Monitoring Banking Sector Fragility
    By Aykut Kibritçiog ˘lu* Introduction The last two decades of the 20th century are mainly characterized by currency and banking crises. Particularly the 1994–95 Mexican, 1997–98 Asian and 1998 Russian financial crises seem to have strongly stimulated academic research on the timing, duration, causes, effects, and cures of both currency and banking crises. As also defined in the International Monetary Fund’s World Economic Outlook (May 1998, p. 74), a currency crisis may be said to occur when a speculative attack on the exchange value of a national currency results in a devaluation (or sharp depreciation) of the currency, or forces the authorities to defend the currency by expending large volumes of international reserves or by sharply raising interest rates. Accordingly, many researchers construct a monthly or quarterly index, which is called a foreign-exchange-market pressure (FEMP) index, to identify and predict currency crisis episodes. Typically, a FEMP index is calculated as the weighted average of (1) percentage changes in nominal exchange rates, (2) negative of percentage changes in foreign exchange reserves, and (3) international interest rate differential. A crisis is then said
    1
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking
    
    episodes are recently reviewed and discussed by Frydl (1999), Eichengreen and Arteta (2000) and Boyd et al. (2001), among others. In the literature, it is usually argued that, contrary to the case of currency crises, building a time series index to identify banking crisis episodes is highly difficult, particularly because of the lack of reliable sectoral data on banks’ financial activities. It is frequently stressed that the data on non-performing loans in many countries are either not available or are systematically distorted (see Hawkins and Klau, 2000). Consequently, existing methods that are widely used to pinpoint banking crisis episodes are generally event-based. That is to say, they usually are based on the available ex-post figures, which are related to banks’ losses and governments’ bailout costs. The years attached to crises reviewed in the literature are those, which are more or less generally accepted by finance experts familiar with the countries (Caprio and Klingebiel, 2003). Furthermore, in many studies, the crisis episodes are also identified with the help of the country-specific banking information that is available in the databases of some international financial organizations (e.g., Bank for International Settlements, the International Monetary Fund, and the World Bank) or that is published in major daily newspapers or popular economy journals (e.g., Wall Street Journal, New York Times, and American Banker).
    
    Management
    
    to arise when the index value exceeds an arbitrarily determined threshold value. In this sense, the identification of currency crisis episodes using the so-called FEMP index is an easy and highly mechanical task, and it is highly useful in empirical research. A bank failure, on the other hand, refers to a situation in which the excessively rising liquidity, credit, interest-rate, or exchange-rate risk pushes the bank to suspend the internal convertibility of its liabilities. If the bank failure problem undermines an entire banking system, the crisis turns out to be systemic. Potential or actual difficulties in the domestic banking sector sometimes may force the government to intervene into the market to prevent their far reaching adverse effects, such as that on the corporate sector and foreign exchange market. The exact timing of government intervention and the extent of the possible bailout costs obviously vary according to time and space. The last two decades have seen a dramatic increase of systemic banking crises, as documented mainly by the comprehensive studies of Caprio and Klingebiel (1996, 1999, 2002 and 2003)2 and Lindgren et al. (1996). The identified domestic crisis
    
    51
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management Advantages Event-Based Approach
    
    Table 1 Comparison of Different Methods to Identify Episodes of Banking Crises and High Banking Fragility
    
    Statistical Approach Employed in This Study
    
    - It is relatively easy to find information on the date of - The banking-sector-fragility (BSF) index is very useful
    both government intervention and change in banking regulations. to monitor and interpret the developments in the sector by using monthly banking data. - The monthly BSF index can easily be employed within a single-country framework. - One can easily define criteria to differentiate between systemic and non-systemic crises or fragility, based on the fluctuations in the BSF index.
    
    Disadvantages
    
    - Pinpointing the dates of crises is possible only for the - Reliable and continuous monthly banking sector data for every country in the world is not available. annual data frequency. Therefore, it is not useful to apply it to predict and discuss banking crises within the frame- - Some of the data may be biased because of the wrong reporting practices (as in the case of non-performing work of monthly data frequency. loans or interest rates) or country-specific legal regula- In general, the usage of crisis years is restricted to limittions. ed dependent variable models (logit, probit, etc.). Hence, it usually requires a multi-country framework to increase - The data, and hence, the BSF index do not necessarily reflects the exact date of government intervention. the number of crises considered. - The date of government intervention, which is used to pinpoint crisis dates, does not necessarily reflect the actual beginning date of a particular crisis. - It is not always easy to judge whether a crisis is systemic or not, particularly if one uses information only on government intervention. - For an individual researcher, it is not easy to collect event-based information on banking sector difficulties across the world.
    
    The event-based mainstream approach, however, clearly has some disadvantages against the statistical approach that may be employed by constructing a banking sector fragility (BSF) index using the available monthly time-series data. Table 1 summarizes and compares both the advantages and disadvantages of these two methods. Apparently, the statistical approach has some overwhelming advantages with respect to the widely used event-based approach. Particularly, the measurement of banking sector vulnerability by using "monthly" data is a highly attractive feature of the time-series-based statistical approach. A monthly BSF index may significantly contribute to policy makers’ efforts towards early detection of approaching banking sector difficulties. It also can be used as a reliable method to identify crisis episodes, even though it cannot completely substitute the event-based approach. Departing from this idea, this paper attempts to propose a weighted BSF index to measure and monitor changes in the banking sector fragility by using monthly data for selected countries. Despite the above-mentioned arguments on the scarcity of some relevant banking sector data, we aim to show that even the existing data for national banking sectors definitely allows us to work with "monthly statistics" instead of "events", if one intends to pinpoint banking crisis episodes in different countries. The rest of the paper is organized as follows. In section 2, the
    
    the tendency towards crisis in the banking system are demonstrated in detail. The discussions there are primarily based on a brief literature review. Conceptually, section 2 focuses on the various financial risks that banks face. An empirically functional BSF index is then explicitly formulated in section 3. Section 4 covers both the statistical details of calculations and visual presentations of the BSF indices for each of the selected 22 countries. Finally, in section 5, the results of the current study are briefly evaluated with respect to previous, i.e. eventbased studies, such as Caprio and Klingebiel (2003) and Lindgren et al. (1996).
    
    Banks’ Net Worth, Economic Risks, and Potential Fragility-Indicators Banks are intermediaries, which aim to earn profits in financial markets by acquiring funds, and investing these funds or lending them to borrowers. Banks’ liabilities are the funds that they acquire from savers in the form of deposits or as borrowings, while their assets mainly include reserves, marketable securities, and loans. The difference between the assets and liabilities of a bank equals its net worth, which in fact shows the bank’s remaining value, or equity capital, after it has met all of its liabilities.3 When the net worth of a bank turns into negative, the bank becomes insolvent.
    
    52 need for and possibility of constructing an index to measure
    
    Explicitly, a bank is exposed to the risk that the values of its assets and/or liabilities change in financial markets. That is, all banks are potentially exposed to different types of economic risks, such as (i) liquidity risk (i.e., massive bank runs), (ii) credit risk (i.e., rising non-performing loans), and (iii) exchange-rate risk (i.e., banks’ increasing unhedged foreign currency liabilities). Therefore, a bank’s net worth, and hence, a bank failure basically can be associated with excessive risk-taking of bank managers. In fact, several empirical studies in the literature show that massive bank runs and withdrawals, enormous lending booms, and/or high increases in the foreign liabilities of the banking sector are among the major leading indicators of impending banking crises.4 1. Bank Runs and Liquidity Risk No matter what the reason, savers’ massive run on deposits may indeed trigger a new (or accelerate the ongoing) increase in the fragility of the banking sector to crisis. However, it should be noted that the presence of a so-called deposit insurance system may prevent depositors from withdrawals, and hence, this may significantly weaken the potential link between bank runs and bank insolvency. Furthermore, Kaminsky and Reinhart (1999) argue that "recent" banking problems worldwide do arise from the assets side (i.e., increases in non-performing loans) instead of the liability side (i.e., bank runs). 2. Lending Boom, Non-Performing Loans, and Credit Risk A lending boom on the assets side of a bank’s balance sheet is likely to be caused by the bank’s poor, or over-optimistic, evaluation regarding the investors’ credit applications. Moreover, a bank can credit risky projects (and thus, it may contribute to a possible credit-boom process in the country), if the borrower is an economic unit, which actually is somehow connected with the bank. This is called insider, or connected, lending in the literature. Additionally, the existence of deposit insurance may encourage bank managers to take excessive risk (moral hazard problem) by loosening the credit taps further than expected. These considerations imply that credit booms easily may be linked to banking crises, at least at the theoretical level. However, Gourinchas et al. (2001) recently emphasized that, while most banking crises may be preceded by a lending boom, most lending booms are not followed by a banking crisis.
    5
    
    currency crises. Referring to crises since the early 1980s, they briefly argue that problems in the banking sector typically precede a currency crisis but they are not necessarily the immediate cause of currency crises. In turn, however, the currency crisis deepens the banking crisis, activating a vicious spiral. Obviously, in the absence of regulations limiting banks’ open foreign currency positions and if the domestic currency is not expected to depreciate (or to be devaluated) in the near future, banks are likely to be motivated to take excessive risk by acquiring funds from international financial markets. If domestic banks have large amounts of unhedged foreign currency debt, then a sudden devaluation may cause a sharp fall in the net worth of banks thereby increasing the vulnerability of the domestic banking sector. Therefore, banks may try to reduce their foreign currency liabilities, if they foresee that the domestic currency will be devaluated soon. Accordingly, they also may attempt to reduce the high debt burden by increasing the credit interest rates.6 Hence, bank credits to the private sector may considerably decline in the aftermath of devaluation. Furthermore, devaluation expectations and/or rises (falls) in foreign (domestic) interest rates may trigger a massive bank run, as also discussed by Calvo et al. (1994) and Obstfeld and Rogoff (1995). To examine the causes of banking crises empirically or to develop a model to monitor and predict impending banking sector problems, one first needs to be able to empirically identify the episodes and severity of previously occurred crises. Our discussions so far show that there is a strong motivation to design an empirically functional BSF index which is able to reflect the changes in the excessive risk-taking behavior of banks for monthly data frequency. Therefore, the next section is devoted to the creation of this type of index.
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management
    
