## 1. Banking Crises Determinants

## Source details

**Canonical URL:** [1. Banking Crises Determinants](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp0596.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp0596.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp0596.pdf.json)

---

### Introduction
- Paper surveys literature on causes and consequences of bank fragility and highlights directions for future research.
- Organisation:
  - Review of basic facts about the recent wave of financial crises.
  - Presentation of two basic methodologies in cross-country empirical studies of banking-crisis determinants.
  - Discussion of crisis prediction applications.
  - Review of factors contributing to bank fragility.
  - Survey of economic effects of banking crises.
  - Conclusions and directions for further research.

### The resurgence of financial instability in the 1990s
- Historical context:
  - Post–World War II: banking crises were rare; calm macroeconomic environment, low inflation, pervasive controls on international capital flows, and restrictive supervisory powers contributed to stability.
  - Breakdown of Bretton Woods and the 1970s: macroeconomic turbulence increased but banking sectors remained largely sound in most countries.
  - Early 1980s financial liberalization and high real interest rates: several crises in Latin America and other developing countries.
- Notable episodes and consequences:
  - Chile: linked to banking system and privatization (Diáz-Alejandro, 1985).
  - U.S. S&L debacle: erosion of bank capital after liberalization, generous deposit insurance, ineffective regulation; fiscal cost borne by taxpayers, but negligible macroeconomic effects.
  - 1990s episodes with major macroeconomic consequences:
    - Scandinavian banking crises: currency devaluation, falling asset prices, banking crises and economic slowdown.
    - Japan: collapse of asset price bubble rendered most of the banking sector insolvent; regulatory forbearance and lax monetary policy prolonged balance sheet repair for more than a decade; growth halted in 1992 and had yet to fully recover by time of writing.
    - Mexico (began December 1994): cost of bailing out the banks reached almost 20 percent of GDP; bank credit to the private sector and economic growth remained lackluster after rescue.
    - East Asian crises (1997–8): economies with sound public finances and high growth were severely affected within months as banks buckled, depositors lost confidence, asset prices collapsed, and foreign capital inflows evaporated.
- Systematic assessment:
  - IMF and World Bank surveys (1996) found that a full three-quarters of their membership experienced significant banking problems during 1980–96.
  - Problems varied: isolated bank insolvencies; loss‑making government-owned institutions needing chronic recapitalization; pervasive weaknesses across regions and development levels.
- Implication: banking fragility is pervasive and multifaceted, warranting systematic empirical investigation.

### Two econometric approaches to identifying determinants of banking crises

H3: The signals approach
- Origin and application:
  - Signals approach developed for business-cycle turning points; first applied to banking crises by Kaminsky and Reinhart (1999) focusing on "twin crises" (simultaneous currency and banking crises).
- Data and definitions:
  - Sample: twenty industrial and emerging countries during 1970–95.
  - Currency crises: identified based on an index of market turbulence.
  - Banking crisis onset assumed to coincide with depositor runs leading to closure/takeover of banks, or with large-scale government intervention that spreads across the financial system.
- Incidence and co-movement:
  - Currency crises more frequent than banking crises in the sample: 76 episodes versus 26.
  - Of these, 19 episodes are twin crises; a wide majority of banking crises are accompanied by an exchange rate crash.
  - Caveat: sample selection (countries with fixed or heavily managed exchange rates) may overemphasize exchange rate importance.
- Pre-crisis behavior (24 months preceding crises):
  - Monetary growth and interest rates (lending and deposit rates) are above normal.
  - Export growth appears below trend before banking crises.
  - Real exchange rate is appreciating.
  - Real output growth falls below trend about eight months before the peak of the banking crisis.
  - Stock prices peak at about the same time as onset, suggesting banking crises are preceded by a cyclical downturn.
- Formal signals methodology:
  - Each variable’s behavior in the 24 months prior to crisis contrasted with “tranquil” times.
  - A variable signals a crisis when it crosses a threshold; a signal is correct if a crisis occurs within the next 24 months.
  - Thresholds chosen to minimize in-sample noise-to-signal ratio (authors use an “adjusted” noise-to-signal ratio: probability of false alarms divided by one minus probability of missing a crisis).
  - Performance metrics: Type I error (probability of missing a crisis), Type II error (probability of false signal), noise-to-signal ratio, and probability of a crisis conditional on a signal.
- Key signal results:
  - Indicators with lowest noise-to-signal ratio and highest probability of crisis conditional on the signal: appreciation of the real exchange rate, followed by equity prices and the money multiplier.
  - These indicators have large Type I error: they fail to issue a signal in 73–79 percent of the observations during the 24 months preceding a crisis.
  - Type II error (false signals) is much lower, ranging between 8 and 9 percent.
  - The real interest rate has the lowest Type I error, signaling in 30 percent of pre-crisis observations.
  - Indicators reflecting real sector developments are more closely associated with banking crises than currency crises.
  - Twin crises are preceded by more acute warning signs than individual crises and have more protracted adverse effects.

H3: The multivariate logit approach
- Motivation and advantages:
  - Signals approach treats covariates in isolation and ignores magnitude information relative to thresholds.
  - Multivariate logit models crisis probability as a function of a vector of explanatory variables, aggregating information consistently.
- Model specification:
  - Dependent variable: binary (zero if no crisis, one if crisis).
  - Probability P(i,t) modeled as F(X(i,t)' β) with F logistic; coefficients affect ln(P/(1–P)).
  - Increase in probability depends on original probability and initial values of independent variables and coefficients.
- Methodological issues:
  - Observations during crisis episodes typically excluded because explanatory variables may be affected by the crisis (feedback effects).
  - Construction of banking crisis dummy:
    - Economies in transition excluded.
    - Episodes identified using Caprio and Klingebiel (1996), Lindgren and others (1996), and case studies.
    - To be classified as a full-fledged crisis at least one condition must hold:
      - Ratio of nonperforming assets to total assets in the banking system exceeded 10 percent.
      - Cost of the rescue operation was at least 2 percent of GDP.
      - Banking sector problems led to a large scale nationalization of banks.
      - Extensive bank runs took place or emergency measures (deposit freezes, prolonged bank holidays, or generalized deposit guarantees) were enacted by the government.
- Sample and estimation notes:
  - Baseline regressions extended through 2002 and to include more countries; number of crisis episodes in baseline rose from 31 to 77.
  - Model estimated without country fixed effects to include noncrisis countries as controls; errors allowed to be correlated within each country by clustering the errors by country.

