚所長給的建議讓本篇文章能夠更臻完善。很感謝研究所的同學和我一同打拚、一同享樂，其中佳慧更是不能沒有妳！澎湖、美食、研討會以及世足賽都感謝妳的熱情參與！另外很感謝的就是網路的發達，這讓我能和亞儒一起在 MSN 上鬼吼鬼叫之外，也能在 Plurk 上與我的農友團朋友大家交流分享生活中的點滴。最終我要感謝的是我的爸媽與我的家人，感謝他們無止盡的付出與再三的包容我的任性，讓我無憂的一路念到研究所結束。謹以此論文獻給我親愛的家人及師長們。. 鍾欣潔謹誌於高雄大學經營管理研究所中華民國九十九年.

(3) Will Liquidity Creation Cause Higher Liquidity Risk? Advisor：Dr. Yi-Kai Chen Department of Finance, National Universityof Kaohsiung. Student: Hsin-Chieh Chung Institute of Business and Management, National University of Kaohsiung. Abstract Generally liquidity risk is an existence in banks for different degree. However, liquidity risk had not been considered as an important risk until the subprime mortgage liquidity risk. Many large banks bankrupt for liquidity risk, thus people take more serious in manage liquidity risk. The liquidity creation is one of traditional functions of the bank. Berger & Bouwman (2009) proposed liquidity creation is positively to bank's value. The issue is deserved to discuss whether the relationship between liquidity risk and liquidity creation exists or not. We take financial system as dummy variable, using two stage least squares model to estimate bank liquidity risk and liquidity creation. The results show that the relationship between liquidity risk and liquidity creation is negative except to in the market system and when the financial crisis occurred.. Keywords:：Liquidity Risk, Liquidity Creation, Financial System. I.

(4) 流動性創造是否會導致更多的流動性風險？指導教授：陳怡凱博士國立高雄大學金融管理學系學生：鍾欣潔國立高雄大學經營管理研究所. 摘要對於銀行而言，流動性風險是一個必然的存在。許多大銀行在次級房貸中倒閉，其中一個主要原因是流動性風險而倒閉，這才引起世人對於流動性風險的重視與管理。而流動性風險和流動性創造都由相對不流動的資產去支應相對流動的負債所產生，流動性創造是銀行的傳統功能之一。Berger & Bouwman (2009)提出了流動性創造與銀行價值呈正相關，探討流動性風險與流動性創造是否相關是一值得探討的議題。我們加入金融體系與金融危機做為虛擬變數，使用兩階段最小平方法來探討銀行的流動性風險與流動性創造之關係。結果流動性創造與流動性風險呈負相關而非正相關。而在加入金融體系後發現，在市場體系的銀行會有比較敏感的關係。. 關鍵詞：流動性風險、流動性創造、金融體系. II.

(5) Table of Contents Abstract ................................................................................................................................I 摘要 ................................................................................................................................... II Table of Contents .............................................................................................................. III List of Tables ...................................................................................................................... V Chapter 1 Introduction ....................................................................................................... 1 1.1 Research Background ............................................................................................ 1 1.2 Research Purpose ................................................................................................... 2 Chapter 2 Literature Review .............................................................................................. 4 2.1 Liquidity Risk ........................................................................................................ 4 2.2 Measurement of Liquidity Risk ............................................................................. 4 2.2.1 Liquidity Ratios ................................................................................................... 5 2.2.2 Sources and Uses of Liquidity ............................................................................. 5 2.2.3 Maturity Ladder/Scenario Analysis ................................................................... 6 2.2.4 Peer Group Ratio Comparisons .......................................................................... 6 2.2.5 Liquidity Index .................................................................................................... 6 2.2.6 Liquidity Planning .............................................................................................. 7 2.2.7 Financing Gap and the Financing Requirement ................................................ 8 2.3 Liquidity Creation ................................................................................................. 8 2.4 Measurement of Liquidity Creation ...................................................................... 8 2.4.1 Liquidity Transformation Gap ........................................................................... 9 2.4.2 Liquidity Creation............................................................................................... 9 Chapter 3 Data and Methodology..................................................................................... 11 3.1 Data Description .................................................................................................. 11 3.2 Variables .............................................................................................................. 11 3.2.1 Liquidity Risk.................................................................................................... 11. III.

(6) 3.2.2 Liquidity Creation............................................................................................. 14 3.2.3 Dummy Variables ............................................................................................. 17 3.3 Determinants of Liquidity Creation .................................................................... 18 3.4 Bank Liquidity Risk and Liquidity Creation Model ........................................... 18 3.5 Panel Unit Root Test and Sargan Test ................................................................ 20 Chapter 4 Empirical Results ............................................................................................. 22 4.1 Panel Unit Root Tests and Sargan Test Results .................................................. 22 4.2 Regression Results ............................................................................................... 22 Chapter 5 Conclusions ...................................................................................................... 24 Reference ........................................................................................................................... 26 Appendix: Survey Questions of Bank Regulation and Supervision ................................. 40. IV.

(7) List of Tables Table 1 Variable Description ................................................................................. 28 Table 2 Liquidity Activities Classification with Bankscope Database Restrictions 29 Table 3 Classification of Financial Systems of Sample Countries.......................... 30 Table 4 Panel Unit Root Test Results Using Unbalanced Panel Data ..................... 30 Table 5 Sargan Test Results of All Samples........................................................... 31 Table 6 Sargan Test Results in Different Conditions .............................................. 31 Table 7 Descriptive Statistics of All Samples ........................................................ 32 Table 8 Descriptive Statistics in Market-Based Financial System .......................... 32 Table 9 Descriptive Statistics in Bank-Based Financial System with Taking Liquidity Creation as Different Variable ................................................................... 33 Table 10 The Regression Result of All Samples .................................................... 34 Table 11 The Results In Different Financial Systems ............................................ 35 Table 12 The Results of the Subsample in Market-Based Financial System .......... 36 Table 13 The Results of the Subsample in Bank-Based Financial System ............. 37 Table 14 The Regression Result in Different Financial Systems not during Financial Crisis ........................................................................................................ 38 Table 15 The Regression Result in Different Financial Systems during Financial Crisis ........................................................................................................ 39. V.

(8) Chapter 1 Introduction 1.1 Research Background Liquidity risk had been overlooked for a long time. The earlier financial institutions’ problems are focus on credit risk, market risk and operational risk. Basel I Accord (Basel committee on Banking Supervision, 1988) takes credit risk and market risk in to account on measuring banks’ risk. And Basel II Accord (Basel committee on Banking Supervision, 2004) takes operational risk into account. Liquidity risk had not received takes seriously until subprime outbreak. One of the major reasons for banks’ bankruptcy is liquidity risk rather than other problems. As a result, the literatures about liquidity risk boomed after this crisis. Generally speaking, liquidity risk arises in two dimensions. One is from the assets, and the other is from the liabilities. Banks finance risky loans with riskless deposits, and the mismatch for liabilities and assets causes banks’ financial fragility (Diamond and Rajan, 2001). The financial fragility causes banks face the liquidity risk. Banks need to manage their liabilities are proportioned to their assets in case of highly discount rate immediately. When the liquidity risk happens, it means the mismatch for liabilities and assets. At the meanwhile, it creates liquidity for banks. One of bank traditional roles is liquidity creator. When one bank transfers illiquid assets to liquid liabilities, it creates positive liquidity creation (Berger and Bouwman, 2009). They said banks’ value is positive linked with liquidity creation, finding the banks have higher liquidity creation would with significantly higher market-to-book and price-earning ratio. But they didn’t describe the precise details. Banks earn profits by liquidity creation. When liquidity risk is under control, 1.

(9) banks make their best efforts in liquidity creation. Once liquidity risk is out of control, it usually cause in serious result. Those banks which overlooked liquidity risk went bankrupt during the financial subprime crisis. We want to know whether the liquidity creation make effect on liquidity risk or not. Higher liquidity creation may mean the bank takes more risk. The same degree of the liquidity creation may come with vary degree of liquidity risk, because banks finance with different kinds of liquidity activities. However, there is no paper to explore the relation between liquidity creation and liquidity risk in Taiwan. And the liquidity creator role of banks is less papers to discuss, the relationship might be got more attention later.. 1.2 Research Purpose The concept of liquidity creation could trace back from Deep and Schaefer (2004). They proposed a formula to count the gap between liquid liabilities and liquid assets to measure the banks’ liquidity as the liquidity transformer role. It contained the asymmetry maturity of funds between providers and borrowers, and banks generate economic value. Simultaneously, this asymmetry maturity may come with liquidity risk. However, they don’t take the off balance sheet items and the properties of assets and liabilities into account. In this paper we compute the liquidity creation followed Berger and Bouwman (2009). They take the off balance sheet items and the properties of assets and liabilities into account. Considering the bank-specific and macroeconomic factors, we hope to find out it would exist a negative relationship between liquidity risk and liquidity creation. And we want to know whether liquidity creation affect liquidity risk or not when bank create liquidity. Furthermore, during fourteen years there are several financial crises, 2.