    Construction of a Monthly Banking Sector Fragility Index In this section, we begin to describe how a monthly BSF index can be constructed, and how it can be used to decide whether a national banking system is/was in crisis at a particular point in time. The brief discussions in the previous sections indicate that there are mainly three7 leading sectoral indicators of banking crises, which may be used in construction of a BSF index: (i) bank deposits, (ii) bank claims on (or credits to) the domestic private sector, and (iii) foreign liabilities of banks. These three variables are proxies, or indirect indicators, of changes in the liquidity risk, credit risk and exchange rate risk in the
    
    3. Banks’ Unhedged Foreign Liabilities, Devaluation, and Exchange-Rate Risk Kaminsky and Reinhart (1999) present one of the broadest frameworks to discuss the potential links between banking and
    
    53
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management
    
    banking sector, respectively. In other words, the fluctuations in these indicators are supposed to represent the changes in the fragility of banking sector in any country. Therefore, considering the economic risks related to banks’ balance sheets, we propose the following general index (BSF3) to measure the fragility of banks to crisis by using monthly banking sector data:
    
    actually corresponds to a large extent to the excessive risk-taking behavior of banks, and hence, to an early period of increasing possibility of crisis in the banking sector. This early warning phase of any approaching crisis is then followed by a rapid decrease in the value of the BSF3, which in turn can be associated with substantial falls (i) in bank deposits (bank withdrawals), (ii) in claims to private sector (as a response to significant increases in non-performing loans), and/or (iii) in foreign liabilities (particularly in the face of an actual or potential
    
    (1)
    
    depreciation in the domestic currency). In this sense, it is obvious that a coincidence of these three events would enhance the severity of the impending banking sector problem. The sudden change in the pattern of risk-taking behavior of banks, or the
    
    where (2)
    
    substantial fall in the BSF3 following an enormous increase, may be triggered by a country-specific event, such as a political scandal or individual failure of a major bank.
    
    (3) Every fall in the BSF index, on the other hand, does not necesand (4) sarily imply that a banking system is moving into a deep systemic crisis. Therefore, we differentiate here between medium and high fragility episodes by defining two arbitrary thresholds. In this study, a national banking system is supposed to be in a medium fragility period, if the value of the BSF3 index is In equation (1), the BSF3 index is defined as an average of standardized8 values of CPS, FL and DEP, where µ and σ stand for the arithmetic average and standard deviation of these three variables, respectively. In equations (2), (3) and (4), LCPS, LFL and LDEP represent banking system’s total real claims on the private sector, the banks’ real foreign liabilities, and the total real deposits on banks, respectively. That is, CPS, FL and DEP are simply the corresponding annual changes in each and every one of these three variables. By using 12-month percent changes in the monthly data instead of using monthly changes, we avoid any seasonality, which may be incorporated into the data. We also hope to be kept away from the risk of deriving misleading interpretations, if we would consider simply month-to-month changes. Indeed, banking crises should be those types of far reaching financial difficulties that cannot be signaled simply by "monthly" fluctuations in banking variables, such as the bank deposits, claims on private sectors, or foreign liabilities. They must be caused by longer term and powerful deteriorations in the banking sector. The BSF3 index is proposed to measure the ups and downs in the domestic banking sector.9 Its mean for the sample period is equal to zero, as implied by equation (1) above. As long as the BSF3 does not deviate significantly from zero, historically there is no reason to expect a severe banking sector problem in the short run. Evidently every deep banking crisis is preceded The cyclical time pattern of a hypothetical banking crisis and its five successive stages described above are summarized in table 2 and illustrated in figure 1. In terms of the thresholds defined in equations (5) and (6), we expect now that banking crises which are identified in event-based studies mentioned above occur in high fragility periods determined by our estimations given below. Accordingly, a banking system is only accepted to be fully recovered from crisis when the BSF index reaches its sample period average (i.e., zero) again. If, however, the value of the BSF3 index is equal to or lower than –0.5, we assume that the relevant banking sector is highly fragile to systemic crisis: –0.5 ≥ BSF3t (6) between 0 and –0.5: 0 > BSF3t > -0.5 (5)
    
    54 by a relatively significant increase in the BSF3 index, which
    
    Table 2 Changes in the BSF Index and the Five Phases of a Hypothetical Baking Crisis
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking
    
    Banks’ Behavior
    
    Direction of the Change in the BSF Index increases significantly above zero suddenly begins to decrease
    
    Banking Fragility Banking Crisis falls * (optimistic, or boom, phase) starts to increase
    
    Probability of Approaching
    
    Phase 1
    
    excessively risk taking
    
    the probability starts to increase * it increases furthermore (probably panic arises) system is approaching the borderline to crisis most probably, a crisis occurs in this phase crisis is over if the BSF is very close or equal to zero again
    
    Management
    
    Phase 2
    
    generally risk avoiding
    
    Phase 3
    
    risk avoiding
    
    falls below zero (but it’s still above 0.5) falls below –0.5
    
    increases significantly (medium fragility) continues to increase (high fragility) it falls again (recovery period)
    
    Phase 4
    
    risk avoiding
    
    Phase 5
    
    gradually they start to take risk again
    
    increases towards zero
    
    * Although increases in the BSF index imply a fall in fragility in the short run, it actually must be interpreted as an alarming indicator for impending crisis, if the increase in the index is significant and continues for a while. Hence, the probability of crisis starts to increase in this initial phase, since banks’ take excessive risks during that period of time.
    
    Figure 1 Time Path of the BSF Index and Five Phases of a Hypothetical Banking Crisis
    Value of the BSF Index
    
    The BSF2 index above is simply calculated by omitting the role of changes in real bank deposits on banks’ financial fragility, and thus any deviation of the BSF2 from the BSF3 will help us in understanding the relative importance of bank runs in banking crises.
    
    Phase 1
    
    P h a s e 2
    
    P h a s e 3
    
    Phase 4
    
    P h a s e 5
    
    Empirical Results The BSF3 and BSF2 indices proposed above are calculated for
    BSF Index
    
    each of the selected 22 countries from which we know that they experienced systemic, or at least significant, banking sector
    Time
    
    0
    
    problems within the last three decades. To ensure the international comparability we use the International Monetary Fund’s International Financial Statistics (IFS) database (CD-ROM version, July 2003) as the common data source: LCPS is taken from IFS’s line 22D, while LFL is taken from line 26C. Finally,
    
    - 0.5
    
    Note: For interpretation of both changes in the BSF index and phases of crisis, see table 2 and the related part of the text. Clearly, it can be accepted that this recovery period starts in some cases as the BSF is increasing but is still below -0.5.
    
    LDEP is considered as the sum of lines 24 and 25 in the IFS. Notice that nominal series are deflated by using the corresponding domestic consumer price index (CPI). If the CPI data (IFS line 64), however, is not available for a particular country, the wholesale, or producer, price index (IFS line 63) is used to deflate the relevant nominal time series.
    
    Before we proceed to the presentation of empirical results, we shortly define an alternative index of banking sector fragility, BSF2, to test the idea that bank runs do not play a major role in modern banking crises:
    
    (1’)
    
    55
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003
    Country Standard Deviations in: Sample Period
    
    Table 3 Basic Characteristics of National Banking Systems According to the Components of the BSF Index
    Selected Banking 12-Month Percent Change in the Consumer Prices Index (LCPS / LFL) 2100.0 2025.5 1391.5 9.8 12.9 105.4 4.9 7.0 10.1 3.5 3.8 40.9 6.0 1345.5 3.7 21.5 10.6 4.0 7.8 Sector Ratios (period averages) Real Claims Real Total Real Foreign on Private Deposits / Liabilities/(Real Sector / Real Real Claims Claims on Foreign on Private Private Sector Liabilities Sector RealTotal Deposits) (LDEP / LCPS) 2.2 9.3 6.5 13.9 6.7 2.8 7.8 2.2 16.3 8.5 1.4 74.1 7.4 5.3 2.9 7.4 8.8 1.6 5.7 (LFL/ 0.9 0.8 0.9 0.6 1.1 1.1 0.9 1.2 1.3 0.9 1.4 1.3 1.2 1.4 1.3 1.3 0.7 1.0 0.9 (LCPS-LDEP)) 5.5 0.5 1.2 0.2 -1.4 -5.2 1.4 -2.2 -0.2 1.3 -1.9 0.0 -0.6 -0.5 -1.3 -0.4 0.4 -14.3 2.2
    