### Empirical findings from multivariate logit specifications
- Macroeconomic correlates:
  - Low GDP growth, high real interest rates, and high inflation are significantly correlated with occurrence of a banking crisis: crises tend to occur during periods of weak economic growth and loss of monetary control.
  - Exposure to real interest rate risk is a source of banking fragility; higher and more volatile real interest rates during the 1980s and 1990s may have contributed to greater incidence of banking crises.
  - Changes in terms of trade and exchange rate depreciation are not significant.
  - Fiscal variable (budget surplus scaled by GDP) has a positive coefficient but is significant only when deposit insurance is omitted.
- Banking sector and external vulnerability:
  - Ratio of broad money to foreign exchange reserves (measuring vulnerability to a run on the currency) enters positively and significantly.
  - Credit to the private sector enters with a positive sign, indicating greater bank exposure to private borrowers is associated with vulnerability.
  - High lagged credit growth, capturing a credit boom, is significantly and positively correlated with crisis probability in all specifications.
- Institutional variables:
  - Level of development (GDP per capita) is negatively correlated with systemic banking sector problems.
  - Presence of an explicit deposit insurance scheme appears to be a risk factor, likely because moral hazard outweighs reductions in self-fulfilling panics.

### IV. Using econometric models of banking crises as early warning systems
- Need: improve monitoring of financial vulnerabilities after 1990s crises.
- Proposed leading indicators and findings:
  - Credit growth to detect credit booms.
  - Equity price declines highlighted as important.
  - Ratio of broad money to foreign exchange reserves suggested as a vulnerability indicator.
- Selected studies and findings:
  - Hoonah (1997): sample of eighteen crisis and six noncrisis countries; divides crisis countries into macroeconomic, microeconomic, and government-behavior types; links high borrowing and central bank lending to government-intervention crises; links high loan-to-deposit ratios, high foreign borrowing-to-deposit ratios, and high credit growth to macroeconomic-origin crises.
  - Rojas-Suarez (1998): adapts CAMEL early warning system to emerging markets; better at identifying weak banks within a system than systemic crises; requires detailed bank-level data.
  - Kaminsky and Reinhart (signals): composite index of indicators crossing thresholds can outperform single indicators in-sample but may perform worse at predicting tranquil observations.
  - Demirgüç-Kunt and Detragiache (2000): multivariate logit yields lower in-sample Type I and Type II errors than Kaminsky and Reinhart signals; propose two monitoring frameworks using forecasted probabilities:
    - Threshold-trigger framework: choose probability threshold to act by minimizing loss trading off costs of acting when no crisis vs. not acting when crisis.
    - Rating framework: rate banking-system fragility with ratings interpreted in terms of crisis probability.
  - Application to six crisis episodes (Jamaica, Indonesia, Korea, Malaysia, Philippines, Thailand):
    - Both actual and forecasted data would have indicated high vulnerability in Jamaica.
    - For Asian countries, assessments were more reassuring because expectations of continued strong growth and stable exchange rates offset negative impacts of high real interest rates and past credit expansion.
- Limits and avenues for improvement:
  - Econometric monitoring tools show limited out-of-sample success; reasons include new crises differing from past crises and rarity of banking crises (few in-sample data points).
  - Suggested improvements:
    - Develop alternative scenarios with high and low forecasts for explanatory variables.
    - Use stress-testing exercises (used in Financial Sector Assessment Programs).
    - Explore high-frequency variables (interbank spreads, bank commercial paper spreads, stock market valuation of banks, corporate vulnerability) before crises—requires significant data collection.

### A. Individual bank measures of fragility and systemic crises
- Early literature uses bank balance-sheet and market information to explain and forecast individual bank failure.
- Findings:
  - Nonperforming loans and capital asset ratios often deteriorate rapidly before bank failure.
  - Little evidence that individual bank failure is driven by overall banking sector fragility (limited contagion in some contexts).
  - CAMEL variables reasonably predict distress; big institutions more likely to become distressed but less likely to be closed; connected institutions more likely to experience trouble.
  - Systemic crises reflect both exogenous shocks and prior individual-bank weaknesses.

### B. Financial liberalization and crises
- Theory: liberalization gives banks greater opportunities to take risk; without strong institutions and prudential regulation, liberalization may increase fragility.
- Empirical findings:
  - Demirgüç-Kunt and Detragiache (1998): crises more likely in liberalized financial systems, controlling for other characteristics.
  - Effect mitigated by strong institutional environment (rule of law, low corruption, good contract enforcement).
  - Subsequent studies find financial liberalization can significantly increase bank fragility.

### C. International shocks, exchange rate regime, and crises
- International shocks (global interest rates, OECD growth) influence bank fragility in developing countries; evidence mixed over time.
- Exchange rate regime effects:
  - Flexible exchange rates may stabilize the financial system by absorbing shocks and discouraging over-borrowing in foreign currency.
  - Fixed exchange rates may increase susceptibility to bank runs but may also discipline policymakers and reduce risk-taking.
- Empirical findings:
  - Arteta and Eichengreen (2002): fixed and flexible exchange rate countries equally susceptible to banking crises.
  - Domaç and Martinez-Peria (2003): adopting a fixed exchange rate diminishes likelihood of crisis in developing countries, but economic cost of a crisis is larger under a fixed exchange rate.
  - Arteta (2003): no evidence that deposit and credit dollarization increases fragility in a large sample.
  - De Nicolo, Honohan and Ize (2003): dollarization positively related to average Z-scores and non-performing loans.