(10) we take the financial crises into account to see whether the relation between liquidity risk and liquidity creation would change or not. We apply panel data instrumental variables regression, using two stage least squares (2SLS) estimators to estimate the determinants of bank liquidity risk model. And we use financial systems and financial crisis as sample respectively. Dependent variable is financing gap ratio (FGAPR) to measure the liquidity risk, and main explanatory variable is the liquidity creation divided by total equity (LC). The result is liquidity creation is negative to liquidity risk in both of market-based and bank-based financial system. In bank-based financial system, the negatively relationship between liquidity risk and liquidity creation is higher than these relationship in market-based system. And which financial system is better is not confirmed in academic. This result may be due to the situations of banks in different financial system would have different banking developments. For the natures of market-based financial system, banks take more liquidity risk due to they would face higher funding risk. But there is not an absolutely relationship between banks and financial systems, liquidity creation would not absolutely cause liquidity risk. The results show that liquidity creation is negatively relationship to liquidity risk. Banks do more liquidity creation, and they face less liquidity risk. It implies banks act more traditional role, and they face less liquidity risk.. 3.

(11) Chapter 2 Literature Review 2.1 Liquidity Risk Liquidity risk is typically defined as the ability for institutions to replace liability runoff and fund asset growth promptly and at a reasonable price. Landskroner and Paroush (2008) found an obvious trade-off relationship between liquidity and return. Banks earn profits by switching deposits to loans, thus liquidity risk exists with asymmetric maturities. Liquidity risk is the possibility of the insolvency, and it exists in commercial banks which receives deposits. Decker (2000) indicated that the sources of liquidity risk are bank-specific problems and macroeconomic problems. Bank-specific problems are banks cannot liquidate their assets for the quality of assets or they cannot obtain adequate funding from depositors. Macroeconomic problems are that banks may sell their assets in very low prices for the thinly traded securities markets or market disruptions.. 2.2 Measurement of Liquidity Risk Basel Committee on Banking Supervision (2000) proposed maturity laddering method for measuring liquidity risk. Saunders and Cornett (2008) indicated that banks can use sources and uses of liquidity, peer group ratio comparisons, liquidity index, financing gap and the financing requirement, and liquidity planning to measure their liquidity exposure. Here we brief discuss each liquidity risk measurements below:. 4.

(12) 2.2.1 Liquidity Ratios In the past, better practices for liquidity risk measures focused on the use of liquidity ratios. The ratios previous studies used include liquid assets to total assets ratio (e.g. Bourke, 1989; Molyneux and Thornton, 1992; Barth et al., 2003; Demirgüç-Kunt et al., 2003), liquid assets to deposits ratio (Shen et al., 2001),liquid assets to customer and short term funding (Kosmidou et al., 2005), loans to total assets ratio (e.g. Demirgüç-Kunt and Huizinga, 1999; Athanasoglou et al., 2006), net loans to customer and short term funding ratio (e.g. Pasiouras and Kosmidou, 2007; Naceur and Kandil, 2009) to assess bank’s liquidity risk. The higher the value of these ratios, the more liquidity risk the banks will suffer.. 2.2.2 Sources and Uses of Liquidity Saunders and Cornett (2006) indicated that the net liquidity statement that lists sources and uses of liquidity is a useful tool to management liquidity risk. It provides measure of banks net liquidity position. It tells that banks can obtain liquid funds in three ways. First, they can sell their liquid assets with less price risk and low transaction cost. Second, they can borrow funds in the money or purchased funds market up to a maximum amount. Third, they can use any excess cash reserves over and above the amount held to meet regulatory imposed reserve requirements. Bank’s managers can deal with liquidity risk through historical sources and use of liquidity statements can assist the managers in potential liquidity problems.. 5.

(13) 2.2.3 Maturity Ladder/Scenario Analysis The Bank for international Settlements (BIS) outlines a Maturity Ladder for measurement liquidity risk. It computes the net cash flow to understand in a period – it can be one day or one month or even longer. It can be illustrated under the cumulative net excess or shortage of funds. There are three major conditions in the graph, it simulates cumulative funding gap in normal condition, general market crisis and individual specific crisis.. 2.2.4 Peer Group Ratio Comparisons Saunders and Cornett (2008) mentioned one financial institutions can compare crucial ratios with the similar size and location financial institutes. It could mean they face the similar liquidity problem from the money market or purchased funds market. Therefore, banks can determine their liquidity exposure, such as loans to deposits, borrowed funds to total assets, and commitments to lend to assets ratio. A high ratio of loan commitments to assets indicates the need for a high degree of liquidity to fund any unexpected execution of these loans. Thus, banks with high commitment often face more liquidity risk exposure.. 2.2.5 Liquidity Index Jim Pierce at Federal Reserve developed the liquidity index to measure liquidity risk. It can measure the potential losses a financial institution could suffer as (or fire-sale) disposal of assets. Define an index I such that:. 6.

(14) 𝑁. 𝐼=. 𝜔𝑖 𝑖=1. 𝑃𝑖 𝑃𝑖∗. 1. Where 𝜔𝑖 is the percent of each asset in the financial institute’s portfolio:. 𝑁. 𝜔𝑖 = 1. 2. 𝑖=1. The smaller differences between immediate fire-sale asset price (Pi) and fair market price (𝑃𝑖∗ ), the more liquidity the financial institute has. The liquidity index lies between 0 and 1. The smaller of the fire-sale asset price, the smaller the liquidity index and higher the liquidity risk the bank faces. Besides, the liquidity index for this bank could also be compared with similar indexes calculated for a peer group of similar bank.. 2.2.6 Liquidity Planning Saunders and Cornett (2008) considered deposit institutions can use liquidit y planning to control liquidity risk. It includes four components. The fist component is the delineation of managerial details and responsibilities. The second component is detailed list of fund providers who are most likely to withdraw. The third component is the identification of the size of potential deposit and fund withdrawals over various time horizons in the future. The fourth component is to set internal limits on separate subsidiaries’ and branches’ borrowings. Liquidity planning provides deposit institution managers details to cope with liquidity risk.. 7.

(15) 2.2.7 Financing Gap and the Financing Requirement Saunders and Cornett (2008) construct the financing gap equation to measure liquidity risk. Although demand depositors have the right to withdraw their fund immediately, they do not do so in normal circumstances. So a deposit institution managers use the average liquid deposit to measure their financing gap. The financing gap can be compute for average loans minus average deposits. When financing gap is positive, deposit institution needs to borrowed funds to smooth away this gap. The borrowed funds are also named financing gap in the method.. 2.3 Liquidity Creation Diamond and Dybvig (1983) gave the first explicit analysis of the demand for liquidity and the ―transformation‖ service provided by bank. When banks transform relatively illiquid assets into relative liquid liability, liquidity is created. However, literatures about liquidity creation are less discussed since most of empirical literatures see banks’ role is risk transformers rather than liquidity creators. The phenomenon of liquidity creation by banks appears everywhere.. 2.4 Measurement of Liquidity Creation Owning to banks are seen as the risk transformer before, there is few of liquidity creation for banks in academic studies. Berger and Bouwman (2009) developed comprehensive measures of bank liquidity creation due to people focused on banks’ transformation role rather than liquidity role. We describe two meanly measurement for liquidity creation. 8.

(16) 2.4.1 Liquidity Transformation Gap Deep and Schaefer (2004) construct a measurement equation for liquidity creation. They separate all assets into liquid asset and illiquid asset. And so is liability. With maturity in a year as the standard, they construct an equation as the following to measure liquidity.. 𝐿𝑖𝑞𝑢𝑖𝑑𝑖𝑡𝑦 𝑇𝑟𝑎𝑛𝑠𝑓𝑜𝑟𝑚𝑎𝑡𝑖𝑜𝑛 𝐺𝑎𝑝 =. 𝐿𝑖𝑞𝑢𝑖𝑑 𝐿𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑦 − 𝐿𝑖𝑞𝑢𝑖𝑑 𝐴𝑠𝑠𝑒𝑡 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡. 3. When banks financed in large part by liquid deposits (liquidity liabilities) and hold less liquid loans would have a high LT gap value, and banks have the larger liquidity risk.. 2.4.2 Liquidity Creation Another measure of liquidity creation is developed by Berger and Bouwman (2009). They proposed two major effects of bank liquidity creation are risk absorption effect in large banks and financial fragility-crowding out effect in small banks. Due to subject to more regulatory scrutiny and market discipline than small banks, large banks may be affected their capacity to absorb risk. They thought large banks would control better risk absorption than small banks. And they pointed out large bank generate more liquidity creation. They mentioned less fragile banks would monitor their borrowers and hence extend loans for them. Simultaneously, if the banks are small banks, loaning more would impede banks’ liquidity creations and. 9.

(17) crows out deposits. Thus, the financial fragility-crowding out effect appears in small banks and additional equity capital impede them create liquidity. Berger and Bouwman (2009) separated assets, liabilities and off-balance sheet activities by category. According to liquidity creation theory, banks create liquidity on the balance sheet when they transform illiquid assets into liquid liabilities. It implies that banks create liquidity by thus method can give public liquid items. Under this theory, they applied positive weights to illiquid asset and liquid liabilities. Following similar logic, they applied negative weights to liquid assets, illiquid liabilities, and liquid equity. So that liquidity was destroyed when illiquid liabilities is financed to liquid asset. They took some factors like the ease, cost and time for banks and customers in to account. The measurement including off-balance activities, it incorporates the uncertainty that LT gap doesn’t mention.. 10.