    Banking Management
    (CPS) 1 Argentina 2 Bolivia 3 Brazil 4 Chile 5 Indonesia 6 Israel 7 Japan 8 Jordan 9 Kenya 10 Malaysia 11 Malta 12 Mexico 13 Pakistan 14 Peru 15 Philippines 16 Poland Jan. 1982 - Dec. 2002 Jan. 1965 - Dec. 2002 June 1989 - Dec. 2002 Dec. 1979 - Dec. 2002 Jan. 1981 - Dec. 2002 Jan. 1982 - Dec. 2002 Sep. 1968 - Dec. 2002 Jan. 1977 - Dec. 2002 Jan. 1969 - Dec. 2002 May 1965 - Dec. 2002 Jan. 1965 - Dec. 2002 Jan. 1983 - Dec. 2002 Jan. 1965 - Dec. 2002 Jan. 1965 - Dec. 2002 Dec. 1987 - Dec. 2002 Jan. 1991 - Dec. 2002
    
    12-Month 12-Month 12-Month Percent Percent Percent Change in Change in Change in Real Claims on Real Foreign Real Total Private Sector Liabilities Deposits (FL) 17.7 42.3 23.7 14.2 22.4 6.1 5.5 11.2 13.1 7.9 14.8 30.9 8.5 25.2 15.5 12.6 16.1 7.9 11.0 (DEP) 35.8 77.2 85.6 51.3 88.5 9.7 23.3 44.7 45.3 30.8 98.5 53.8 33.4 100.3 34.6 28.4 40.7 15.2 33.8 (INF) 25.2 48.3 20.6 9.8 11.1 6.5 5.2 10.1 11.9 6.7 6.6 38.0 9.3 21.5 7.8 7.6 15.7 5.2 8.7
    
    17 South Korea Jan. 1968 - Dec. 2002 18 Sweden 19 Thailand 20 Trinidad and Tobago 21 Turkey 22 Venezuela Dec. 1965 - Dec. 2002 Jan. 1979 - Dec. 2002 Sep. 1968 - Dec. 2002 Jan. 1971 - Dec. 2000 Jan. 1961 - Dec. 2002
    
    11.4 20.1 19.7 16.3
    
    42.3 95.4 59.3 51.3
    
    9.4 15.4 15.4 14.4
    
    5.3 24.0 24.4 325.8
    
    16.3 3.7 48.6 11.8
    
    1.3 1.0 1.3 1.1
    
    -0.2 -23.6 -0.1 -1.8
    
    Sample Average
    
    Source: IMF, International Financial Statistics, CD-ROM version, July 2003; author’s own calculations.
    
    The list of countries considered in this study, corresponding sample periods imposed by the availability of reliable country data, and country-specific standard deviations in CPS, FL, DEP and inflation rates are all given in table 3. Note that the standard deviation figures in this table show that, for each and every one of the countries covered here, the FL variable has the highest volatility among the three variables.
    
    of FL is so low that we do not necessarily need to consider FL in the fragility index explicitly. Hence, we create a third version of the fragility index, BSF2*, by excluding the FL from the BSF3 index:
    
    (1’’) As also mentioned above, we use the standardized values of the three variables in construction of the BSF3 to avoid the possibility that one of the three components dominates the BSF3 index. Therefore, one may think that there is nothing wrong with the fact that the fluctuations in one of the variables are significantly higher than those of the others. However, after checking the three ratios calculated in the last three columns in Note that the BSF2* index is calculated only for those countries, which have relatively high LCPS / LFL and quite low LFL / (LCPS – LDEP) ratios (see table 3). These countries are Chile, Kenya, Mexico, Trinidad and Tobago, and Venezuela.
    
    56 table 3, we conclude that for some countries the absolute value
    
    Figure 2 Banking Sector Fragility in Argentina
    4.5 4.0 3.5 3.0 Jul. ’82 2.5 2.0 1.5 1.0 Sep. ’85 0.5 0.0 -0.5 -1.0 May ’86 -1.5 -2.0 Jul. ’83 -2.5 -3.0 -3.5
    Jan-82 Jan-83 Jan-84 Jan-85 Jan-86 Jan-87 Jan-88 Jan-89
    
    Figure 5 Banking Sector Fragility in Chile
    3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5
    Dec-79 Dec-80 Dec-81
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking
    Jan. ’96
    
    May ’89
    
    Jan. ’82
    
    Oct. ’92 May. ’94
    
    Mar. ’98 Feb. ’01
    
    Management
    Aug. ’98
    
    May ’90
    
    Dec. ’94 Nov. ’90 Nov. ’83
    Dec-82 Dec-83 Dec-84 Dec-85 Dec-86 Dec-87 Dec-88 Dec-89 Dec-90 Dec-91 Dec-92 Dec-93 Dec-94 Dec-95 Dec-96 Dec-97 Dec-98
    
    Oct. ’95 Dec. ’01
    
    Jun. ’00 Nov. ’02
    
    Dec-99
    
    Dec-00 Jan-01 Jan-01
    
    Dec-01 Jan-02 Apr. ’02 Jan-02
    
    May ’90
    Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02
    
    High Fragility
    
    BSF3
    
    BSF2
    
    BSF2*
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Figure 3 Banking Sector Fragility in Bolivia
    12.0 10.5 9.0 7.5 6.0 4.5 3.0 1.5 0.0 -1.5 -3.0
    Jan-65 Jan-67 Jan-69 Jan-71 Jan-73 Jan-75 Jan-77 Jan-79 Jan-81 Jan-83
    
    Figure 6 Banking Sector Fragility in Indonesia g g y
    4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 -3.0
    Jan-81
    
    Aug. ’86
    
    Aug. ’90 Jan. ’98
    
    Mar. ’83
    
    Oct. ’88 Jun. ’01
    
    Jul. ’69 Mar. ’78
    
    Sep. ’82 Mar. ’93
    
    Mar. ’84
    
    Mar. ’89
    
    Sep. ’91 Apr. ’02 Jun. ’99
    
    Aug. ’73
    
    Apr. ’80
    
    Aug. ’88 Aug. ’85
    Jan-85 Jan-87 Jan-89 Jan-91 Jan-93
    
    Jul. ’96
    Jan-95 Jan-97 Jan-99
    
    Feb. ’02
    Jan-01
    
    Jan-82
    
    Jan-83
    
    Jan-84
    
    Jan-85
    
    Jan-86
    
    Jan-87
    
    Jan-88
    
    Jan-89
    
    Jan-90
    
    Jan-91
    
    Jan-92
    
    Jan-93
    
    Jan-94
    
    Jan-95
    
    Jan-96
    
    Jan-97
    
    Jan-98
    
    Jan-99
    
    High Fragility
    
    BSF3
    
    BSF2
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Figure 4 Banking Sector Fragility in Brazil
    3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 Jul. ’90
    Jun-89 Dec-89 Jun-90 Dec-90 Jun-91 Dec-91 Jun-92 Dec-92 Jun-93 Dec-93 Jun-94 Dec-94 Jun-95 Dec-95 Jun-96 Dec-96 Jun-97 Dec-97 Jun-98 Dec-98 Jun-99 Dec-99 Jun-00 Dec-00 Jun-01 Dec-01 Jun-02 Dec-02
    
    Figure 7 Banking Sector Fragility in Israel
    
    Dec. ’82 Oct. ’83
    
    Jul. ’92
    
    2.0 1.5 Sep. ’94 Jan. ’99 Mar. ’01 1.0 0.5 0.0 -0.5 Aug. ’96
    
    Nov. ’93
    May. ’85 May ’87 Nov. ’87
    
    Oct. ’98
    
    Nov. ’85
    
    Jan. ’88
    
    -1.5 -2.0 -2.5
    
    Nov. ’83
    
    Feb. ’00
    
    Apr. ’02
    
    -1.0
    
    Jun. ’82
    
    Aug. ’96 Nov. ’94 Dec. ’89
    
    Jun. ’86 Dec. ’85
    Jan-85 Jan-86 Jan-87 Jan-88 Jan-89 Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00
    
    Jan-82
    
    Jan-83
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Jan-84
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Jan-00
    
    Dec-02
    
    57
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management
    2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0
    Sep-68 Sep-70 Sep-72 Sep-74 Sep-76 Sep-78 Sep-80
    
    Figure 8 Banking Sector Fragility in Japan
    Jul. ’87 May ’80 3.5 3.0 2.5 2.0 1.5 May ’96 Feb. ’70 Nov. ’81 Aug. ’77 Mar. ’94
    Sep-82 Sep-84 Sep-86 Sep-88 Sep-90 Sep-92 Sep-94 Sep-96
    
    Figure 11 Banking Sector Fragility in Malaysia
    
    Nov. ’71
    
    Feb. ’97 Oct. ’80 Dec. ’68 Dec. ’73 Dec. ’83 Sep. ’92
    
    Mar. ’01
    
    1.0 0.5 0.0 -0.5 -1.0
    
    Nov. ’82 Apr. ’70 Jan. ’66
    May-65 May-67 May-69 May-71 May-73 May-75 May-77 May-79 May-81 May-83 May-85
    
    Aug. ’99
    Sep-98 Sep-00 Sep-02
    
    Jan. ’75 Sep. ’87
    May-87 May-89 May-91
    
    -1.5 -2.0
    
    Jan. ’95 Jan. ’99
    May-93 May-95 May-97 May-99 Jan-99 Jan-01 May-01 Jan-02 Jan-01
    