### D. Bank ownership and structure and crises
- State ownership:
  - Greater state ownership associated with reduced competition, poorer productivity, lower growth, and higher probability of a banking crisis in 1980–97.
- Foreign ownership:
  - Foreign entry improves operating efficiency and financial intermediation.
  - Empirical evidence does not support fears that foreign banks flee during crises; foreign banks often had stronger and less volatile loan growth during and after crises in several cases.
  - Performance of foreign banks during crises can depend on origin (e.g., non‑regional vs. regional foreign banks).
- Competition and concentration:
  - More concentrated banking systems, fewer regulatory restrictions, and institutions encouraging competition are associated with fewer crises; concentration result may reflect better risk diversification by larger banks.

### E. The role of institutions
- Institutional development:
  - Weaker institutions (lower GDP per capita, weaker law and order) are related to higher probability of banking crises.
  - Low transparency/high corruption increases crisis likelihood in liberalized markets.
- Deposit insurance:
  - Theoretical trade-off: reduces panics but increases moral hazard.
  - Demirgüç-Kunt and Detragiache (2002): explicit deposit insurance associated with higher probability of banking crisis, especially with deregulated bank interest rates and weak institutions.
  - Design matters: features associated with lower fragility include lower coverage, co-insurance, private sector involvement in management, ex-post funding, and mandatory membership.
  - Results are less robust for developing-country–only samples; explicit deposit insurance can increase volatility of financial development indicators in weak institutional environments.
  - Market discipline is stronger in countries with better institutions; generous deposit insurance can curtail discipline and increase fragility.
- Regulation and supervision:
  - Practices forcing accurate disclosure, empowering private monitoring, and fostering incentives for private corporate control promote bank performance and stability.
  - Poorly designed explicit deposit insurance leads to greater probability of banking crises even after controlling for regulation and supervision.

### F. The political system and crises
- Political economy affects timing and cost of government intervention:
  - Recommendations to improve policy and reduce crisis costs include disseminating information about policy costs, ensuring competition among interest groups, increasing transparency, improving legislative oversight, and allowing foreign bank entry.
  - Political concerns delay government intervention: failing banks are less likely to be taken over or lose licenses before elections than after elections; effect stronger when ruling party is politically weak.

### Evidence on real effects, credit crunch, and firm-level impacts (1997–8 episodes)
- Firm-level evidence:
  - Small and medium-sized firms suffered more than large firms during Malaysia and Korea crises (1997–8); they are more dependent on bank credit—evidence consistent with a credit crunch.
  - Thai firm survey suggests poor demand rather than lack of credit caused production declines, though many firms complained about high interest rates.
  - Indonesia and Korea: evidence of a credit crunch only in the first few months of the crisis in aggregate credit demand/supply tests.
  - Korea: chaebol firms lost preferential access to credit during the banking crisis, not necessarily indicating a generalized credit crunch.
- Cross-country panel evidence:
  - Dell’Ariccia and others (2005): difference-in-difference panel with industry-level data finds more financially dependent sectors suffer more during banking crises, supporting credit crunch hypothesis.
  - Robustness: results hold controlling for flight-to-quality, concomitant currency crises, and bank-portfolio exposure.
  - Magnitude: more financially dependent sectors lose about 1 percentage point of growth in each crisis year compared to less financially dependent sectors.
  - Heterogeneity: effects stronger in developing countries, where private sector has less access to foreign finance, and where crises are more severe.
- Measurement issues:
  - Aggregate stock of real credit to private sector is not a good measure of flow of credit around crises due to valuation effects from inflation or exchange-rate changes.
  - Declines in stock of credit may result from restructuring that transfers nonperforming loans outside the banking system.

### Intervention policies and the costs of crises
- Compiling accurate intervention-policy information across crises is laborious; sequence, timing, and modalities of bank support are crucial and hard to capture quantitatively.
- Honohan and Klingebiel (2003): database of fiscal cost estimates for 40 banking crises; classify policies into five categories (blanket guarantees to depositors, liquidity support to banks, bank recapitalization, financial assistance to debtors, and forbearance); conclude more generous bailouts resulted in higher fiscal costs.
- Political economy: when voters are better informed, elections are close, and number of veto players is large, governments make smaller fiscal transfers and are less likely to exercise forbearance.
- Claessens, Klingebiel, and Laeven (2003): generous support to banks does not reduce output cost of banking crises (output loss relative to trend); interpretation ambiguous due to potential omitted-variable bias.

### Conclusions and directions for future research
- Cross-country econometric research has advanced understanding of how systemic bank fragility is influenced by macroeconomic shocks, banking market structure, institutions, credit‑market institutions, and political economy.
- Limitations:
  - Existing studies based on relatively small number of episodes; broader samples needed to assess robustness.
  - Improving banking-crisis definitions (distinguishing long-simmering problems from sudden shock-triggered events) may enhance model performance.
  - Empirical models have been more useful for identifying associated factors than for out-of-sample prediction; many models were not designed for forecasting.
  - Early-warning indicators should move toward high-frequency data because macroeconomic correlates lose significance if lagged by one year, indicating short transmission to banking systems.
- Further research directions:
  - Study how compliance with banking regulation and the introduction of the Basel II Capital Agreement might affect financial stability, particularly in developing countries.
  - Investigate effects of policy choices such as liberalization, foreign bank entry, and resulting market structures on fragility as globalization and consolidation reshape banking systems.
  - Integrate bank-level information into cross-country empirical models to bridge open economy macroeconomics and microeconomics of banking and regulation.