(18) Chapter 3 Data and Methodology 3.1 Data Description In this paper, we compute the liquidity creation by the measurement from Berger and Bouwman (2009). We use the data lying on 1995 to 2008 in this paper with the Bankscope Database. Due to the limitations of attainable details, we adjust our computed items. We choose the G10 countries (Belgium, Canada, France, Germany, Italy, Japan, Netherlands, Sweden, Switzerland, United Kingdom and United States) in World Outlook Database (IMF). And we follow Barth, Caprio, and Levine (2004) to measure regulatory and supervisory variables.. 3.2 Variables All the used variables are listed and explained in Table 1. We take liquidity creation as endogenous variables for that some variables of liquidity creation would affect liquidity risk. We set two stage least squares to know the relationship between liquidity risk and liquidity creation.. 3.2.1 Liquidity Risk Saunders and Cornrett (2008) defined the financing gap as the difference between average loans and average core deposits. They thought the average deposits as the minus item because the average deposits are stable. Average deposits is the core source for institutions to do loans. We can calculate the average deposits by customer. 11.

(19) deposit. As a result, the financing gap can be written as following:. 𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 𝑔𝑎𝑝 = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑙𝑜𝑎𝑛𝑠 − 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑐𝑜𝑟𝑒 𝑑𝑒𝑝𝑜𝑠𝑖𝑡𝑠. 4. The banks have to fund the positive financing gap by its liquid asset and/or borrowing funds in the money market. Thus the equation can see as:. 𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 𝑔𝑎𝑝 = −𝐿𝑖𝑞𝑢𝑖𝑑 𝐴𝑠𝑠𝑒𝑡𝑠 + 𝐵𝑜𝑟𝑟𝑜𝑤𝑒𝑑 𝐹𝑢𝑛𝑑𝑠. 5. We can write relationship as:. 𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 𝑔𝑎𝑝 + 𝐿𝑖𝑞𝑢𝑖𝑑 𝐴𝑠𝑠𝑒𝑡𝑠 = 𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑚𝑒𝑛𝑡𝑠 𝐵𝑜𝑟𝑟𝑜𝑤𝑒𝑑 𝐹𝑢𝑛𝑑𝑠. 6. In this method, financing gap uses the difference between average loans and average deposits. In order to standardize for all banks, we divide financing gap to total assets.. 𝐹𝐺𝐴𝑃𝑅 =. 𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 𝑔𝑎𝑝 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑙𝑜𝑎𝑛𝑠 − 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑐𝑜𝑟𝑒 𝑑𝑒𝑝𝑜𝑠𝑖𝑠 = 𝑡𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠 𝑡𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠. 12. 7.

(20) We use financing gap ratio (FGAPR) to measure liquidity risk. And the loans are classified by maturity in Saunders and Cornett (2008). We additional deduct cash to account more exactly pure liquidity risk. And we adjust the items for the limitation by database. 1 Individual banks causes of liquidity risk include bank size, less risky liquid assets and bank capital. We use natural logarithm of bank’s gross total assets (LNGTA) to measure bank size.2 For large banks may benefit from implicit guarantees, they acquire lowly funding cost and them could invest in riskier assets (Iannotta et al., 2007). Banks can sell or collateralize its liquid assets to obtain liquid funds, thus they can get away from bank’s liquidity risk. So we expect that large banks usually hold more loans and thus have larger financing gap ratio and higher. But banks may are confronted to credit freeze, they are difficult to sell or collateralize their liquid assets. So we compute the less risky liquid assets, and we divided less risky assets by gross total assets separately to standardize, finally get less risky assets to gross total assets ratio (LRA). Banks can sell their less risky liquid assets with reasonable prices in short term. Macro economy causes of liquidity risk include economic growth and financial restrictions. Aspachs, Nier and Tiesset (2005) indicated that bank hoard liquidity during periods of economic downturn, when lending opportunities may not be as good and they run down liquidity buffers during economic expansions when lending opportunities may have picked up. Thus we expect that higher economic growth make banks run down their liquidity buffer and induce banks to lend more. However, banks. 1. 2. Under the limitations from database, we proxy the core deposits by the total of deposit-saving and deposit-demand. I In general, people used to proxy bank’s size by natural logarithm of bank’s total assets rather than natural logarithm of bank’s gross total assets rather. Gross total assets are total assets plus loan loss reserve in this paper, and we use gross total assets for consisting with liquidity creation accounts. 13.

(21) will attract less deposit during economic expansions, consequently increasing their financing gap. We use annual percent change of GDP (GDPCt) to capture the effect of the macroeconomic environment. Economic growth may encourage banks to unwind their limitations for loans, as to create their liquidity while they may face higher liquidity risk. Financial crisis has been avoided by limiting the ability of financial institutions to do some risky activities. These limitations make banks to adjust their activities to pursue profit. Such regulation and supervision on banks pressure on bank managers responding to increased costs of funding by reducing default risk. (Angkinand and Wihlborg, 2008) We follow Barth, Caprio, and Levine (2004) to measure regulation and supervision for banks. We use official supervisory power index (OSP), private monitoring index (PMI), and overall bank activities and ownership restrictiveness (BAR) to proxy government regulation and supervisory practices. And we examine the effects of supervisory and regulatory variables with the interaction with annual percent change of GDP (GDCt). It produces new supervisory and regulatory variables, which are interactive with macroeconomic variable, are annual percent change of GDP and official supervisory power index (GDPCt×OSP), interactions between annual percent change of GDP and private monitoring index (GDPC t×PMI), interactions between annual percent change of GDP and overall bank activities and ownership restrictiveness (GDPCt×BAR).. 3.2.2 Liquidity Creation We use the method established by Berger and Bouwman(2009) to measure the liquidity creation. First they classified activities as liquid activities, semiliquid. 14.

(22) activities, or illiquid activities. They classified both activities on balance sheet activities (like assets, liabilities plus equity) and off-balance sheet activities (like guarantees and derivatives). And they assigned weights to the classified activities. And the detail categories are shown on the Table 2. They based on liquidity theory that banks transform illiquid assets in to liquid liabilities. As a result, they applied positive weights to illiquid assets, liquid liabilities and illiquid guarantees. Following similar logic, they applied negative weights to liquid assets, illiquid liabilities plus equity, liquid guarantees and liquid derivatives. They applied the intermediate weights of 0 to semiliquid assets, semiliquid liabilities, and semiliquid guarantees. These items are considered fall halfway between liquid and illiquid activities. When banks transform $1 of illiquid assets into $1 of liquid liabilities, they assign a weight of. 1 2. to both items for simple dollar-for-dollar adding up constraints.. At the same constrains, they assign a weight of. 1 2. for illiquid asset, liquid liabilities 1. and illiquid guarantee. And they assign a weight of -2 for liquid asset, illiquid liabilities and liquid guarantee. semiliquid assets, semiliquid liabilities and semiliquid guarantees are assigned a weight of 0. The liquidity creation can white as following:. 15.

(23) 𝐿𝑖𝑞𝑢𝑖𝑑𝑖𝑡𝑦 𝐶𝑟𝑒𝑎𝑡𝑖𝑜𝑛 =. 1 1 ∗ 𝑖𝑙𝑙𝑖𝑞𝑢𝑖𝑑 𝑎𝑠𝑠𝑒𝑡𝑠 + 0 ∗ 𝑠𝑒𝑚𝑖𝑙𝑖𝑞𝑢𝑖𝑑 𝑎𝑠𝑠𝑒𝑡𝑠 − ∗ 𝑙𝑖𝑞𝑢𝑖𝑑 𝑎𝑠𝑠𝑒𝑡𝑠 2 2. 1 + ∗ 𝑙𝑖𝑞𝑢𝑖𝑑 𝐿𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 + 0 ∗ 𝑠𝑒𝑚𝑖𝑙𝑖𝑞𝑢𝑖𝑑 𝑙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 𝑎𝑛𝑑 𝑒𝑞𝑢𝑖𝑡𝑦 2 1 1 − ∗ 𝑖𝑙𝑙𝑖𝑞𝑢𝑖𝑑 𝑙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 + ∗ 𝑖𝑙𝑙𝑖𝑞𝑢𝑖𝑑 𝑔𝑢𝑎𝑟𝑎𝑛𝑡𝑒𝑒𝑠 + 0 2 2 1 1 ∗ 𝑠𝑒𝑚𝑖𝑙𝑖𝑞𝑢𝑖𝑑 𝑔𝑢𝑎𝑟𝑎𝑛𝑡𝑒𝑒𝑠 − ∗ 𝑙𝑖𝑞𝑢𝑖𝑑 𝑔𝑢𝑎𝑟𝑎𝑛𝑡𝑒𝑒𝑠 − 2 2 ∗ 𝑙𝑖𝑞𝑢𝑖𝑑 𝑑𝑒𝑟𝑖𝑣𝑎𝑡𝑖𝑣𝑒𝑠. 8. For the limitation form the database, we adjust the detail activities in the Table 2. The adjusted the liquidity creation equation can be written as like the following:. 𝐿𝑖𝑞𝑢𝑖𝑑𝑖𝑡𝑦 𝐶𝑟𝑒𝑎𝑡𝑖𝑜𝑛 =. 1 1 ∗ 𝑖𝑙𝑙𝑖𝑞𝑢𝑖𝑑 𝑎𝑠𝑠𝑒𝑡𝑠 + 0 ∗ 𝑠𝑒𝑚𝑖𝑙𝑖𝑞𝑢𝑖𝑑 𝑎𝑠𝑠𝑒𝑡𝑠 − ∗ 𝑙𝑖𝑞𝑢𝑖𝑑 𝑎𝑠𝑠𝑒𝑡𝑠 2 2. 1 + ∗ 𝑙𝑖𝑞𝑢𝑖𝑑 𝐿𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 + 0 ∗ 𝑠𝑒𝑚𝑖𝑙𝑖𝑞𝑢𝑖𝑑 𝑙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 2 1 1 − ∗ 𝑖𝑙𝑙𝑖𝑞𝑢𝑖𝑑 𝑙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 + 0 ∗ 𝑠𝑒𝑚𝑖𝑖𝑙𝑙𝑖𝑞𝑢𝑖𝑑 𝑔𝑢𝑎𝑟𝑎𝑛𝑡𝑒𝑒𝑠 + 2 2 ∗ 𝑖𝑙𝑙𝑖𝑞𝑢𝑖𝑑 𝑓𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝑔𝑢𝑎𝑟𝑎𝑛𝑡𝑒𝑒𝑠. 9. In our paper, we take the variables appear in liquidity creation as our endogenous variables. 3. 3. The endogenous variable is gross total asset (GTA), equity to gross total asset (EGTA), and loss reserve to gross loans (LLRGL). We follow Berger and Bouwman (2009) to define GTA equals total assets plus the allowance for loan and lease losses and the allocated transfer risk reserve. For limitation, we adjust the GTA equals to total assets plus loan loss reserve. We use LLRGL to proxy credit risk. 16.