    High Fragility
    
    BSF3
    
    BSF2
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Figure 9 Banking Sector Fragility in Jordan
    4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 Mar. ’79 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0
    
    Figure 12 Banking Sector Fragility in Malta g
    Aug. ’67 Mar. ’69 Jan. ’72 Oct. ’72 Jan. ’77
    
    g
    
    y
    
    Mar. ’92 May ’88
    
    Mar. ’95
    
    May ’01
    
    Mar. ’88
    
    Aug. ’80 Jan. ’78 Feb. ’86 Sep. ’89 Jun. ’97
    
    Jan. ’96 May. ’99
    
    May ’02
    
    Jan. ’78 Mar. ’68 Dec. ’73 Aug. ’65
    Jan-65 Jan-67 Jan-69 Jan-71 Jan-73 Jan-75 Jan-77 Jan-79 Jan-81 Jan-83 Jan-85 Jan-87 Jan-89 Jan-91
    
    Jan. ’95
    Jan-93 Jan-95 Jan-97 Jan-00
    
    Mar. ’02
    
    Jan-77 Jan-78 Jan-79 Jan-80 Jan-81 Jan-82 Jan-83 Jan-84 Jan-85 Jan-86 Jan-87 Jan-88 Jan-89 Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02
    
    High Fragility
    
    BSF3
    
    BSF2
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Figure 10 Banking Sector Fragility in Kenya g
    4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 -3.0
    Jan-69
    
    Figure 13 Banking Sector Fragility in Mexico
    4.0 3.5 Jan. ’96
    
    g
    
    y
    
    y
    Apr. ’90 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5
    Jan-83 Jan-84 Jan-85 Jan-86 Jan-99 Jan-01
    
    Dec. ’70 Jul. ’74
    
    Aug. ’77
    
    Sep. ’86 Feb. ’92
    
    Dec. ’94 Mar. ’85 Aug. ’98
    
    Jun. ’72 May ’75 Aug. ’82 Nov. ’90 Mar. ’94
    Jan-71 Jan-73 Jan-75 Jan-77 Jan-79 Jan-81 Jan-83 Jan-85 Jan-87 Jan-89 Jan-91 Jan-93 Jan-95 Jan-97
    
    Sep. ’00
    
    Nov. ’86
    Jan-87 Jan-88 Jan-89 Jan-90 Jan-91 Jan-92 Jan-93 Jan-94
    
    Jan. ’01 Feb. ’96
    Jan-95 Jan-96 Jan-97 Jan-98 Jan-99
    
    High Fragility
    
    BSF3
    
    BSF2
    
    BSF2*
    
    High Fragility
    
    BSF3
    
    BSF2
    
    BSF2*
    
    58
    
    Figure 14 Banking Sector Fragility in Pakistan
    3.0 2.5 2.0 1.5 1.0 0.5 0.0
    Jul. ’69
    
    Figure 17 Banking Sector Fragility in Poland g g y
    Oct. ’97 May. ’96 Jun. ’00 Nov. ’92 0.0 -0.5 Oct. ’93 Nov. ’97 Aug. ’02 -1.0 -1.5 -2.0
    Jan-91 Jul-91 Jan-92 Jul-92 Jan-93 Jul-93 Jan-94 Jul-94 Jan-95 Jul-95 Jan-96 Jul-96 Jan-97 Jul-97 Jan-98 Jul-98 Jan-99 Jul-99 Jan-00 Jul-00 Jan-01 Jan-02 Jul-02 Jul-01
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management
    Jun. ’02 Dec. ’98
    
    Mar. ’84 Dec. ’76 Jun. ’65
    
    2.0 Jul. ’86 1.5 1.0
    
    Dec. ’72
    
    Oct. ’96 Aug. ’99 Mar. ’83 Aug. ’88 Dec. ’90
    
    0.5
    
    Jun. ’66
    
    Sep. ’71
    
    -0.5 -1.0 -1.5 -2.0 -2.5
    
    Dec. ’80 Jul. ’74
    
    Jun. ’01 Oct. ’94 Nov. ’91
    
    Jan-65 Apr-66 Jul-67 Oct-68 Jan-70 Apr-7 Jul-72 Oct-73 Jan-75 Apr-76 Jul-77 Oct-78 Jan-80 Apr-81 Jul-82 Oct-83 Jan-85 Apr-8 Jul-87 Oct-88 Jan-90 Apr-91 Jul-92 Oct-93 Jan-95 Apr-96 Jul-97 Oct-98 Jan-00 Apr-01 Jul-02
    
    High Fragility
    
    BSF3
    
    BSF2
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Figure 15 Banking Sector Fragility in Peru
    
    Figure 18 Banking Sector Fragility in South Korea
    
    4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5
    
    Jun. ’74
    
    3.5 3.0 2.5 2.0 1.5 Sep. ’98 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0
    
    Mar. ’75 Jan. ’70
    Apr. ’72 May ’73
    
    May ’67 May ’82
    
    Jan. ’79 Sep. ’82 Sep. ’85 Oct. ’84 Jan. ’81 Jan. ’88
    
    Dec. ’91
    
    Jun. ’68
    
    Feb. ’75
    
    Jun. ’77
    
    Aug. ’83 Jan. ’89
    
    Feb. ’00
    
    Jan. ’76 Apr. ’74
    
    Jan. ’94 Nov. ’98
    
    Jan-65 Apr-66 Jul-67 Oct-68 Jan-70 Apr-71 Jul-72 Oct-73 Jan-75 Apr-76 Jul-77 Oct-78 Jan-80 Apr-81 Jul-82 Oct-83 Jan-85 Apr-86 Jul-87 Oct-88 Jan-90 Apr-91 Jul-92 Oct-93 Jan-95 Apr-96 Jul-97 Oct-98 Jan-00 Apr-01 Jul-02
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Figure 16 Banking Sector Fragility in Philippines
    2.5 2.0 1.5 1.0 Dec. ’90 Nov. ’93 Oct. ’00 Aug. ’94 Jul. ’88 Sep. ’91 Jan. ’99
    Dec-87 Jun-88 Dec-88 Jun-89 Dec-89 Jun-90 Dec-90 Jun-91 Dec-91 Jun-92 Dec-92 Jun-93 Dec-93 Jun-94 Dec-94 Jun-95 Dec-95 Jun-96 Dec-96 Jun-97 Dec-97 Jun-98 Dec-98 Jun-99 Dec-99 Jun-00 Dec-00 Jun-01 Dec-01 Jun-02 Dec-02
    
    3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0
    
    Nov. ’96
    
    0.5 0.0 -0.5 -1.0 -1.5 Oct. ’01 -2.0 -2.5 Nov. ’76 Mar. ’81 May ’75 Oct. ’83 Feb. ’86 Mar. ’92 Feb. ’94
    Jan-71 Apr-72 Jul-73 Oct-74 Jan-76 Apr-77 Jul-78 Oct-79 Jan-81 Apr-82 Jul-83 Oct-84 Jan-86 Apr-87 Jul-88 Oct-89 Jan-91 Apr-92 Jul-93 Oct-94 Jan-96 Apr-97 Jul-98 Oct-99 Jan-01 Apr-02
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Jan-68 Apr-69 Jul-70 Oct-71 Jan-73 Apr-74 Jul-75 Oct-76 Jan-78 Apr-79 Jul-80 Oct-81 Jan-83 Apr-84 Jul-85 Oct-86 Jan-88 Apr-89 Jul-90 Oct-91 Jan-93 Apr-94 Jul-95 Oct-96 Jan-98 Apr-99 Jul-00 Oct-01
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Figure 19 Banking Sector Fragility in Sweden
    
    Jul. ’88 Apr. ’90 Jan. ’72 Jun. ’74 Dec. ’79 Feb. ’83
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Mar. ’97 Nov. ’97
    
    Aug. ’91
    
    Nov. ’98 Sep. ’00
    
    Sep. ’99
    
    59
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management
    2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 Jul. ’66 Dec. ’72 Feb. ’71 Mar. ’64
    
    Figure 20 Banking Sector Fragility in Thailand g
    Jan. ’79 Jul. ’97 Jul. ’92 Mar. ’86 Jun. ’80 Jan. ’99 Sep. ’00
    
    Figure 23 Banking Sector Fragility in Venezuela g
    Mar. ’94 Nov. ’88 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 Dec. ’75 Jul. ’91 Feb. ’84 Jan. ’87
    
    g
    
    y
    
    g
    
    y
    Oct. ’97
    
    Mar. ’84
    
    Feb. ’01
    
    Jan. ’85 Feb. ’80 Nov. ’93 May ’89 Jun. ’99 Jun. ’96
    
    Jan-61 Apr-62 Jul-63 Oct-64 Jan-66 Apr-67 Jul-68 Oct-69 Jan-71 Apr-72 Jul-73 Oct-74 Jan-76 Apr-77 Jul-78 Oct-79 Jan-81 Apr-82 Jul-83 Oct-84 Jan-86 Apr-87 Jul-88 Oct-89 Jan-91 Apr-92 Jul-93 Oct-94 Jan-96 Apr-97 Jul-98 Oct-99 Jan-01 Apr-02
    
    High Fragility
    
    BSF3
    
    BSF2
    
    Source: IMF, International Financial Statistics, CD-ROM version, July 2003; author’s own calculations.
    