*Source: _wp0596 - 1. Banking Crises Determinants.*

### 1. Banking Crises Determinants .................................................................................... 21

### 1. Banking Crises Determinants

### Introduction
- Paper surveys the rapidly growing literature on causes and consequences of bank fragility and highlights directions for future research.
- Organisation of the paper:
  - Review of basic facts about the recent wave of financial crises.
  - Presentation of the two basic methodologies adopted in cross-country empirical studies of the determinants of banking crises.
  - Discussion of how these models have been used for crisis prediction.
  - Review of how various factors contribute to bank fragility.
  - Survey of work on the economic effects of banking crises.
  - Conclusions and directions for further research.

### The resurgence of financial instability in the 1990s
- Historical context:
  - Post–World War II period: banking crises were rare; calm macroeconomic environment, low inflation, pervasive controls on international capital flows, and restrictive supervisory powers contributed to stability.
  - Breakdown of Bretton Woods and the 1970s: macroeconomic turbulence increased but banking sector remained largely sound in most countries.
  - Early 1980s financial liberalization and high real interest rates: several financial crises in Latin America and other developing countries.
- Notable episodes and consequences:
  - Diáz-Alejandro (1985) traced the Chilean crisis to the banking system and its privatization.
  - U.S. savings and loans (S&L) debacle illustrated erosion of bank capital after liberalization, generous deposit insurance, and ineffective regulation; fiscal cost borne by taxpayers, but negligible macroeconomic effects.
  - 1990s episodes with major macroeconomic consequences:
    - Scandinavian banking crises: currency devaluation, falling asset prices, banking crises and economic slowdown (Drees and Pazarbasioglu, 1998).
    - Japan: collapse of asset price bubble rendered most of the banking sector insolvent; regulatory forbearance and lax monetary policy prolonged balance sheet repair for more than a decade (Hoshi and Kashyap, 2004). After over 40 years of rapid expansion, growth ground to a halt in 1992 and had yet to fully recover by the time of writing.
    - Mexico (“tequila” crisis, began December 1994): combination of faltering banking system, dollar-denominated debt, and political shocks led to devaluation and financial meltdown; cost of bailing out the banks reached almost 20 percent of GDP; despite the rescue, bank credit to the private sector and economic growth remained lackluster.
    - East Asian crises (1997–8): economies with sound public finances and high growth were severely affected within months as banks buckled, depositors lost confidence, asset prices collapsed, and foreign capital inflows evaporated (Lindgren and others, 1999).
- Systematic assessment:
  - IMF and World Bank surveys (1996) discovered that a full three-quarters of their membership had experienced significant banking problems during 1980–96.
  - Problems varied across cases: insolvency of one or two large banks; loss-making government-owned institutions needing chronic recapitalization; pervasive weaknesses across regions and levels of development.
- Implication: banking fragility is pervasive and multifaceted, warranting systematic empirical investigation.

### Two econometric approaches to identifying determinants of banking crises

H3: The signals approach
- Origin and application:
  - Signals approach originally developed to identify turning points in business cycles.
  - First applied to banking crises by Kaminsky and Reinhart (1999), focusing on "twin crises" (simultaneous currency and banking crises).
- Data and definitions:
  - Sample: twenty industrial and emerging countries during 1970–95.
  - Currency crises identified based on an index of market turbulence (Eichengreen and others, 1995).
  - Onset of a banking crisis assumed to coincide with depositor runs leading to closure or takeover of one or more banks, or with large-scale government intervention to assist, take over, merge, or close one or more financial institutions leading to more intervention elsewhere in the financial system.
- Incidence and co-movement:
  - Currency crises more frequent than banking crises in the sample: 76 episodes versus 26.
  - Of these, 19 episodes are twin crises; a wide majority of banking crises are accompanied by an exchange rate crash.
  - Caveat: sample selection (countries with fixed or heavily managed exchange rates) may overemphasize exchange rate importance.
- Pre-crisis behavior (24 months preceding crises):
  - Monetary growth and interest rates (both lending and deposit rates) are above normal.
  - Export growth appears below trend before banking crises.
  - Real exchange rate is appreciating.
  - Real output growth falls below trend about eight months before the peak of the banking crisis.
  - Stock prices peak at about the same time as the onset, suggesting banking crises are preceded by a cyclical downturn.
- Formal signals methodology:
  - Behavior of each relevant variable during the 24 months prior to a crisis is contrasted with behavior during "tranquil" times.
  - A variable signals a crisis any time it crosses a particular threshold; if a crisis occurs within the next 24 months the signal is correct; otherwise it is a false alarm.
  - Thresholds chosen to minimize the in-sample noise-to-signal ratio (authors use an “adjusted” noise-to-signal ratio computed as the ratio of the probability of false alarms to one minus the probability of missing a crisis).
  - Performance yardsticks: Type I error (probability of missing a crisis), Type II error (probability of a false signal), noise-to-signal ratio, and probability of a crisis conditional on a signal.
- Key signal results:
  - Indicators with lowest noise-to-signal ratio and highest probability of crisis conditional on the signal: appreciation of the real exchange rate, followed by equity prices and the money multiplier.
  - These indicators have large Type I error: they fail to issue a signal in 73–79 percent of the observations during the 24 months preceding a crisis.
  - Type II error (false signals) is much lower, ranging between 8 and 9 percent.
  - The real interest rate has the lowest Type I error, signaling in 30 percent of pre-crisis observations.
  - Indicators reflecting real sector developments are more closely associated with banking crises than currency crises.
  - Twin crises are preceded by more acute warning signs than individual crises and have more protracted adverse effects.