(24) Financial crises sometimes occur in the inappropriate treatment for internal supervisory practices or external government regulations. Gualandri, Landi, and Venturelli (2009) mentioned that due to financial innovation has allowed credit risk to be transferred to financial investors by means in financial market, the nexuses between credit, market, and liquidity risk have been closer. As a result, the supervisions and regulations for the financial intermediations bring about considerably effect on controlling liquidity risk. And Berger and Bouwman (2005) also said supervision and regulation should be used to protect banks from failure when some potential policies may injure banks. In our study, we take the macro factors into account. We examine the effects of supervisory and regulatory variables with the interaction with annual percent change of GDP (GDPCt). We take liquidity creation as exogenous and endogenous variables in different financial system after endogenous tests.. 3.2.3 Dummy Variables In the paper, we use two dummy variables. One is to split the financial system to bank-based and market-based financial system, and the other is whether financial crisis is breakout or not in past two years. In different financial system, loans’ borrowers may take different channel to assess funds. And we wonder to know whether the liquidity risk and liquidity creation change or not in financial crisis breakout. We take Asian Financial crisis, Dot-com bubble and Subprime cisis into account, and the crisis would affect the next year. Those years are we defined financial crisis outbreak.. 17.

(25) 3.3 Determinants of Liquidity Creation This model provides an economic analysis of the determinants of liquidity creation. Besides, we divide the causes of liquidity risk into internal and external factors. In order to examine the relationship between liquidity creation and the bank-specific, supervisory and regulatory variables, the panel fixed effect regression model has been developed:. 𝐿𝑖𝑡 = 𝑐𝑖 + 𝜆1 𝐿𝑁𝐺𝑇𝐴𝑖𝑡−1 + 𝜆2 𝐸𝐺𝑇𝐴𝑖𝑡−1 + 𝜆3 𝐿𝐿𝑅𝐺𝐿𝑖𝑡−1 + 𝛿4 𝐺𝐷𝑃𝐶𝑗𝑡 −1 + 𝜀𝑖𝑡. 10. where Lit is liquidity creation of ith bank at time t, with i = 1,…, N, t = 1,…,T. 𝐿𝑁𝐺𝑇𝐴𝑖𝑡−1 , 𝐸𝐺𝑇𝐴𝑖𝑡−1 and 𝐿𝐿𝑅𝐺𝐿𝑖𝑡−1 are bank-specific variables. 𝐺𝐷𝑃𝐶𝑗𝑡 −1 macro-economy variable.. is. j refers to the country in which bank i operates; c is a. constant term; εit is the error term. Equation estimated through fixed effects regression taking each bank’s liquidity creations as the dependent variable. We have been tested with the Hausman test and reject the null hypothesis of random effects is suitable model. Thus, we use fixed effects rather than random effects model.. 3.4 Bank Liquidity Risk and Liquidity Creation Model This model provides an economic analysis of the relationship between bank liquidity risk and liquidity creation. Besides, from the causes of liquidity creation. 18.

(26) model, we can realize that there are several factors may affect bank liquidity creation. In our study, we thus regard liquidity creation as an endogenous determinant of bank liquidity risk, and apply panel data instrumental variables regression to estimate this model. In order to examine the relationship between bank liquidity risk and liquidity creation, the panel instrumental variables regression model has been developed:. 𝐵. 𝐾. 𝑆. 𝜃𝑏 𝑋𝑖𝑡𝑏 +. 𝑅𝑖𝑡 = 𝑐 + 𝛽𝑙 𝐿𝑖𝑡 + 𝑏=1. 𝜔𝑘 𝑋𝑗𝑡𝑘 + 𝑗 =1. 𝑀. 𝜈𝑠 𝑋𝑗𝑡𝑠 + 𝑠=1. 𝜂𝑚 𝑋𝑗𝑡𝑚 + 𝜀𝑖𝑡. 11. 𝑚 =1. the reduced form equation for Lit is. 𝐵. 𝑆. 𝜆𝑏 ∏𝑏𝑖𝑡. 𝐿𝑖𝑡 = 𝑐 + 𝑏. 𝑀. 𝛿𝑠 ∏𝑗𝑡𝑠. + 𝑠=1. 𝜈𝑚 ∏𝑚 𝑗𝑡 + 𝜀𝑖𝑡. +. 12. 𝑚 =1. where Pit is bank liquidity risk of ith bank at time t, with i = 1,…, N, t = 1,…,T. 𝑋𝑖𝑡𝑏 , 𝑋𝑗𝑡𝑘 , 𝑋𝑗𝑡𝑠 and𝑋𝑗𝑡𝑚 are bank-specific, market structure, supervisory and regulatory and macroeconomic variables with b = 1,…, B, k = 1,…, K, s = 1,…, S, m = 1,…, M, respectively. j refers to the country in which bank i operates. Lit is liquidity creation, 𝑠 and regarded as endogenous variable. ∏𝒃𝒊𝒕 , ∏𝒋𝒕 ,and ∏𝒎 𝒋𝒕 are instrumental variables. causing liquidity risk. c is a constant term; εit is the error term.. 𝑅𝑖𝑡 = 𝑐 + 𝛽𝑙 𝐿𝑖𝑡 + 𝜃1 𝐿𝑁𝐺𝑇𝐴𝑖𝑡 + 𝜗2 𝐸𝐺𝑇𝐴𝑖𝑡 + 𝜈1 𝑂𝑆𝑃𝑗𝑡 × 𝐺𝐷𝑃𝐶𝑡 + 𝜈2 𝑃𝑀𝐼𝑗𝑡 × 𝐺𝐷𝑃𝐶𝑡 + 𝜈3 𝐵𝐴𝑅𝑗𝑡 × 𝐺𝐷𝑃𝐶𝑡 + 𝜂1 𝐺𝐷𝑃𝐶𝑗𝑡 + 𝜀𝑖𝑡. the reduced form equation for Lit is. 19. 13.

(27) 𝐿𝑖𝑡 = 𝑐𝑖 + 𝜆1 𝐿𝑁𝐺𝑇𝐴𝑖𝑡−1 + 𝜆2 𝐸𝐺𝑇𝐴𝑖𝑡−1 + 𝜆3 𝐿𝐿𝑅𝐺𝐿𝑖𝑡−1 + 𝛿4 𝐺𝐷𝑃𝐶𝑗𝑡 −1 + 𝜀𝑖𝑡. 14. We have more instrumental variables than endogenous variables. Therefore, the endogenous variables are over-identified. Equations are estimated through two stage least squares (2SLS) estimator taking each bank’s liquidity risk as the dependent variable.. 3.5 Panel Unit Root Test and Sargan Test The most common unit root tests for panel data in practice are Levin, Lin, and Chu (2002), Im, Pesaran and Shin (2003) and Maddala and Wu (1999). Those tests are based on the following Augmented Dickey-Fuller regression:. ∆𝑦𝑖𝑡 = 𝛼𝑖 + 𝛿𝑖 𝑡 + 𝜌𝑖 𝑦𝑖,𝑡−1 + 𝜀𝑖𝑡 , 𝑖 = 1, … , 𝑁, 𝑡 = 1, … , 𝑇. 15. where αi and δi allow for fixed and unit specific time trends for each i. εit is error term and εit ~ i.i.d.(0,σ2). All series in the panel are nonstationary processes under the null hypothesis, and the alternative hypothesis is different for each test. It has individual i =1, …, N1 where yit is stationary and individual i = N1+1, N1+2, …, N where yit is nonstationary. Levin et al. (2002), Im et. al (2003) are used for balanced panel data. Maddala and Wu (1999) proposed the Fisher test to test the unit root for unbalanced panel data, and the null and alternative hypothesis is specified as:. 𝐻0 : 𝜌𝑖 = 0, ∀𝑖 𝐻𝑎 : 𝜌 𝑖 < 0, ∀𝑖 = 1, … , 𝑁1 and 𝜌𝑖 = 0, ∀𝑖 = 𝑁1 + 1, 𝑁1 + 2, … , 𝑁 20.