    Figure 21 Banking Sector Fragility in Trinidad & Tobago g g y g
    
    Note:
    
    4.0 3.5 3.0 Feb. ’70 2.5 Sep. ’76 2.0 Dec. ’72 Nov. ’82 1.5 Aug. ’67 1.0 0.5 0.0 -0.5 Nov. ’80 -1.0 Jul. ’68 -1.5 Apr. ’74 Sep. ’87 -2.0
    
    The national BSF3 and BSF2 indices - and if calculated, the
    Aug. ’97
    
    BSF2* index - are graphically shown in figures 2 to 23. The
    Oct. ’91
    
    episodes of medium and high banking-sector fragility are calMay ’00
    
    culated according to the above-described criteria by considering the BSF2 (or BSF2*) index. The country-specific high fragility periods determined according to equation (6) are marked by gray vertical bands in figures.
    
    Dec. ’92
    
    Dec-65 Mar-67 Jun-68 Sep-69 Dec-70 Mar-72 Jun-73 Sep-74 Dec-75 Mar-77 Jun-78 Sep-79 Dec-80 Mar-82 Jun-83 Sep-84 Dec-85 Mar-87 Jun-88 Sep-89 Dec-90 Mar-92 Jun-93 Sep-94 Dec-95 Mar-97 Jun-98 Sep-99 Dec-00 Mar-02
    
    High Fragility
    
    BSF3
    
    BSF2
    
    BSF2*
    
    Concluding Comparison and Final Remarks Figure 22 Banking Sector Fragility in Turkey
    3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0
    Jan-79 Jan-80
    
    In recent years, there has been a high interest in research on the timing, duration, causes, effects, and cures of banking crises. In this paper, we proposed a monthly, weighted banking-sector fragility (BSF) index that may easily be used to measure and monitor the changes in the banking sector fragility to crisis.
    Oct. ’00 Feb. ’01
    
    Aug. ’82
    
    Dec. ’82
    
    Aug. ’97 Feb. ’87 Nov. ’90 Jun. ’79
    Nov. ’91
    
    Oct. ’93
    
    Apparently, this type of index is capable of providing more information about the ups and downs in the banking sector with respect to certain crisis-years in event-based studies. Table 4 compares our findings shown in figures 2-23 with the
    
    Nov. ’83 May ’80
    Jan-81 Jan-82
    
    Jan. ’86 Sep. ’88
    
    Apr. ’94 Oct. ’94
    Jan-93 Jan-94 Jan-95 Jan-96 Jan-97
    
    Jul. ’99 Feb. ’02
    Jan-98 Jan-99 Jan-00 Jan-01 Jan-02
    
    results of major studies in the event-based tradition.
    
    Jan-83 Jan-84 Jan-85
    
    Jan-86 Jan-87
    
    Jan-88 Jan-89 Jan-90
    
    High Fragility
    
    Jan-91 Jan-92
    
    BSF3
    
    BSF2
    
    60
    
    Sep-68 Dec-69 Mar-71 Jun-72 Sep-73 Dec-74 Mar-76 Jun-77 Sep-78 Dec-79 Mar-81 Jun-82 Sep-83 Dec-84 Mar-86 Jun-87 Sep-88 Dec-89 Mar-91 Jun-92 Sep-93 Dec-94 Mar-96 Jun-97 Sep-98 Dec-99 Mar-01 Jun-02
    
    High Fragility
    
    BSF3
    
    BSF2
    
    BSF2*
    
    BSF3 and BSF2 are two alternative indices of banking sector fragility that are calculated as defined in equations (1) and (1’), respectively. For interpretation of changes in indices, please see table 2. Note that gray vertical bands in figures designate the periods of high fragility (see equation (6)) with respect to the BSF2 index. For Chile, Kenya, Mexico, Trinidad and Tobago, and Venezuela, we consider BSF2* instead of BSF2, as justified in section 4 above.
    
    Table 4 Episodes of Major Banking Crises and High Fragility in Selected Countries
    Caprio and Klingebiel (1996, 1999 2002 and 2003) Argentina 1980-1982 1980-1982 Lindgren, Garcia and Saal, (1996) Hardy and Pazarba?? o∂lu (1998) DemirgüçKunt and Detragiache (1997 and 1998) Kaminsky and Reinhart (1996 and 1999) Beginning of the Crisis Mar. 1980 May 1985 1989-1990 1995 2001-present Bolivia 1989-1990 Dec. 1994 Mar. 1995 July 1982 June 1989 1989-1990 1995-1997 1989 1995 1980-1982 1980 d.n.a. Oct. 1985 Jun. 1989 Jun. 1994 Mar. 2001 Apr. 1978 Oct. 1982 1986-1988 1986-1987** 1994-? Brazil 1990 1994-1999 1994-pres.** Dec. 1994 Mar. 1996 1994-pres.** Nov. 1985 Nov. 1985 1990 1994-1997 1990 1994 Oct. 1987 June 1988 1986-1987 1994-1997 Sep. 1986 April 1993 d.n.a. d.n.a. Apr. 1995 Feb. 1999 Chile 1976 1981-1986 1981-1987 1981-1987 Sep. 1981 Mar. 1983 1981-1987 1976 1981-1983 1976 1981 d.n.a. Feb. 1982 June 1990 Sep. 1998 Mar. 2002 Indonesia 1994* 1997-present Israel 1977-1983 1983-1984** 1983-1984 Oct. 1983 June 1984 1994-pres.** 1992* 1997 1992-1994 Nov. 1992 present 1994 1997 1992 1997-1998 1977 Sep. 1990 Feb. 1998 d.n.a. Jan. 1983 Dec. 1987 Japan 1991-present 1992-pres.** 1992* 1992-1994 1992-pres 1992-1997 1992 Dec. 1971 Aug. 1987 June 1996 Jordan 1989-1990* 1989 1989-1990 1989-1990 Mar. 1977 Jun-88 Apr. 1995 Kenya Aug. 1974 Sep. 1977 1985-1989 1992 1993-1995 1996-?* Malaysia 1985-1988* 1997-present Malta no crisis between 1990-1995 Mexico 1981-1991 1982 1982 1994 1982 1994-1995 Sep. 1982 Oct. 1992 June 1984 Mar. 1996 1985-1988 1985 1985-1988 July 1985 Sep. 1997 Aug. 1986 present 1985-1988 1985-1988 1997 no crisis between 1992-1997 1981-1991 1995-1997 1981 1994 1985 1998 1993* 1993* 1993 1993-1995 1992-1997 Mar. 1992 Feb. 1996 Jan. 1984 Mar. 1997 Nov. 1972 Feb. 1977 Jun. 1999 Apr. 1985 Jan. 1995 Sep. 1998 Pakistan 1980-pres.** Jan. 1973 Aug. 1986 Sep. 1999 Peru June 1967 July 1974 1983-1990 1983-1990** 1981-1987 1981-1987 1981 1983* 1981-1987 1983-1990 Jan. 1981 Mar. 1983 June 1985 Apr. 1983 1981-1987 1983-1990 1981 1983 d.n.a.. June 1982 d.n.a.. Jan. 1991 1998-present 1997* July 1997 present 1997 Dec. 1996 1985-1989 1985-1989 Oct. 1986 Jul. 1983 May 1986 May 1990 Oct. 1995 Dec. 2001 Apr. 1980 Aug. 1985 Aug. 1988 Feb. 2002 d.n.a. July 1990 Aug. 1996 Feb. 2000 d.n.a. Nov. 1983 Nov. 1990 June 2000 Nov. 2002 Sep. 1991 June 1999 d.n.a. Dec. 1985 Dec. 1989 Aug. 1977 Mar. 1994 Aug. 1999 Jan. 1978 Sep. 1989 Jun. 1997 May 1975 Aug. 1982 Nov. 1990 Jan. 1983 - Apr. 1987 Aug. 1990 - July 1991 Feb. 2000 - Jan. 2001 Jul. 2002 - Dec. 2002 medium fragility Sep. 1998 - Dec. 2002 d.n.a. Nov. 1983 - Jun. 1986 Jan. 1988 - Aug. 1991 Dec. 1976 - July 1978 July 1991 - July 1995 June 1997 - Dec. 2002 Nov. 1977 - Mar. 1978 Dec. 1988 - June 1991 Sep. 1996 - Aug. 1998 Jan. 1975 - Jan. 1976 Oct. 1980 - July 1984 Mar. 1988 - Oct. 1988 Nov. 1989 - Apr. 1991 medium fragility Jan. 2000 - Mar. 2000 Jul. 1983 - Mar. 1985 Apr. 1986 - Nov. 1986 Oct. 1988 - Jan. 1992 medium fragility Sep. 2001 - Feb. 2002 Aug. 1979 - Oct. 1981 Mar. 1983 - Sep. 1985 May 1988 - Dec. 1989 Feb. 2000 - Jun. 2002 Peak of the Crisis Martinez Peria (2000) Glick and Hutchison (2000) Bordo and Eichengreen (2002) Current Study (the BSF2 , or BSF2* , index) Beginning of the Distress Date of Highest Fragility Episode of High Fragility (if applicable)
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management
    
    see the next row Mar. 1994 Sep. 2000 Sep. 1987 Jan. 1999 Dec. 1973 Jan. 1978 Mar. 2002 Nov. 1986 Feb. 1996 Jan. 2001 July 1974 Aug. 1988 Aug. 2002 June 1968 June 1977 Mar. 1990 - June 1994 Jun. 2000 - May 2001 Oct. 1986 - Mar. 1989 Apr. 1998 - Dec. 2002 Feb. 1973 - Sep. 1975 Nov. 1977 - Jan. 1980 Jul. 2001 - Jun. 2002 Dec. 1985 - Dec. 1988 Aug. 1995 - Dec. 1996 Aug. 2000 - Apr. 2001 June 1973 - June 1975 Apr. 1988 - Nov. 1988 Sep. 2001 - Dec. 2002 June 1968 - Nov. 1968 June 1977 - July 1979
    