H3: The multivariate logit approach
- Motivation and advantages:
  - Signals approach treats covariates in isolation; does not aggregate information across indicators and ignores magnitude information relative to thresholds.
  - Multivariate logit approach models the probability of crisis as a function of a vector of explanatory variables, producing an estimated crisis probability that aggregates information consistently.
- Model specification:
  - Dependent variable: binary—zero if no crisis, one if crisis.
  - Probability that a crisis occurs P(i,t) modeled as F(X(i,t)' β), with F assumed logistic.
  - Estimated coefficients reflect effects on ln(P(i,t)/(1–P(i,t))).
  - Increase in probability depends on the original probability and on initial values of all independent variables and their coefficients.
- Methodological issues:
  - Observations during crisis episodes are typically excluded from the sample because explanatory variables may be affected by the crisis (feedback effects), e.g., real interest rate falling due to rescue-related monetary loosening.
  - Construction of the banking crisis dummy:
    - Economies in transition excluded.
    - Episodes of banking sector distress identified using Caprio and Klingebiel (1996), Lindgren and others (1996), and other case studies.
    - For an episode to be classified as a full-fledged crisis in the panel, at least one of the following four conditions had to hold:
      - Ratio of nonperforming assets to total assets in the banking system exceeded 10 percent.
      - Cost of the rescue operation was at least 2 percent of GDP.
      - Banking sector problems led to a large scale nationalization of banks.
      - Extensive bank runs took place or emergency measures such as deposit freezes, prolonged bank holidays, or generalized deposit guarantees were enacted by the government in response to the crisis.
- Sample and estimation notes:
  - Baseline regressions extended through 2002 and to include more countries; number of crisis episodes in the baseline specification rose from 31 to 77.
  - Model estimated without country fixed effects to include noncrisis countries as controls; errors allowed to be correlated within each country by clustering the errors by country (in new regressions). Earlier paper used robust standard errors.

### Empirical findings from multivariate logit specifications
- Macroeconomic correlates:
  - Low GDP growth, high real interest rates, and high inflation are significantly correlated with the occurrence of a banking crisis: crises tend to occur during periods of weak economic growth and loss of monetary control.
  - Exposure to real interest rate risk is a source of banking fragility; higher and more volatile real interest rates during the 1980s and 1990s may have contributed to greater incidence of banking crises.
  - Changes in the terms of trade and exchange rate depreciation are not significant.
  - Fiscal variable (budget surplus scaled by GDP) has a positive coefficient, but is significant only when deposit insurance is omitted.
- Banking sector and external vulnerability:
  - Ratio of broad money to foreign exchange reserves (measuring vulnerability to a run on the currency) enters positively and significantly, suggesting bank exposure to currency crises plays a role.
  - Credit to the private sector enters with a positive sign, indicating countries where banks have larger exposure to private sector borrowers are more vulnerable, perhaps due to mismanaged liberalization.
  - High lagged credit growth, capturing a credit boom, is significantly and positively correlated with crisis probability in all specifications.
- Institutional variables:
  - Level of development (GDP per capita) is negatively correlated with systemic banking sector problems, indicating that developing countries are more vulnerable to bank fragility.
  - Presence of an explicit deposit insurance scheme appears to be a risk factor, probably because the moral hazard effect outweighs any reduction in self-fulfilling panics.

*Source: _wp0596 - 1. Banking Crises Determinants.*

### Section V.

### V. STUDIES OF THE DETERMINANTS OF BANKING CRISES

### IV. USING ECONOMETRIC MODELS OF BANKING CRISES AS EARLY WARNING SYSTEMS
- Need to improve monitoring capabilities of financial vulnerabilities became acute as banking crises spread in the 1990s.
- Proposed leading indicators and findings:
  - Gavin and Houseman (1995) and Sachs, Tornell, and Velasco (1996) proposed using credit growth to detect credit booms.
  - Mishkin (1996) highlighted equity price declines.
  - Calvo (1996) suggested monitoring the ratio of broad money to foreign exchange reserves.
- Hoonah (1997):
  - Sample: eighteen crisis and six noncrisis countries.
  - Divides crisis countries into three types: macroeconomic, microeconomic, government-behavior related.
  - Findings:
    - Crises due to government intervention associated with high levels of borrowing and central bank lending to the banking system.
    - Macroeconomic-origin crises associated with high loan-to-deposit ratios, high foreign borrowing-to-deposit ratios, and high growth rates of credit.
    - Microeconomic-origin crises not associated with abnormal behavior in the seven indicators examined.
- Rojas-Suarez (1998):
  - Proposes adapting CAMEL early warning system to emerging markets; adds non-CAMEL indicators such as deposit interest rates, the spread between lending and deposit rates, growth rate of credit, and growth rate of interbank lending.
  - Limitation: better at identifying weak banks within a system than systemic crises; requires detailed bank-level information, making cross-country application difficult.
- Signals approach (Kaminsky and Reinhart, 1999; Kaminsky, 1999; Goldstein, Kaminsky and Reinhart, 2000):
  - Composite index constructed as the number of indicators crossing thresholds at any time; weighted variant uses signal-to-noise ratios.
  - Best composite indicator outperforms the real exchange rate in predicting crises in-sample but is worse at predicting tranquil observations.
- Demirgüç-Kunt and Detragiache (2000):
  - Multivariate logit framework yields lower in-sample type I and type II errors than Kaminsky and Reinhart signals.
  - Use out-of-sample forecasts of crisis probabilities based on logit coefficients and forecasts of right-hand-side variables from professional forecasters or international institutions.
  - Two monitoring frameworks for using forecasted probabilities:
    - Threshold-trigger framework: choose probability threshold to act by minimizing a loss function trading off costs of acting when no crisis (type I) vs. not acting when crisis (type II); optimal trigger depends on in-sample predictive power and costs of mistakes.
    - Rating framework: rate fragility of banking system with ratings interpreted in terms of crisis probability to guide differing actions.
  - Application to six crisis episodes (Jamaica, Indonesia, Korea, Malaysia, Philippines, and Thailand):
    - Both actual and forecasted data would have indicated high vulnerability in Jamaica.
    - For the Asian countries the picture would have been much rosier; Thailand and the Philippines showed some signs of fragility but overall assessments were fairly reassuring because expectations of continued strong economic growth and stable exchange rates offset negative impacts of relatively high real interest rates and strong past credit expansion.
- Limits and avenues for improvement:
  - Econometric monitoring tools show limited out-of-sample success; possible reasons include:
    - New crises differing from past crises (coefficients from in-sample estimation may be of limited out-of-sample use).
    - Banking crises are rare events, so in-sample estimates have relatively few data points.
  - Suggested improvements:
    - Develop alternative scenarios with high and low forecasts for explanatory variables.
    - Use stress-testing exercises (used in Financial Sector Assessment Programs by the IMF and World Bank).
    - Explore movements in high-frequency variables (spreads on interbank market or commercial paper issued by banks, stock market valuation of banks, corporate vulnerability) before crises—requires significant data collection for wide country coverage.