(28) Basing on Fisher (1932), Maddala and Wu (1999) proposed the Fisher Pλ test. The test suggests combining the p-value of the test statistic for unit root in each cross-sectional unit, and Fisher Pλ test does not require balanced panel data, so T can differ over cross-sections. The statistic of Fisher Pλ test is given by:. 𝑁. 𝑙𝑛 𝑝𝑖 ~𝜒 2 2𝑁. 𝑃𝜆 = −2. 16. 𝑖=1. The Fisher test statistics are shown in Table 4, and all of our variables are rejected to null hypothesis. It stands for our unbalance panel data is stationary, thus we continue our study. Sargan test is a test of overidentifying restrictions. The null hypothesis is that the instruments are valid instruments, and that the excluded instruments are correctly excluded from the estimated equation. Sargan test can exam whether our model is valid or not.. 21.

(29) Chapter 4 Empirical Results 4.1 Panel Unit Root Tests and Sargan Test Results Our data is unbalanced panel data, thus we use Fisher test for all variables to examine whether our variables are stationary or not. Table 4 shows that all variables are rejected to null hypothesis of nonstationarity at 5% significant level. After we confirm our data is stationary, we use Sargan test to examine whether liquidity creation is endogenous variable or not in different conditions. Table 5 and Table 6 show that liquidity creation is endogenous variable in overall condition without separation and bank-based financial system. For endogenous and exogenous model with different variables, the data consist of 30361 observations for endogenous overall financial system, 17279 observations for exogenous market-based financial system, 16402 observations for endogenous bank-based financial system and 45314 observations for exogenous bank-based financial system. The descriptive statistics are shown all conditions in Table 7 to Table 9. The Sargan test results show in overall data and in bank-based financial system, the liquidity creation is endogenous factor. Our analysis is focus on exogenous market-based financial system and endogenous bank-based financial system.. 4.2 Regression Results Endogenous test reveals that liquidity creation is the endogenous variable in overall condition in Table 5. Table 10 shows overall the relationship between liquidity risk and liquidity creation, liquidity creation is negative to liquidity risk at 5% 22.

(30) significant level. It shows relationship is on the opposite side of people think. When bank create liquidity, bank receives less liquidity risk in average. Adding financial system variable, liquidity creation is exogenous variable in market-based financial system and liquidity creation is endogenous variable in bank-based financial system. Table 11 shows the results of market-based financial system at Panel A and the results of bank-based financial system at Panel B. Underlying the Sargan test suggested, liquidity creation would be affected by some factors last year in bank-based financial system. And liquidity creation is more sensitive in bank-based financial system. Table 12 shows the results of in no financial crisis and financial crisis outbreak in market-based financial system. Liquidity creation is negative to liquidity risk in market-based financial system. And the coefficients are more sensitive under financial crisis outbreak rather than under no financial crisis. Table 13 shows the results of in no financial crisis and financial crisis outbreak in bank-based financial system. In bank-based financial system, liquidity creation is also negative to liquidity risk, but the relationship is more sensitive with no financial crisis than with financial crisis outbreak. This result is inversed to which in market-based financial system. Table 14 and Table 15 show that there is much more sensitive relation in bank-based financial system under no financial crisis, little sensitive relation in under financial crisis outbreak. In bank-based financial system, banks are higher sensitive connection in liquidity risk and liquidity creation. Other financing channels are more prosperous in market-based financial system than in bank-based financial system, so banks are more difficult to liquidate their assets in market-based financial system. Banks create more liquidity would receive less liquidity risk when financial crisis does not exist in bank-based financial system than those in market-based financial. 23.

(31) system. When financial crisis happens, the difference relationship between liquidity risk and liquidity creation in two financial systems has diminished.. 24.

(32) Chapter 5 Conclusions Government can use monetary policies to control economic activities via banks. When banks want to reduce their net lending position, there is an inverse relationship between deposits and loan interest rates. Meanwhile, depositors reduce their money on banks and loan interest rates increase. A monetary policy tightening induces a decrease in created liquidity (Rauch, Steffen, Hackethal and Tyrell, 2009). Liquidity creator is one of banks’ traditional roles. However, people had not been discussed about it for a long time due to a comprehensive measurement of liquidity creation until Berger and Bouwman (2005) constructed a comprehensive measurement for bank liquidity creation. Liquidity risk and liquidity creation may occur at one time, because once liquidity creation appears, it means banks relocate their activities in balance sheet and off balance sheet. The empirical result is liquidity creation is negative to liquidity risk in market-based and bank-based financial system. Purda (2007) indicated banks are suffered higher risk in bank-based financial system than other banks in market-based system due to the financial system remains have an important influence in firms’ credit rating assignment. Banks have been considered as effective monitors to firms, because bank can renegotiate with firms. Our empirical results show there are much higher sensitive correlations in bank-based financial system than in market-financial system. More liquidity creator banks acted, less liquidity risk banks suffered in bank-based financial system. According to the issue of the too big to fail, large banks have the advantage to do more commitment loans. It makes larger banks have more capability to possess more risky assets. This implies larger banks would have more liquidity risk. Our empirical results are consistent with this viewpoint. The relationships between liquidity risk and 25.

(33) bank size are all positive. And there is a contagion of banking failure due to bank's commitment problem, and more commitment problem would exacerbate a liquidity shortage (Diamond and Rajan, 2005). Contagion risk would be arisen due banks’ transactions in the interbank and capital markets (Wong and Hui, 2009). Under financial crisis outbreak, there is not an obvious difference in two financial systems. But a negative correlation exists under financial crisis outbreak. It implies when banks do more liquidity creation, banks face less pressure of insolvency. Contagion of banking failure could be prevented by doing more liquidity creation. Liquidity risk and liquidity creation is not the same in our empirical proof. Instead of increasing liquidity risk, doing more liquidity creation would decrease liquidity risk. There is not an internationally recognized“Liquidity Accord”since liquidity is not absolutely determined by assets and liabilities (Fernando, 2009). For the nature of banks, they based on public information and confidence affecting the end liquidity risk degree. Banks would take appropriate degree of liquidity risk for cost considering, thus less liquidity risk is not absolute good for banks. When banks benefit from asymmetric maturities in assets and liabilities, they should avoid taking too much liquidity risk in case of insolvency.. 26.

(34) Reference Acerbi, C., & Scandolo, G. (2007). Liquidity Risk Theory and Coherent Measures of Risk. Quantitative Finance, 8(7), 681. Athanasoglou, P. P., Delis, M. D., & Staikouras, C. (2008 ). Determinants of Bank Profitability in the South Eastern European Region. Journal of Financial Decision Making, Vol. 2, pp. 1-17. Barth, J. R., Caprio, G., & Levine, R. (2004). Bank Regulation and Supervision: What Works Best? Journal of Financial Intermediation, 13(2), 205-248. Barth, J. R., Nolle, D. E., Phumiwasana, T., & Yago, G. (2003). A Cross-Country Analysis of the Bank Supervisory Framework and Bank Performance. Financial Markets, Institutions & Instruments, 12(2), 67. Basel Committee on Banking Supervision, (1988), Basel I: International Convergence of Capital Measurement and Capital Standards, Bank for International Settlements. Basel Committee on Banking Supervision, (2004). Basel II: International Convergence of Capital Measurement and Capital Standards: A Revised Framework, Bank for International Settlements. Ben Naceur, S., & Kandil, M. (2009). The Impact of Capital Requirements on Banks' Cost of Intermediation and Performance: The Case of Egypt. Journal of Economics and Business, 61(1), 70-89. Berger, A. N., & Bouwman, C. H. (2008). Financial Crises and Bank Liquidity Creation. Working Paper Series. Berger, A. N., & Bouwman, C. H. (2009). Bank Liquidity Creation. The Review of Financial Studies, 22(9), 59. Bourke, P. (1989). Concentration and Other Determinants of Bank Profitability in Europe, North America and Australia. Journal of Banking & Finance, 13(1), 15. Brunnermeier, M. K. (2009). Deciphering the Liquidity and Credit Crunch 2007-2008. The Journal of Economic Perspectives, 23(1), 77. Decker, P. A. (2000). The Changing Character of Liquidity and Liquidity Risk Management: A Regulator's Perspective. The Journal of Lending & Credit Risk Management, 82(8), 9. Deep, A., & Schaefer, G. K. (2004). Are Banks Liquidity Transformers? SSRN Working Paper Series. Demirguc-Kunt, A., & Levine, R. (1999). Bank-Based and Market-Based Financial Systems: Cross-Country Comparisons. World Bank Policy Working Paper. Diamond, D. W., & Dybvig, P. H. (1983). Bank Runs, Deposit Insurance, and Liquidity. Journal of Political Economy, 91(3), 401.. 27.