    1994-1997 1994-pres.**
    
    Jan. 1989 July 1987 - Jan. 1991Philippines d.n.a. Sep. 1991 Jan. 1999 Mar. 1991 - Mar. 1992 Aug. 1998 - Dec. 2002
    
    61
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management
    Poland South Korea mid-1980s** 1997-present Sweden 1991 Thailand 1983-1987 1997-present Trinidad & Tobago 1982-1993* Early 19821993** 1983-1987 1983 1997 1983-1987 1990-1993 1992 1990-1993 Nov. 1991 Mar. 1979 Oct. 1983 May 1996 1997 Caprio and Klingebiel (1996, 1999 2002 and 2003) 1990s Lindgren, Garcia and Saal, (1996) Hardy and Pazarba?? o∂lu (1998) DemirgüçKunt and Detragiache (1997 and 1998)
    
    Table 4 Episodes of Major Banking Crises and High Fragility in Selected Countries (continued)
    Kaminsky and Reinhart (1996 and 1999) Beginning of the Crisis d.n.a.. June 1973 Feb. 1979 Oct. 1985 1997 1997-1998 Apr. 1997 Mar. 1983 Sep. 1992 Mar. 1979 June 1985 present 1983-1987 1997 1983 1990-1993 Aug. 1988 Feb. 1979 Apr. 1984 Apr. 1994 Jan. 1973 1982-1993 Dec. 1982 Nov. 1991 Turkey 1982-1985 1982 1982 1982-1985 1982 July 1979 Sep. 1982 Mar. 1987 1991 1994* 1994** 1991 1994-1995 Jan. 1991 Mar. 1991 1991 1994-1995 Dec. 1990 Nov. 1993 Sep. 1997 2000-present Venezuela Late 1970s & 1980s* 1978-1986 1980 Jan. 1976 Feb. 1987 1994-1995 1994 1993-1995 Oct. 1993 Aug. 1994 1994-1997 1993 Aug. 1991 Nov. 1997 Mar. 2001 Feb. 1980 May 1989 June 1996 June 1999 Dec. 2002? Oct. 1979 - Aug. 1980 Aug. 1987 - Mar. 1990 Jan. 1993 - Feb. 1997 Sep. 1998 - Jan. 2000 Feb. 2002 - Dec. 2002 Nov. 2000 Nov. 1991 Apr. 1974 Jan. 1981 Jan. 1988 Nov. 1998 Oct. 1983 Feb. 1994 June 1980 Mar. 1986 Sep. 2000 Apr. 1974 Sep. 1987 Dec. 1992 May 1980 Nov. 1983 Sep. 1988 Nov. 1991 Oct. 1994 July 1999 Feb. 2002 Jul. 1991 - Jun. 1995 July 1973 - Sep. 1974 Mar. 1980 - June 1980 Mar. 1987 - Apr. 1989 Mar. 1998 - Feb. 1999 Oct. 1983 - June 1984 Apr. 1991 - Feb. 1995 Dec. 1979 - Mar. 1981 Nov. 1985 - Nov. 1987 Feb. 1998 - Dec. 2002 Nov. 1973 - Oct. 1974 June 1983 - Feb. 1991 Apr. 1992 - Nov. 1994 Jan. 1979 - Nov. 1980 medium fragility Apr. 1988 - Oct. 1989 Nov. 1991 - Mar. 1992 Apr. 1994 - Apr. 1995 Mar. 1999 - Mar. 2000 June 2001 - Dec. 2002 Peak of the Crisis Martinez Peria (2000) Glick and Hutchison (2000) Bordo and Eichengreen (2002) Current Study (the BSF2 , or BSF2* , index) Beginning of the Distress Date of Highest Fragility Episode of High Fragility (if applicable)
    
    * Borderline, or non-systemic, banking crisis. ** Significant, or extensive, unsoundness short of a crisis. d.n.a.: Data not available.
    
    It should be noted that, in this table, we interpret the first month, when a country-specific fragility index started to decline before entering an episode of high fragility, as the first sign, or beginning date, of an approaching banking sector distress. Table 4 also shows both the specific months of highest fragility and episodes of high fragility, which are marked by gray vertical bands in figures from 2 to 23. Considering the comparative information presented both in figures 2-23 and in table 4, we briefly conclude that: (a) Both the definition of banking crisis and the identification of crisis episodes are essential, if one attempts to predict and explain banking crises empirically. Depending on the timing and duration of a crisis that is to be explained, the result of the analysis is expected to vary remarkably. The crisis episodes in most of the subsequent studies are primarily adapted from the information given in Caprio and Klingebiel (1996 or 2003) and/or Lindgren et al. (1996).
    
    crisis episodes given in these two studies and between the episodes considered in those studies that are also mentioned in table 4. The Mexican banking crisis in the 1980s, for example, is to be said by different researchers to occur in 1982, between 1982 and 1984, or in the 1981-1991 period. It is obvious that the result of an empirical analysis of the Mexican crisis will strongly depend on which year or years we initially assumed as crisis years. (b) Many studies in the literature (see table 4) do not differentiate between systemic and non-systemic (borderline) crises. However, the analysis of a banking crisis must also be affected by the initial assumption on the extent of the crisis. The BSF index proposed here not only captures crisis times in terms of the defined high fragility periods in this study, but it also roughly describes the whole development process of a banking sector problem, even if it is only a significant unsoundness short of a crisis.
    
    62
    
    However, there are some important differences between the
    
    (c) Overall, the high fragility episodes determined according to the BSF calculations in this study overlap with the crisis episodes mentioned in those studies that are considered in table 4. Moreover, medium fragility episodes dated here are to a large extent in accordance with borderline-crisis episodes mentioned in the literature. Clearly, the information content of changes in a monthly BSF index is signifi-
    
    cantly higher than that of the simple years of crisis that are identified based on country-specific information or relevant events. A monthly BSF index explicitly detects the ups and downs even within a single year, and hence, it eliminates the risk of labeling an entire year as crisis year even if the crisis has arisen, let’s say, only on the last two months of that year.
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management
    
    Table 5 Country-Specific Literature on Banking Sector Performance and Fragility
    Study Drees and Pazarbas,ıog (1995) ˘lu García-Herrero (1997) Country Coverage Finland, Norway, and Sweden Argentina, Paraguay, and Venezuela South Korea Sweden Period 1980s & 1990s 1990s Major Issues Macroeconomic determinants of banking crises; role of financial liberalization in financial crises Causes of banking crises stemming from both macroeconomic and bank-specific factors; macro economic effects of banking crises The sources of the 1997 Korean financial crisis, and the measures taken to deal with it Discussion of the 1985 deregulation and other causes of the banking crisis in early 1990s; the relation between the European Exchange-Rate Mechanism (ERM) crisis in 1992 and Swedish banking crisis Policy responses of Indonesia, Korea, and Thailand to the 1997 Asian crisis and comparison of their actions with those of Malaysia and the Philippines, which were buffeted by the crisis Causes of banking crisis; reasons for the unnecessary prolongation of the recovery process Effects of banks’ unbooked losses on banking stress; government corruption and duration of banking crises Possible effects of banking sector concentration on financial development, economic growth and banking sector fragility Macro- and microeconomic roots of international illiquidity in countries considered Korean stabilization and reform program implemented in response to the currency and banking crisis in 1997-98; recovery from deep recession; the lessons learned Rapid growth of non-performing loans; debt restructuring between creditor banks and borrowing corporate sector companies Chronological evaluation of developments in the weak of the 1997 banking crisis, and the effects of government policies on the recovery process Brazil’s January 1999 currency crisis; links between banking and currency crises The effects of the 1988-1992 financial sector reform Bonaccorsi di on the profitability and efficiency of the Pakistani banking system The chronology and causes of currency and banking crises in South Korea; recovery from a twin crisis The chronology of events and the policy responses by the authorities; identification of factors that explain why it has taken so long to bring the crisis under control; lessons learnt from the crisis Review of financial sector performances and quantitative analysis of bank soundness in the Caribbean Causes of the slowdown in bank credit to the private sector in the 1990s Effects of currency and interest-rate shocks on the vulnerability of the Indonesian banking system, measures taken to deal with it, and the lessons learned Recent disinflation attempt in Turkey and its negative effects on the baking sector fragility
    
    Baliño and Ubide (1999) Englund (1999)
    
    1993-1997 1980s & 1990s
    
    Lindgren et al. (1999)
    
    Indonesia, South Korea, Thailand, Malaysia, and the Philippines Japan African countries Chile East Asian and Latin American countries South Korea
    
    1997-1999
    
    Kanaya and Woo (2000) Kane and Rice (2000) Levine (2000) Chang and Velasco (2001) Chopra et al. (2001)
    
    1990s 1980-1999 1980-1999 1997-1998 1997-1998
    
    Duvan (2001) Enoch et al.. (2001)
    