### A. Individual Bank Measures of Fragility and Systemic Crises
- Early literature (since early 1970s) uses bank balance sheet and market information to explain and forecast individual bank failure.
- González-Hermosillo (1999):
  - Uses bank-specific and macroeconomic data for U.S. regions, Mexico, and Colombia.
  - Finds nonperforming loans and capital asset ratios often deteriorate rapidly before bank failure.
  - Finds little evidence that individual bank failure is driven by overall banking sector fragility (limited contagion).
- Bongini, Claessens, and Ferri (1999) on Asian crises:
  - Analyze CAMEL variables, bank size, corporate connections, and country dummies.
  - Findings:
    - CAMEL variables reasonably predict distress.
    - Big financial institutions are more likely to become distressed but less likely to be closed.
    - Connected institutions are more likely to experience trouble.
    - Conclude systemic crisis in Asia reflected both exogenous shocks and significant prior weaknesses at individual bank level.

### B. Financial Liberalization and Crises
- Theory: Financial liberalization gives banks greater opportunities to take on risk; without strong institutions and prudential regulation, liberalization may increase fragility (Caprio and Summers, 1993; Stiglitz, 1994).
- Demirgüç-Kunt and Detragiache (1998):
  - Find banking crises are more likely in countries that have liberalized financial systems, controlling for other characteristics.
  - Effect is mitigated by strong institutional environment (respect for rule of law, low corruption, good contract enforcement).
- Subsequent empirical studies (Mehrez and Kaufmann, 1999; Glick and Hutchison, 2001; Arteta and Eichengreen, 2002; Noy, 2004) similarly find financial liberalization can significantly increase bank fragility.

### C. International Shocks, Exchange Rate Regime, and Crises
- Research examines impact of worldwide shocks and exchange rate regime on bank fragility; historical episodes cited (Volcker disinflation 1979–81; U.S. monetary tightening 1994 and Mexican crisis).
- Eichengreen and Rose (1998):
  - First empirical paper on role of international shocks in banking crises.
  - Finds strong effect of interest rates and, to a smaller extent, GDP growth in advanced economies on bank fragility in developing countries.
- Arteta and Eichengreen (2002):
  - When sample extended to more recent years, evidence of an OECD effect becomes weaker; conclude mid-1990s crises were different with external factors playing a smaller role relative to domestic factors.
- Exchange rate regime considerations:
  - Flexible exchange rates may stabilize financial system by absorbing real shocks and discouraging over-borrowing in foreign currency (Mundell, 1961; Eichengreen and Hausmann, 1999).
  - Fixed exchange rates limit lender of last resort operations and may increase susceptibility to bank runs and financial panics (Eichengreen and Rose, 1998; Wood, 1999).
  - Conversely, commitment to a peg may discipline policymakers and reduce risk-taking, possibly reducing crisis probability (Eichengreen and Rose, 1998; Calvo, 1999).
- Empirical findings on exchange rate regime and dollarization:
  - Arteta and Eichengreen (2002): countries with fixed and flexible exchange rates are equally susceptible to banking crises.
  - Domaç and Martinez-Peria (2003): adopting a fixed exchange rate diminishes likelihood of a banking crisis in developing countries, but once a crisis occurs its economic cost is larger under a fixed exchange rate.
  - Arteta (2003): no evidence that deposit and credit dollarization increases fragility for a large sample of developing and transition countries.
  - De Nicolo, Honohan and Ize (2003): using average Z-scores and non-performing loans, find dollarization positively related to both measures of bank fragility.

### D. Bank Ownership and Structure and Crises
- State ownership:
  - La Porta, Lopez-de-Silanes, and Shleifer (2002) and Barth, Caprio, and Levine (2001): greater state ownership associated with reduced competition, poorer productivity, lower growth.
  - Caprio and Martinez-Peria (2000): greater state ownership at beginning of 1980s associated with greater probability of a banking crisis during 1980–97.
  - Barth, Caprio, and Levine (2001) confirm via simple cross-sectional regressions.
- Foreign ownership:
  - Foreign entry improves operating efficiency and financial intermediation (Claessens, Demirgüç-Kunt, and Huizinga, 2001).
  - Concerns about foreign banks fleeing not supported by empirical evidence:
    - Demirgüç-Kunt, Levine, and Min (1998): presence of foreign banks associated with lower risk of banking crisis.
    - Dages and others (2000): foreign banks in Argentina and Mexico had stronger and less volatile loan growth during and after Tequila Crisis (1994–9).
    - Peek and Rosengren (2000): similar conclusions for Argentina, Brazil, Mexico, 1994–1999.
    - Detragiache and Gupta (2004) for Malaysia: foreign banks performed better during the crisis only if from outside the region; Asian-focused foreign banks did not perform significantly better than domestic banks.
- Competition and concentration:
  - Beck, Demirgüç-Kunt, and Levine (2004):
    - Find banking crises are less likely in economies with more concentrated banking systems, fewer regulatory restrictions on bank competition and activities, and national institutions that encourage competition.
    - Conclude no evidence that greater competition damages stability; concentration result likely reflects better risk diversification by larger banks in concentrated systems.