(35) Diamond, D. W., & Rajan, R. G. (2001). Banks and Liquidity. The American Economic Review, 91(2), 422-425. Diamond, D. W., & Rajan, R. G. (2005). Liquidity Shortages and Banking Crises. The Journal of Finance, 60(2), 615-647. Freixas, X., Martin, A., & Skeie, D. R. (2010). Bank Liquidity, Interbank Markets, and Monetary Policy. Working Paper Series. Gualandri, E., Landi, A., & Venturelli, V. (2009). Financial Crisis and New Dimensions of Liquidity Risk: Rethinking Prudential Regulation and Supervision. Working Paper Series. Kosmidou, K., Tanna, S., & Fotios. (2005). Determinants of Profitability of Domestic UK Commercial Banks: Panel Evidence from the Period 1995-2002. . Money Macro and Finance (MMF) Research Group Conference. Matz, L., & Neu, P. (2007). Liquidity Risk Measurement and Management: A practitioner's guide to global best practice. Singapore: John Wiley & Sons (Asia) Pte Ltd. Molyneux, P. p., & Thornton, J. p. (1992). Determinants of European Bank Profitability: A Note. Journal of Banking & Finance, 16(6), 6. Motyka, R., Leuca, A., & Fawson, J. (2005). Liquidity Risk: A Fresh Look. UBS Research Paper. Naceur, S. B., & Kandil, M. (2009). The Impact of Capital Requirements on Banks' Cost of Intermediation and Performance: The Case of Egypt. Journal of Economics and Business, 61(1), 70-89. Pasiouras, F., & Kosmidou, K. (2007). Factors Influencing the Profitability of Domestic and Foreign Commercial Banks in the European Union. Research in International Business and Finance, 21(2), 222-237. Purda, L. D. (2008). Risk Perception and the Financial System. Journal of International Business Studies, 39(7), 19. Rauch, C., Steffen, S., Hackethal, A., & Tyrell, M. (2009). Savings Banks, Liquidity Creation and Monetary Policy. Working Paper Series. Saunders, A., & Cornett, M. M. (2008). Financial Institutions Management:A Risk Management Approach. Boston McGraw-Hill. Shen, C.-H., Kuo, C.-J., & Chen, H.-J. (2001). Determinants of Net Interest Margins in Taiwan. Journal of Financial Studies, 9, 47-83. Stange, S., & Kaserer, C. (2009). Market Liquidity Risk - An Overview. Working Paper Series. Wong, T.-C., & Hui, C.-H. (2009). A Liquidity Risk Stress-Testing Framework with Interaction between Market and Credit Risks. Working Paper Series.. 28.

(36) Table 1 Variable Description Variables. Notations. Description/Calculation. Liquidity Risk. LR. The ratio of financing gap to total assets. Financing gap defined as a bank's loans minus core deposit and cash.. Liquidity Creation to Total Equity. LC. Follow Berger and Bouwman (2009) to calculate the liquidity creation, and we divide the volume by total equity. For restrictions with Bankscope, we adjust some calculated items.. Gross Total Assets. LNGTAt. To measure a bank’s size, we follow Berger and Bouwman (2009) to calculate the gross total assets (total assets plus allowance for loan and lease losses and the allocated transfer risk reserve), and we smooth the volume by natural log. For the restrictions from database, we calculate the GTA by total assets plus loan loss reserve.. LNGTAt-1 The bank’s size of last year. Equity to Gross. EGTAt. The ratio of equity to gross total assets.. Total Assets. EGTAt-1. The equity to gross total assets ratio of last year.. Less Risky Assets. LRA. The ratio of risky liquid assets to gross total assets. Risky liquid assets is sum of the liquid assets with little or no discount.. Loan Loss Reserve. LLRGLt-1. The loan loss reserve to loans ratio of last year.. GDPCt. Annual percent change of GDP. GDPCt-1. GDP annual percent change of last year. OSP. It is used to measure of legal power of the supervisory agency. The value is principal. to Gross Loans Macroeconomy Official Supervisory Power. component indicator of fourteen variables. Private Monitoring Index. PMI. It is used to measures regulations that empower private monitoring of banks. Principal component indicator of nine variables.. Overall Bank Activities and Ownership Restrictiveness. BAR. Indicator of bank's ability to engage in business of securities underwriting, insurance underwriting and selling, and in real estate investment, management, and development.. 29.

(37) Table 2 Liquidity Activities Classification with Bankscope Database Restrictions Assets Illiquid (weight=. 1 2. ). HP / Lease Other Loans Loans to Group Companies / Associates. 1. Semi-liquid (weight=0). Liquid (weight= -. Mortgages Loans to Banks Loans to Municipalities / Government. Deposits with Banks Cash and Due from Banks Due from Other Banks Due from Other Credit. 2. Loans to Other Corporate Overdue Loans Restructured Loans Other non-performing Loans. Institutions Total Securities Treasury Bills Other Bills Bonds. Trust Account Lending Other Lending Intangible Assets Other Non Earning Assets Land and Buildings. CDs Equity Investments Other Investments. ). Total Fixed Assets Liabilities plus Equity liquid (weight=. 1. 1. Semi-liquid (weight=0). Illiquid (weight= - 2 ). Deposits – Demand Deposits – Savings Banks Deposits Municipalities / Government Deposits. Time Deposit Commercial Paper Other Funding Other Securities Debt Securities. Subordinated Debt Other Liabilities Total Equity Other Bonds Mortgage Bonds. Commercial Deposits Other Deposits. Securities Loaned. Convertible Bonds. Semi-liquid (weight=0). Liquid (weight= -. 2. ). Off-balance sheet activities Illiquid (weight= Acceptances. 1 2. ). Documentary Credits. Guarantees Other. 30. 1 2. ).

(38) Table 3 Classification of Financial Systems of Sample Countries Country. Financial System. Belgium. Bank-Based. Canada. Market-Based. France. Bank-Based. Germany. Bank-Based. Italy. Bank-Based. Japan. Bank-Based. Netherlands. Market-Based. Switzerland. Market-Based. Sweden. Market-Based. United Kingdom. Market-Based. United States. Market-Based. Note: All countries are sorted by Demirguc-Kunt and Levine (1999).. Table 4 Panel Unit Root Test Results Using Unbalanced Panel Data This table contains all the variables of Fisher test statistics. Fisher test is used to check whether one variable is stationary or not. The null hypothesis is that all all series in the panel are nonstationary processes, and the bellow results of all variables are rejected to null hypothesis. D.F.. chi2. Prob> chi2. LR. 12720. 2.60E+04. 0.0000. LCE. 12720. 2.10e+04. 0.0000. LNGTAt. 12720. 1.35E+04. 0.0000. LNGTAt-1. 11790. 1.23e+04. 0.0011. EGTAt. 12720. 2.32E+04. 0.0000. EGTAt-1. 11790. 1.94e+04. 0.0000. LRA. 12720. 2.04E+04. 0.0000. LLRGLt-1. 6656. 1.51e+04. 0.0000. GDPCt. 12720. 3.08E+04. 0.0000. GDPCt-1. 12720. 2.35E+04. 0.0000. OSP×GDPCt. 12720. 3.08E+04. 0.0000. PMI×GDPCt. 12720. 3.08E+04. 0.0000. BAR×GDPCt. 12720. 3.08E+04. 0.0000. Notes: All variables in these regressions have been defined in Table 1. ***, ** and * denote significance at the 1%, 5% and 10% levels.. 31.

(39) Table 5 Sargan Test Results of All Samples Sargan test is a test of overidentifying restrictions. The null hypothesis is that the instruments are valid instruments, and that the excluded instruments are correctly excluded from the estimated equation. Overall With No Separation Statistic. D.F.. P-value. (1) (2). 0.318 0.465. 3 3. 0.9565 0.9266. (3) (4). 0.519 0.375. 3 3. 0.9146 0.9453. Note: We use LR as dependent variable, and LC as explanatory variable. Model 1 only cosists of LNGTAt, EGTAt, LRA and GDPCt control variables. And we separately add OSP×GDPCt control variable in model 2, PMI×GDPCt control variable in model 3 and BAR×GDPCt control variable in model 4. ***, ** and * denote significance at the 1%, 5% and 10% levels.. Table 6 Sargan Test Results in Different Conditions Sargan test is a test of overidentifying restrictions. The null hypothesis is that the instruments are valid instruments, and that the excluded instruments are correctly excluded from the estimated equation. Market-Based Financial System Model. Statistic. Bank-Based Financial System. D.F. P-value. Model. Overall condition. Statistic. D.F. P-value. Overall condition. (1) (2). 32.727 32.875. 3 3. 0.0000*** 0.0000***. (1) (2). 0.436 0.56. 3 3. 0.9328 0.9054. (3) (4). 33.342 31.805. 3 3. 0.0000*** 0.0000***. (3) (4). 0.597 0.562. 3 3. 0.8971 0.9052. With No Financial Crisis. With No Financial Crisis. (1) (2). 8.939 10.783. 3 3. 0.0301* 0.0130*. (1) (2). 2.958 3.052. 3 3. 0.3981 0.3837. (3) (4). 8.128 15.667. 3 3. 0.0434* 0.0013**. (3) (4). 2.913 2.937. 3 3. 0.4052 0.4015. With Financial Crisis Outbreak. With Financial Crisis Outbreak. (1) (2). 25.67 25.871. 3 3. 0.0000*** 0.0000***. (1) (2). 1.598 1.694. 3 3. 0.6599 0.6383. (3) (4). 25.598 30.173. 3 3. 0.0000*** 0.0000***. (3) (4). 1.767 1.669. 3 3. 0.6220 0.6440. Note: We use LR as dependent variable, and LC as explanatory variable. Model 1 only cosists of LNGTAt, EGTAt, LRA and GDPCt control variables. And we separately add OSP×GDPCt control variable in model 2, PMI×GDPCt control variable in model 3 and BAR×GDPCt control variable in model 4. ***, ** and * denote significance at the 1%, 5% and 10% levels.. 32.