    Turkey Indonesia
    
    1999-2001 1988-1999
    
    Gruben and Welch (2001) Hardy and Bonaccorsi di Patti (2001) Koo and Kiser (2001) Nakaso (2001)
    
    Brazil Pakistan South Korea Japan
    
    late 1990s 1980s & 1990s 1997-1998 1990s
    
    Worrell, Cherebin and Polius-Mounsey (2001) Barajas and Steiner (2002)
    
    Caribbean countries (incl. Trinidad and Tobago) Argentina, Bolivia, Brazil, Chile, Colombia, Mexico, Peru, and Venezuela Indonesia
    
    1990s 1960-2000
    
    Pangestu and Habir (2002)
    
    1990s
    
    Ertug ˘rul and Yeldan (2002)
    
    Turkey
    
    2000-2001
    
    63
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management
    
    (d) For an individual researcher, the interpretation, or justification, of variations in large numbers of national BSF indices is not an easy undertaking because it requires some degree of country expertise as well as additional microeconomic information related to the relevant sector. Thus, a certain group of studies in the financial crises literature (see table 5) can be considered as a benchmark to examine the chronological and institutional background of changes in the BSF index. The country-specific chronological explanations in those studies, which are listed in table 5, strongly support both the results of and the main motivation behind the current study. To be more precise, let us consider the case of the Swedish banking crisis in the early 1990s. The total length of this crisis spans from one year to four years, depending on what study is considered as a reference for crisis episodes (see table 4). Even the timing of the same crisis varies across the different studies.10 Figure 19 however implies that, after 1985, real bank deposits, banks’ foreign liabilities, and credits to the private sector all started to increase simultaneously, and that it reached a peak in July 1988. The BSF approach used here suggests that the following falls in the BSF2 (or BSF3) after this date can be interpreted as a serious increase in the Swedish banks’ vulnerability to crisis. The period from autumn of 1988 to autumn of 1990 is a period where all of the three components of the index were clearly decreasing. Now, according to Englund’s (1999) comprehensive analysis of the Swedish banking crisis, these occurrences can be justified and understood as follows: "Newly deregulated credit markets after 1985 stimulated a competitive process between financial institutions where expansion was given priority. Combined with an expansive macro policy, this contributed to an asset price boom. The subsequent crisis resulted from a highly leveraged private sector being simultaneously hit by three major exogenous events: a shift in monetary policy with an increase in pretax interest rates, a tax reform that increased after tax interest rates, and the ERM crisis. Combined with some overinvestment in commercial property, high real interest rates contributed to breaking the boom in real estate prices, triggering a downward price spiral resulting in bankruptcies
    3 2 1
    
    figure 19 in this study. Not to expand the extent of the current study unnecessarily, we prefer to restrict our countryspecific remarks here to the case of Sweden. But interested readers easily may examine the reliability of the BSF index proposed here by considering the explanations in countryspecific studies, such as those that are listed in table 5, among others. (e) As shown in figures 1-23, for many countries, the BSF3 and BSF2 curves appear to have followed a very similar pattern, roughly implying that bank runs in many countries may not have an important role in triggering banking crises. Nevertheless, for particular countries, such as Mexico, the developments in bank deposits must be closely watched to detect possible banking sector problems. To sum up, all of the discussions above show that the BSF index proposed in this study is highly helpful in monitoring and identifying the banking sector difficulties by using monthly data. Since the BSF index is reflecting the changes in the sectoral climate more precisely and timely, it significantly reduces the possibility that the crisis or high fragility episodes are misidentified, contrary to the case of event-based identification strategies. The BSF index presents the chance to be able to work with higher frequency data on banking crises. Its information content is significantly high. Therefore, in the future, studies that aim to empirically investigate the causes, timing and effects of banking crises can easily depart from the timeseries-based statistical approach developed here.
    
    Notes
    To construct an index of vulnerability to currency crisis, some researchers employ the difference between domestic and foreign interest rates, or percentage changes in domestic interest rates, while many others avoid using it because many developing countries do not have reliable interest-rate data. Caprio and Klingebiel frequently update their well-known table of episodes of systemic and borderline banking crises and publish it also on the web, i.e. http://www.worldbank.org. Note that some of the crises episodes mentioned in later versions of the table differ from those episodes which are given in earlier versions. That is, the bank’s net worth includes the capital contributed by the bank’s shareholders and accumulated profits from doing business as intermediary in financial markets. See Kaminsky and Reinhart (1996, 1999), Demirgüç-Kunt and Detragiache (1998, 1999, 2000), Kaminsky (1999), Hardy and Pazarba_ıo_lu (1998, 1999), the IMF’s World Economic Outlook (May 1998, Ch. 4), Hutchison (1999), Goldstein et al. (2000), Martinez Peria (2000), Bordo and Schwartz (2000), Gourinchas et al. (2001), Hutchison and Neuberger (2002), and Bordo and Eichengreen (2002).
    
    and massive credit losses. The government rescued the banking system by issuing a general guarantee of bank obligations. The total direct cost to the taxpayer of the salvage has been estimated at around 2 per cent of GDP." Englund’s explanations, which are only partially quoted here, perfectly illustrate the macroeconomic background
    4
    
    64
    
    behind the time path of the Swedish BSF curve shown in
    
    5
    
    6
    
    7
    
    8
    
    9
    
    That means, credit overexpansions may reflect fundamental improvements in investing opportunities that are beneficial to output growth in the long run. In this case, this may lead to output recession, if bank credit is the primary funding source of activities in the real economy as it is often observed in many developing economies (see Disyatat, 2001). We neglect here (real) interest rates (or real interest rate differential) as the fourth component of the BSF index because many developing countries do not have internationally comparable and continuous time series on market-determined interest rates. One may also argue that interest-rate-risk (i.e., difficulties in maturity transformation) actually is indirectly considered in calculations here by the mean of deposits, claims, or foreign liabilities. By using the standardized values of CPS, FL and DEP, we equalize the variance of the three components, and thus avoid the possibility that any one of three components dominates the BSF3 index. Apparently, Hawkins and Klau’s (2000) index of banking system vulnerability is the most similar one to the BSF index proposed here. The authors use the following five proxies to measure the banking system vulnerability, by departing from the suggestion that banking crises are typically preceded by overvalued exchange rates, inadequate international reserves, recessions, high real interest rates, and excessive credit growth: (i) the rate of growth of domestic bank credit, (ii) the growth of borrowing from international banks, (iii) the external borrowing by banks as a percentage to domestic credit, (iv) the level of real interest rates, and (v) "stand-alone" credit ratings of the leading banks. Their index, however, differs from the BSF3 index proposed here in many aspects. It is calculated for 24 emerging market economies but the sample period is limited to 1996-1998 for quarterly data frequency. The BSF3 index, on the other hand, aims to cover a broader period of time for another set of 22 countries, including also a few developed market economies, and the data frequency is decided to be monthly. Furthermore, the Hawkins-Klau index is based on a weighted scoring methodology, contrary to the calculation methodology of the BSF3 index, which is actually similar to that of the FEMP index that is used to measure the pressure in foreign currency market. Note that Caprio and Klingebiel (2003) accept the Swedish banking crisis as occurred in 1991, although they considered the whole 19911994 period as the crisis episode in an earlier version of their useful table of banking crises.
    
    - Boyd, J. H., P. Gomis, S. Kwak, and B. D. Smith (2001). A User’s Guide to Banking Crises. Internet:
    www.eco.utexas.edu/facstaff/Smith/bankingcrisestextnew.pdf, manuscript.
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management
    
    10
    
    References
    - Baliño, T. J. T., and A. Ubide (1999). The Korean Financial Crisis of 1997 – A Strategy of Financial Sector Reform. Washington, DC: International Monetary Fund, Working Paper No. 99/28. - Barajas, A., and R. Steiner (2002). Credit Stagnation in Latin America. Washington, DC: International Monetary Fund, Working Paper No. 02/53. - Bordo, M., and A. J. Schwartz (2000). Measuring Real Economic Effects of Bailouts: Historical Perspectives on How Countries in Financial Distress Have Fared With and Without Bailouts. Cambridge, MA: National Bureau of Economic Research Working Paper Series, No. 7701. - Bordo, M., and B. Eichengreen (2002). Crises Now and Then: What Lessons from the Last Era of Financial Globalization? Cambridge, MA: National Bureau of Economic Research Working Paper Series, No. 8716. S,
    