### E. The Role of Institutions
- Institutional development matters:
  - Demirgüç-Kunt and Detragiache (1998): proxies for institutional development (GDP per capita and index of law and order) show weaker institutions related to higher probability of banking crises.
  - Mehrez and Kaufmann (1999): low transparency (or high corruption) increases likelihood of banking crises in financially liberalized markets.
- Deposit insurance:
  - Theoretical trade-off: reduces self-fulfilling panics but creates moral hazard (Kane, 1989).
  - Demirgüç-Kunt and Detragiache (2002):
    - Explicit deposit insurance associated with higher probability of banking crisis in a large sample of countries, especially if bank interest rates are deregulated and institutional environment is weak.
    - Impact varies with design: features associated with lower fragility include lower coverage, co-insurance, private sector involvement in management, ex-post funding, and mandatory membership.
  - Arteta and Eichengreen (2002): find these results less robust when focusing only on developing countries and ignoring design differences.
  - Cull, Senbet, and Sorge (2005): explicit deposit insurance increases volatility of financial development indicators in countries with weak institutional development.
  - Demirgüç-Kunt and Huizinga (2004): at bank-level, market discipline (via interest rates and deposit growth) is stronger in countries with better institutions; generously designed deposit insurance can curtail market discipline, resulting in fragility.
- Regulation and supervision:
  - Barth, Caprio, and Levine (2004):
    - Using comprehensive survey database on regulation and supervision measures, find that practices forcing accurate information disclosure, empowering private sector monitoring, and fostering incentives for private agents to exert corporate control promote bank performance and stability.
    - Regulatory and supervisory regimes with these features have suffered fewer crises in past two decades.
    - Confirm that poorly designed explicit deposit insurance leads to greater probability of banking crises even after controlling for regulation and supervision.

### F. The Political System and Crises
- Political economy influences government intervention timing and cost of crises:
  - Kroszner (1997), studying U.S. savings and loan crisis, recommends measures to improve government financial sector policy and reduce crisis costs:
    - Disseminate information about costs of inefficient policy.
    - Ensure competition among interest groups.
    - Increase transparency of government decisions.
    - Improve legislative oversight of regulatory process.
    - Allow entry of foreign banks.
  - Brown and Dinc (2004):
    - Using individual bank failure data in developing countries, find political concerns delay government intervention.
    - Failing banks are less likely to be taken over by government or lose licenses before elections than after elections.
    - Effect is stronger when the ruling party is politically weak.

*Source: _wp0596 - Section V.*

### conclusion for Malaysia and Korea during 1997–8. They found small and medium-sized

### _wp0596 - conclusion for Malaysia and Korea during 1997–8. They found small and medium-sized

### Evidence on firm-level and sectoral effects during 1997–8 crises
- Small and medium-sized firms suffered more than large firms during the crisis in Malaysia and Korea during 1997–8.
- Small and medium-sized firms are usually more dependent on bank credit than large firms; the observed disproportionate suffering is evidence of a credit crunch.
- A survey of Thai firms suggests poor demand rather than lack of credit caused the decline in production, although many firms complained about high interest rates (Dollar and Hallward‑Driemeier, 2000).
- For Indonesia and Korea, Ghosh and Ghosh (1999) test an aggregate model of credit demand and supply and find evidence of a credit crunch, but only in the first few months of the crisis.
- Using firm-level data from Korea, Borensztein and Lee (2002) show that firms belonging to industrial groups (chaebols) lost their preferential access to credit during the banking crisis, although this was not necessarily evidence of a credit crunch.

### Cross-country panel evidence on the credit crunch hypothesis
- Dell’Ariccia and others (2005) apply a difference-in-difference approach with a panel of countries and industry-level data to identify real effects of banking crises.
- Main result: more financially dependent sectors suffer more during banking crises, supporting the credit crunch hypothesis.
- Robustness: results hold after controlling for flight-to-quality during recessions, concomitant currency crises, and exposure of bank portfolios to bank-dependent industries.
- Magnitude: more financially dependent sectors lose about 1 percentage point of growth in each crisis year compared to less financially dependent sectors.
- Heterogeneity: differential effects are stronger in developing countries, in countries where the private sector has less access to foreign finance, and where the crises are more severe.

### Measurement issues for credit around banking crises
- Changes in the aggregate stock of real credit to the private sector are not a good measure of the flow of credit available to the economy, especially around banking crises, because of valuation effects caused by inflation or exchange rate changes.
- A decline in the stock of credit may result from restructuring operations that transfer non‑performing loans to agencies outside the banking system (Demirgüç‑Kunt and others, forthcoming).

### Intervention policies and the costs of crises
- Compiling accurate information on intervention policies for a large sample of crises is laborious; sequence, timing, and modalities of bank support strategies are crucial and hard to capture quantitatively.
- Honohan and Klingebiel (2003): database of fiscal cost estimates for 40 banking crises, classifying policies into five categories (blanket guarantees to depositors, liquidity support to banks, bank recapitalization, financial assistance to debtors, and forbearance). They conclude more generous bailouts resulted in higher fiscal costs.
- Keefer (2001): when voters are better informed, elections are close, and the number of veto players is large, governments make smaller fiscal transfers to the financial sector and are less likely to exercise forbearance; transparency, information dissemination, and competition among interest groups shape crisis response policies.
- Claessens, Klingebiel, and Laeven (2003): generous support to the banking system does not reduce the output cost of banking crises (output loss relative to trend). Interpretation is ambiguous because omitted exogenous shocks may cause both stronger output decline and more generous interventions. Results remain after controlling for GDP growth prior to crisis, existence of deposit insurance, inflation rate at onset, state ownership of banks, degree of dollarization, and others.