(40) Table 7 Descriptive Statistics of All Samples Overall With No Separation Variable. Mean. S.D.. Min. Max. LR LC LRA LNGTAt. -8.82623 9.159409 35.29101 13.96921. 31.72837 237.0005 20.84883 1.943265. -183.378 -85.5014 0 7.073389. 99.38158 41224.39 388.786 21.74467. EGTAt GDPCt OSP×GDPCt PMI×GDPCt BAR×GDPCt. 9.621741 1.295872 11.90485 8.761166 2.903772. 7.842396 1.575653 17.52663 11.37983 4.414864. 0.000447 -1.81096 -23.5424 -14.4876 -5.97615. 100 5.22 45.74921 31.32 13.01324. LNGTAt-1 LLRGLt-1 EGTAt-1 GDPCt-1. 13.89496. 1.945326. 7.197469. 21.68295. 2.63229 9.646595 1.321307. 4.340915 7.826887 1.578959. -1.54 0.000447 -1.81096. 100 96.84032 5.22. Obs. 30361. Notes: All variables in these regressions have been defined in Table 1. ***, ** and * denote significance at the 1%, 5% and 10% levels.. Table 8 Descriptive Statistics in Market-Based Financial System Market-Based Financial System. Variable. Mean. S.D.. Min. Max. LR LC LRA. -5.71183 6.809581 33.7977. 31.70568 14.17131 25.70861. -191.792 -672.592 0. 99.9799 596.4706 259.1788. LNGTAt EGTAt GDPCt OSP×GDPCt PMI×GDPCt. 13.83956 11.3807 1.833547 23.2991 14.28857. 1.955941 11.34319 0.838142 10.91538 6.468926. 7.073389 0.138631 -0.45277 -3.4331 -3.16936. 21.74467 100 5.22 45.74921 31.32. BAR×GDPCt. 4.432289. 2.494428. -0.81498. 9.803402. Obs. 17279. Notes: All variables in these regressions have been defined in Table 1. ***, ** and * denote significance at the 1%, 5% and 10% levels.. 33.

(41) Table 9 Descriptive Statistics in Bank-Based Financial System with Taking Liquidity Creation as Different Variable. Bank-Based Financial System and Take Liquidity Creation As Endogenous Variable. Variable. Mean. S.D.. Min. Max. LR LC LRA. -12.9109 10.97489 38.74642. 33.30545 322.2897 17.70519. -133.848 -85.5014 0. 99.38158 41224.39 388.786. LNGTAt EGTAt GDPCt OSP×GDPCt. 14.00569 8.497834 0.885853 2.275966 4.304039. 1.971925 5.619944 1.916553 16.23083 12.71938. 8.875154 0.000447 -1.81096 -23.5424 -14.4876. 21.60818 91.34595 5.205295 31.23177 31.23177. 1.572604 13.94256 3.392534 8.603806. 5.175608 1.981614 4.347244 5.808842. -5.97615 8.503508 0 0.000447. 13.01324 21.68295 76.03 95.94977. GDPCt-1. 0.948588. 1.939252. -1.81096. 5.205295. Obs. 16402. PMI×GDPCt BAR×GDPCt LNGTAt-1 LLRGLt-1 EGTAt-1. Bank-Based Financial System s and Take Liquidity Creation As Exogenous Variable. Variable. Mean. S.D.. Min. Max. LR. -12.1482. 28.39889. -133.848. 99.85476. LC LRA LNGTAt EGTAt GDPCt. 10.61682 13.54222 7.574565 39.43165 0.919055. 194.2112 1.739893 7.291206 17.74888 1.363273. -813.419 7.221386 0.000447 0 -1.81096. 41224.39 21.82674 100.1903 388.786 5.205295. OSP×GDPCt PMI×GDPCt BAR×GDPCt. 6.804367 4.717914 1.485745. 12.27417 8.710292 3.465381. -23.5424 -14.4876 -5.97615. 31.23177 31.23177 13.01324. Obs. 45314. Notes: All variables in these regressions have been defined in Table 1. ***, ** and * denote significance at the 1%, 5% and 10% levels.. 34.

(42) Table 10 The Regression Result of All Samples Overall the Relationship Between Liquidity Risk and Liquidity Creation Pre. Sign. (1). (2). (3). (4). CONSTANT. ?. LC. -. -37.419* (-1.89) -0.344* (-1.93). -34.453* (-1.74) -0.335* (-1.94). -33.848* (-1.73) -0.330** (-1.96). -39.434** (-1.99) -0.345* (-1.9). LNGTAt. +. EGTAt. -. LRA. -. 4.620*** (3.42) 0.039 (0.17) -0.977***. 4.403*** (3.32) 0.034 (0.15) -0.971***. 4.366*** (3.35) 0.040 (0.19) -0.970***. 4.791*** (3.5) 0.041 (0.18) -0.973***. (-11.05). (-11.46). (-11.64). (-11.09). 1.007 (0.96). -1.748 (-0.72) 0.283 (1.04). -8.124 (-1.22). -5.045 (-1.28). GDPCt. -. OSP×GDPCt. -. PMI×GDPCt. -. BAR×GDPCt. -. 1.304 (1.31) 2.523 (1.4). R2. 0.0242. 0.025. 0.0258. 0.0246. Obs. 30361. 30361. 30361. 30361. Notes: All variables in these regressions have been defined in Table 1. ***, ** and * denote significance at the 1%, 5% and 10% levels.. 35.

(43) Table 11 The Results In Different Financial Systems Pre. Sign. (1). (2). (3). (4). Panel A: Market-Based Financial System CONSTANT. ?. -27.623***. -27.665***. -27.456***. -27.187***. LC. -. -0.031***. -0.031***. -0.031***. -0.032***. LNGTAt. +. 3.148***. 3.154***. 3.131***. 3.168***. EGTAt. -. 0.376***. 0.376***. 0.375***. 0.381***. LRA. -. -0.750***. -0.750***. -0.750***. -0.754***. GDPCt. -. -0.201. -0.029. -1.432. 1.860***. OSP×GDPCt. -. PMI×GDPCt. -. BAR×GDPCt. -. -0.015 0.164 -0.997***. R2. 0.449. 0.4496. 0.4482. 0.4814. Obs. 17279. 17279. 17279. 17279. Panel B: Bank-Based Financial System CONSTANT. ?. -45.203. -39.265. -40.858. -48.472*. LC. -. -0.216**. -0.204**. -0.201**. -0.201**. LNGTAt. +. 5.680***. 5.440***. 5.580***. 6.033***. EGTAt. -. -0.264. -0.230. -0.191. -0.135. LRA. -. -1.133***. -1.111***. -1.115***. -1.130***. GDPCt. -. 1.389. -4.989. -18.687*. -9.787*. OSP×GDPCt. -. PMI×GDPCt. -. BAR×GDPCt. -. 2. 0.807* 3.105* 4.370**. R. 0.0371. 0.0445. 0.0469. 0.0455. Obs. 16402. 16402. 16402. 16402. Notes: All variables in these regressions have been defined in Table 1. ***, ** and * denote significance at the 1%, 5% and 10% levels.. 36.

(44) Table 12 The Results of the Subsample in Market-Based Financial System Pre. Sign. (1). (2). (3). (4). Panel A : Market-Based Financial System without financial crisis ? + -. -41.446*** -0.028* 3.795*** 0.408***. -41.894*** -0.027* 3.842*** 0.410***. -41.277*** -0.028* 3.780*** 0.408***. -40.686*** -0.035** 3.841*** 0.419***. GDPCt. -. -0.645*** 0.342. -0.645*** -1.996. -0.660*** 7.628***. OSP×GDPCt. -. -0.644*** 3.383*** -0.249***. PMI×GDPCt. -. BAR×GDPCt. -. CONSTANT LC LNGTAt EGTAt LRA. 0.305 -3.191***. R2. 0.408. 0.4158. 0.4068. 0.4949. Obs. 10034. 10034. 10034. 10034. Panel B : Market-Based Financial System with financial crisis ? -. -8.764** -0.040***. -8.851** -0.040***. -9.291** -0.039***. -8.876** -0.040***. GDPCt. + -. 2.444*** 0.332*** -0.977*** -0.795***. 2.494*** 0.335*** -0.978*** 1.918. 2.430*** 0.329*** -0.975*** -1.590***. OSP×GDPCt. -. 2.455*** 0.333*** -0.978*** -0.477 -0.028. PMI×GDPCt. -. BAR×GDPCt. -. CONSTANT LC LNGTAt EGTAt LRA. 2. R. Obs. -0.361 0.396 0.4906 7245. 0.4915 7245. 0.4917 7245. 0.4788 7245. Notes: All variables in these regressions have been defined in Table 1. ***, ** and * denote significance at the 1%, 5% and 10% levels.. 37.