    - Caprio, G., and D. Klingebiel (1996). Bank Insolvencies: CrossCountry Experiences. Washington, DC: The World Bank, Policy Research Working Paper No. 1620. - Caprio, G., and D. Klingebiel (1999). Episodes of Systemic and Borderline Financial Crises. Washington, DC: The World Bank, manuscript. - Caprio, G., and D. Klingebiel (2002). Episodes of Systemic and Borderline Banking Crises. In: D. Klingebiel and L. Laeven, eds., Managing the Real and Fiscal Effects of Banking Crises. Washington, DC: The World Bank, Discussion Paper No. 428, p. 31-49. - Caprio, G., and D. Klingebiel (2003). Episodes of Systemic and Borderline Financial Crises. Washington,DC: The World Bank, manuscript. - Calvo, G. A., L. Leiderman, and C. M. Reinhart (1994). The Capital Inflows Problem: Concepts and Issues. Contemporary Economic Policy, 12 (3): 54-66. - Chang, R., and A. Velasco (2001). The 1997-98 Liquidity Crisis: Asia Versus Latin America. Central Bank of Chile, Working Papers No. 120. - Chopra, A., K. Kang, M. Karasulu, H. Liang, H. Ma, and A. Richards (2001). From Crisis to Recovery in Korea: Strategy, Achievements, and Lessons. Washington, DC: International Monetary Fund, Working Paper No. 01/154. - Demirgüç-Kunt, A., and E. Detragiache (1998). The Determinants of Banking Crises: Evidence from Developing and Developed Countries. IMF Staff Papers 45 (1): 81-109.. Formerly printed as: International Monetary Fund, Working Paper No. 97/106, 1997, Washington, DC. - Demirgüç-Kunt, A., and E. Detragiache (1999). Financial Liberalization and Financial Fragility. In: Pleskovic, B., and J. E. Stiglitz, eds., Annual World Bank Conference on Development Economics 1998. Washington, D.C.: World Bank. Formerly printed as: International Monetary Fund, Working Paper No. 98/83, 1998, Washington, DC. - Demirgüç-Kunt, A., and E. Detragiache (2000). Monitoring Banking Sector Fragility: A Multivariate Logit Approach. The World Bank Economic Review, 14 (2): 287–307. - Dress, B., and C. Pazarba_ıo_lu (1995). The Nordic Banking Crises: Pitfalls in Financial Liberalization? Washington, DC: International Monetary Fund, Working Papers, No. 95/61. (http://www.imf.org/external/pubs/cat/longres.cfm?sk=1138.0) Also published as: IMF Occasional Paper, No. 161, 1998. - Disyatat P. (2001). Currency Crises and the Real Economy: The Role of Banks. Washington, DC: International Monetary Fund, Working Paper No. 01/49. - Duvan, B. (2001). Türk Bankacılık Sisteminde Tahsili Gecikmi_ Alacakların Tasfiyesi ve Sirket Borçlarının Yapılandırılması , (= Resolution of Non-Performing Loans in the Turkish Banking System, and Restructuring of Corporate Sector Debts). Ankara: State Planning Organization. - Edison, Hali J. (2000). Do Indicators of Financial Crises Work? An Evaluation of An Early Warning System. Board of Governors of the Federal Reserve System, International Finance Discussion Papers, No. 675. - Eichengreen, B., and C. O. Arteta (2000). Banking Crises in Emerging Markets: Presumptions and Evidence. Berkeley, CA:
    
    65
    
    The Arab Bank REVIEW Vol. 5, No. 2 October 2003 Banking Management -
    
    -
    
    University of California at Berkeley, Center for International and Development Economics Research Working Paper C00-115. Forthcoming in: Blejer, M., and M. Skreb, eds., Exchange Rate and Financial Vulnerability in Emerging Markets, The MIT Press. Eichengreen, B., and A. Rose (1998). Staying Afloat When the Wind Shifts: External Factors and Emerging-Market Banking Crises. Cambridge, MA: National Bureau of Economic Research, Working Paper Series No. 6370. Englund, P. (1999). The Swedish Banking Crisis: Roots and Consequences. Oxford Review of Economic Policy, 15 (3): 80-97. Enoch, C., B. Baldwin, O. Frécaut, and A. Kovanen (2001). Indonesia: Anatomy of a Banking Crisis – Two Years of Living Dangerously, 1997-99. Washington, DC: International Monetary Fund, Working Paper No. 01/52. ˘ Ertug rul, A., and E. Yeldan (2002). On the Structural Weaknesses of the Post-1999 Turkish Dis-Inflation Program. Ankara: Bilkent University, manuscript, internet: http://www.bilkent.edu.tr/~yeldane/Er&y2002.pdf. Frydl, E. (1999). The Length and Cost of Banking Crises. Washington, DC: International Monetary Fund, Working Paper No. 99/30. García-Herrero A. (1997). Banking Crises in Latin America in the 1990s: Lessons from Argentina, Paraguay, and Venezuela. Washington, DC: International Monetary Fund, Working Paper No. 97/140. Goldstein, M., G. L. Kaminsky, and C. M. Reinhart (2000). Assessing Financial Vulnerability: An Early Warning System for Emerging Markets. Washington, DC: Institute for International Economics. Gourinchas, P.-O., R. Valdes, and O. Lenderretche (2001). Lending Booms: Latin America and the World. Cambridge, MA: National Bureau of Economic Research, Working Paper Series, No. 8249. Gruben, W. C., and J. H. Welch (2001). Banking and Currency Crisis Recovery: Brazil’s Turnaround of 1999. Federal Reserve Bank of Dallas, Economic and Financial Review, Fourth Quarter, 12-23. Hardy, D., and E. Bonaccorsi di Patti (2001). Bank Reform and Bank Efficiency in Pakistan. Washington, DC: International Monetary Fund, Working Paper Series No. 01/138. Hardy, D., and C. Pazarba_ıo_lu (1998). Leading Indicators of Banking Crises: Was Asia Different? Washington, DC: International Monetary Fund, Working Paper Series No. 98/91. ˘ Hardy, D., and C. Pazarbas,ıog lu (1999). Determinants and Leading Indicators of Banking Crises: Further Evidence. Washington, DC: International Monetary Fund Staff Papers, 46 (3): 247-258. Hawkins, J., and M. Klau (2000). Measuring Potential Vulnerabilities in Emerging Market Economies. Basel: Bank for International Settlements, Working Papers, No. 91. Hutchison, M. M. (1999). Determinants of European Banking Distress: Evaluating the Risks from EMU and Financial Integration. Prepared for the workshop "Structural Change and European Economic Integration" held at Liseleje, Denmark, September 10-11, 1999, manuscript. Hutchison, M. M., and I. Neuberger (2002). How Bad Are Twins? Output Costs of Currency and Banking Crises. Santa Cruz, CA: University of California, manuscript. International Monetary Fund (1998). World Economic Outlook (Ch. IV: Financial Crises: Characteristics and Indicators of Vulnerability). No. 5: 74-97. International Monetary Fund (2002). International Financial Statistics – Database and Browser; CD-ROM. Washington, DC: IMF.
    
    -
    
    -
    
    - Kaminsky, G. (1999). Currency and Banking Crises: The Early Warnings of Distress. Presented at the IMF Institute course on An Early Warning System for Financial Crises, May 4, 1999, manuscript. - Kaminsky, G. and C. Reinhart (1996). Banking and Balance-ofPayments Crises: Models and Evidence. Washington, DC: Board of Governors of the Federal Reserve System, manuscript. - Kaminsky, G. and C. Reinhart (1999). The Twin Crises: The Causes of Banking and Balance-of-Payments Problems. American Economic Review, 89 (3): 473–500. Formerly published as: Board of Governors of the Federal Reserve System, International Finance Discussion Papers No. 544, 1996, Washington, DC. - Kanaya, A., and D. Woo (2000). The Japanese Banking Crisis of the 1990s. Washington, DC: International Monetary Fund, Working Paper No. 00/07. - Kane, E. J., and T. Rice (2000). Bank Runs and Banking Policies: Lessons for African Policymakers. Chestnut Hill, MA: Boston College, manuscript, internet: www2.bc.edu/~kaneeb/BankRuns.pdf. - Koo, J., and S. L. Kiser (2001). Recovery from a Financial Crisis: The Case of South Korea. Federal Reserve Bank of Dallas, Economic and Financial Review, Fourth Quarter, 24-36. - Levine, R. (2000). Bank Concentration: Chile and International Comparisons. Central Bank of Chile, Working Papers No. 62. - Lindgren, C.-J., G. Garcia, and M. I. Saal (1996). Bank Soundness and Macroeconomic Policy. Washington, D.C.: IMF. - Lindgren, C. J., T. J. T. Baliño, C. Enoch, A.-M. Gulde, M. Quintyn, and L. Teo (1999). Financial Sector Crisis and Restructuring: Lessons from Asia. Washington, DC: International Monetary Fund, Occasional Paper No. 188. - Martinez Peria, M. S. (2000). The Impact of Banking Crises on Money Demand and Price Stability. Washington, DC: The World Bank, Policy Research Working Paper No. 2305. - Nakaso, H. (2001). The Financial Crisis in Japan During the 1990s: How the Bank of Japan Responded and the Lessons Learnt. Basel: Bank for International Settlements, BIS Papers No. 6. - Obstfeld, M. and K. Rogoff (1995). The Mirage of Fixed Exchange Rates. Journal of Economic Perspectives, 9 (4): 73-96. - Pangestu, M., and M. Habir (2002). The Boom, Bust, and Restructuring of Indonesian Banks. Washington, DC: International Monetary Fund, Working Paper No. 02/66. - Sachs, J., A. Tornell, and A. Velasco (1996). Financial Crises in Emerging Markets: The Lessons from 1995. Brookings Papers on Economic Activity, 0 (1): 147–98. - Worrell, D., D. Cherebin, and T. Polius-Mounsey (2001). Financial System Soundness in the Caribbean: An Initial Assessment. Washington, DC: International Monetary Fund, Working Paper No. 01/123.
    
    -
    
    -
    
    -
    
    -
    
    -
    
    -
    
    -
    
    -
    
    -
    
    ˘ * Aykut Kibritçioglu is Associate Professor at the Department of Economics, Faculty of Political Sciences, Ankara University.
    
    -
    
    -
    
    66

Add Comment

To add a comment, you need to login or signup (you can signup without leaving this page)