(45) Table 13 The Results of the Subsample in Bank-Based Financial System Pre. Sign. (1). (2). (3). (4). Panel A : Bank-Based Financial System without financial crisis ? +. 187.785** -3.761*** -7.022. 185.531** -3.719*** -6.902. 185.987** -3.724*** -6.911. 183.494** -3.702*** -6.746. -1.422* -1.536*** -0.416. -1.403* -1.529*** -0.759 0.049. -1.404* -1.530*** -2.583. -1.374* -1.531*** -3.008. PMI×GDPCt. -. BAR×GDPCt. -. CONSTANT LC LNGTAt EGTAt LRA GDPCt OSP×GDPCt. 2. 0.348 1.047 0.0655 9287. R. Obs. 0.0672 9287. 0.0678 9287. 0.069 9287. Panel B : Bank-Based Financial System with financial crisis CONSTANT. ?. -46.818***. -46.033***. -44.255***. -47.098***. LC. -. -0.049**. -0.048**. -0.048**. -0.048**. LNGTAt. + -. 5.349*** 0.087 -1.091***. 5.312*** 0.094 -1.090***. 5.260*** 0.099 -1.092***. 5.473*** 0.136 -1.095***. -. 2.415***. 1.907 0.049. -2.352. -3.275. EGTAt LRA GDPCt OSP×GDPCt PMI×GDPCt BAR×GDPCt 2. 0.678 2.117*. R. 0.2231. 0.2243. 0.2165. 0.2191. Obs. 7115. 7115. 7115. 7115. Notes: All variables in these regressions have been defined in Table 1. ***, ** and * denote significance at the 1%, 5% and 10% levels.. 38.

(46) Table 14 The Regression Result in Different Financial Systems not during Financial Crisis Pre. Sign. (1). (2). (3). (4). Panel A : Market-Based Financial System without financial crisis CONSTANT. ?. -41.446***. -41.894***. -41.277***. -40.686***. LC. -. -0.028*. -0.027*. -0.028*. -0.035**. LNGTAt. +. 3.795***. 3.842***. 3.780***. 3.841***. EGTAt. -. 0.408***. 0.410***. 0.408***. 0.419***. LRA. -. -0.645***. -0.644***. -0.645***. -0.660***. GDPCt. -. 0.342. 3.383***. -1.996. 7.628***. OSP×GDPCt. -. PMI×GDPCt. -. BAR×GDPCt. -. -0.249*** 0.305 -3.191***. R2. 0.408. 0.4158. 0.4068. 0.4949. Obs. 10034. 10034. 10034. 10034. Panel B : Bank-Based Financial System without financial crisis CONSTANT. ?. 187.785**. 185.531**. 185.987**. 183.494**. LC. -. -3.761***. -3.719***. -3.724***. -3.702***. LNGTAt. +. -7.022. -6.902. -6.911. -6.746. EGTAt. -. -1.422*. -1.403*. -1.404*. -1.374*. LRA. -. -1.536***. -1.529***. -1.530***. -1.531***. GDPCt. -. -0.416. -0.759. -2.583. -3.008. OSP×GDPCt. -. PMI×GDPCt. -. BAR×GDPCt. -. 0.049 0.348 1.047. R2. 0.0655. 0.0672. 0.0678. 0.069. Obs. 9287. 9287. 9287. 9287. Notes: All variables in these regressions have been defined in Table 1. ***, ** and * denote significance at the 1%, 5% and 10% levels.. 39.

(47) Table 15 The Regression Result in Different Financial Systems during Financial Crisis Pre. Sign. (1). (2). (3). (4). Panel A : Market-Based Financial System with financial crisis ? +. -8.764** -0.040*** 2.444***. -8.851** -0.040*** 2.455***. -9.291** -0.039*** 2.494***. -8.876** -0.040*** 2.430***. GDPCt. -. 0.332*** -0.977*** -0.795***. 0.335*** -0.978*** 1.918. 0.329*** -0.975*** -1.590***. OSP×GDPCt. -. 0.333*** -0.978*** -0.477 -0.028. PMI×GDPCt. -. BAR×GDPCt. -. CONSTANT LC LNGTAt EGTAt LRA. 2. -0.361 0.396 0.4906 7245. R. Obs. 0.4915 7245. 0.4917 7245. 0.4788 7245. Panel B : Bank-Based Financial System with financial crisis CONSTANT. ?. -46.818***. -46.033***. -44.255***. -47.098***. LC. GDPCt. + -. -0.049** 5.349*** 0.087 -1.091*** 2.415***. -0.048** 5.312*** 0.094 -1.090*** 1.907. -0.048** 5.260*** 0.099 -1.092*** -2.352. -0.048** 5.473*** 0.136 -1.095*** -3.275. OSP×GDPCt. -. PMI×GDPCt. -. BAR×GDPCt. -. LNGTAt EGTAt LRA. 2. R. Obs. 0.049 0.678 2.117* 0.2231 7115. 0.2243 7115. 0.2165 7115. 0.2191 7115. Notes: All variables in these regressions have been defined in Table 1. ***, ** and * denote significance at the 1%, 5% and 10% levels.. 40.

(48) Appendix: Survey Questions of Bank Regulation and Supervision I. Based on Barth, Caprio, and Levine (2004), the survey questions used to construct the official supervisory power: 1. Does the supervisory agency have the right to meet with external auditors to discuss their report without the approval of the bank? Yes/No 2. Are auditors required by law to communicate directly to the supervisory agency any presumed involvement of bank directors or senior managers in elicit activities, fraud, or insider abuse? Yes/No 3. Can supervisors take legal action against external auditors for negligence? Yes/No 4. Can the supervisory authority force a bank to change its internal organizational structure? Yes/No 5. Are off-balance sheet items disclosed to supervisors? Yes/No 6. Can the supervisory agency order the bank’s directors or management to constitute provisions to cover actual or potential losses? Yes/No 7. Can the supervisory agency suspend the directors’ decision to distribute dividends? Yes/No 8. Can the supervisory agency suspend the directors’ decision to distribute bonuses? Yes/No 9. Can the supervisory agency suspend the directors’ decision to distribute management fees? Yes/No 10. Can the supervisory agency legally declare—such that this declaration supersedes the rights of bank shareholders—that a bank is insolvent? Yes/No 11. Does the Banking Law give authority to the supervisory agency to intervene—that is, suspend some or all ownership rights—a problem bank? Yes/No Regarding bank restructuring and reorganization, can the supervisory agency or any other government agency do the following:? Yes/No 12. Supersede shareholder rights? Yes/No 13. Remove and replace management? Yes/No 14. Remove and replace directors? Yes/No II. Based on Barth, Caprio, and Levine (2004), the survey questions used to construct the private monitoring index: 1. Whether bank directors and officials are legally liable for the accuracy of information disclosed to the public? Yes/No 2. Whether banks must publish consolidated accounts? Yes/No 3. Do banks must be audited by certified international auditors? Yes/No 41.

(49) 4. Whether 100% of the largest 10 banks are rated by international rating agencies? Yes/No 5. Are off-balance sheet items are disclosed to the public? Yes/No 6. Whether banks must disclose their risk management procedures to the public? Yes/No 7. Whether accrued, though unpaid interest/principal enter the income statement while the loan is still non-performing? Yes/No 8. Is subordinated debt is allowable as part of capital? Yes/No 9. Whether there is no explicit deposit insurance system and no insurance was paid the last time a bank failed? Yes/No III. Based on Barth, Caprio, and Levine (2004), the survey questions used to construct overall bank activities and ownership restrictiveness: Bank activities restrictiveness: 1. What is the level of regulatory restrictiveness for bank participation in securities activities (the ability of banks to engage in the business of securities underwriting, brokering, dealing, and all aspects of the mutual fund industry)? 2. What is the level of regulatory restrictiveness for bank participation in insurance activities (the ability of banks to engage in insurance underwriting and selling)? 3. What is the level of regulatory restrictiveness for bank participation in real estate activities (the ability of banks to engage in real estate investment, development, and management)? Unrestricted = 1: full range of activities can be conducted directly in the bank. Permitted = 2: full range of activities can be conducted, but some or all must be conducted in subsidiaries. Restricted = 3: less than full range of activities can be conducted in the bank or subsidiaries. Prohibited = 4: the activity cannot be conducted in either the bank or subsidiaries. Ownership restrictiveness: 1. What is the level of regulatory restrictiveness for bank ownership of nonfinancial firms? Unrestricted = 1: a bank may own 100 percent of the equity in any nonfinancial firm. Permitted = 2: a bank may own 100 percent of the equity of a nonfinancial firm, but ownership is limited based on a bank’s equity capital. Restricted = 3: a bank can only acquire less than 100 percent of the equity in a nonfinancial firm. Prohibited = 4: a bank may not acquire any equity investment in a nonfinancial firm. Source: World Bank guide questions, which is available from World Bank research (Bank Regulation and Supervision) or Barth, Caprio, and Levine (2004). 42.

(50)