### Conclusions and directions for future research
- Cross-country econometric research on systemic banking crises has advanced rapidly, improving understanding of how systemic bank fragility is influenced by macroeconomic shocks, banking market structure, broad institutions, credit‑market institutions, and political economy variables.
- Existing studies are based on a relatively small number of episodes; as broader samples become available, assessing robustness of current conclusions is important.
- Improving the definition of a banking crisis may enhance model performance: distinguishing long‑simmering problems revealed gradually from sudden events triggered by severe exogenous shocks may clarify relationships with macroeconomic variables.
- Empirical models have been more useful in identifying factors associated with crisis occurrence than in out‑of‑sample prediction; many models were not designed as forecasting tools.
- Early‑warning indicators should move toward high‑frequency data, such as market data, because macroeconomic correlates of crises tend to lose significance if lagged by one year, indicating transmission to the banking system can be quite short.
- Further research directions:
  - Study how compliance with banking regulation and the introduction of the Basel II Capital Agreement might affect financial stability, particularly in developing countries.
  - Investigate effects of policy choices such as liberalization, foreign bank entry, and resulting market structures on bank fragility as globalization and consolidation reshape banking systems.
  - Integrate bank‑level information into cross‑country empirical models to bridge open economy macroeconomics and microeconomics of banking and regulation.

*Source: IMF working paper content provided in the supplied PDF excerpt.*

### REFERENCES

### _wp0596 - REFERENCES

### Banking crises: causes, propagation, and costs
- Akerlof, G.P., and Paul Romer, 1993, “Looting: The Economic Underworld of Bankruptcy for Profit,” Brookings Papers on Economic Activity, pp. 1–73 (Washington: Brookings Institution).
- Bell, James, and Darren Pain, 2000, “Leading Indicator Models of Banking Crises—A Critical Review,” Financial Stability Review, Bank of England, Issue 9, Article 3, pp. 113–29.
- Bernanke, Ben S., 1983, “Nonmonetary Effects of the Financial Crisis in the Propagation of the Great Depression,” American Economic Review, Vol. 73, pp. 257–76.
- Bernanke, Ben S., C. S. Lown, and B. M. Friedman, 1991, “The Credit Crunch,” Brookings Papers on Economic Activity, Vol. 2, pp. 205–39 (Washington: Brookings Institution).
- Borensztein, Eduardo, and Jong-Wha Lee, 2002, “Financial Crisis and Credit Crunch in Korea: Evidence from Firm-Level Data,” Journal of Monetary Economics, Vol. 49, pp. 853–75.
- Bordo, Michael, Barry Eichengreen, Daniela Klingebiel, and Maria Soledad Martinez-Peria, 2001, “Is the Crisis Problem Growing More Severe?” Economic Policy, Vol. 32, pp. 51–82.
- Borio, Claudio, and Philip Lowe, 2002, “Assessing the Risk of Banking Crises,” BIS Quarterly Review, pp. 43–54 (Basel, Switzerland: Bank for International Settlements).
- Calomiris, Charles, and J. R. Mason, 1997, “Contagion and Bank Failures During the Great Depression: The Chicago 1932 Bank Panic,” American Economic Review, Vol. 87, No. 5, pp. 883–63.
- Drees, Burkhard, and Ceyla Pazarbasioglu, 1998, “The Nordic Banking Crises: Pitfalls in Financial Liberalization?” Working Paper No. 161 (Washington: International Monetary Fund).
- Gorton, Gary, 1988, “Banking Panics and Business Cycles,” Oxford Economic Papers, Vol. 40, pp. 751–81.
- Haubrich, Joseph G., 1990, “Nonmonetary Effects of Financial Crises: Lessons from the Great Depression,” Journal of Monetary Economics, Vol. 25, No. 2, pp. 223–52.
- Hoggarth, Glenn, Ricardo Reis, and Victoria Saporta, 2002, “Costs of Banking System Instability: Some Empirical Evidence,” Journal of Banking and Finance, Vol. 26, pp. 825–55.
- Honohan, Patrick, 1997, “Banking System Failures in Developing and Transition Countries: Diagnosis and Prediction,” BIS Working Paper 39 (Basel, Switzerland: Bank for International Settlements).
- Hoshi, Takeo, and Anil K. Kashyap, 2004, “Japan’s Financial Crisis and Economic Stagnation,” Journal of Economic Perspectives, Vol. 18, No. 1, pp. 3–26.
- Lindgren, Carl-Johan, Gillian Garcia, and Matthew Saal, 1996, Bank Soundness and Macroeconomic Policy (Washington: International Monetary Fund).
- Rojas-Suarez, Liliana, and Steven Weisbrod, 1995, Financial Fragilities in Latin America, Occasional Paper No. 132 (Washington: International Monetary Fund).
- Wood, Geoffrey, 1999, “Great Crashes in History: Have They Lessons for Today?” Oxford Review of Economic Policy, Vol. 15, pp. 98–109.

### Determinants, monitoring, and early warning indicators
- Bell, James, and Darren Pain, 2000, “Leading Indicator Models of Banking Crises—A Critical Review,” Financial Stability Review, Bank of England, Issue 9, Article 3, pp. 113–29.
- Demirgüç-Kunt, Aslí, and Enrica Detragiache, 1998, “The Determinants of Banking Crises: Evidence from Developing and Developed Countries,” IMF Staff Papers, Vol. 45, pp. 81–109.
- Demirgüç-Kunt, Aslí, and Enrica Detragiache, 2000, “Monitoring Banking Sector Fragility: A Multivariate Logit Approach,” World Bank Economic Review, Vol. 14, No. 2, pp. 287–307.
- Gonzáles-Hermosillo, Brenda, Ceyla Pazarbasioglu, and Robert Billings, 1997, “Determinants of Banking Sector Fragility: A Case Study of Mexico,” IMF Staff Papers, Vol. 44, No. 3 (Washington: International Monetary Fund).
- Hardy, Daniel, and Ceyla Pazarbasioglu, 1999, “Determinants and Leading Indicators of Banking Crises: Further Evidence,” Staff Papers, Vol. 46, No. 3, pp. 247–58 (Washington: International Monetary Fund).
- Kaminsky, Graciela, 1999, “Currency and Banking Crises: The Early Warnings of Distress,” Working Paper 99/178 (Washington: International Monetary Fund).
- Rojas-Suarez, Liliana, 1998, “Early Warning Indicators of Banking Crisis: What Works for Developing Countries?,” (Washington: Research Department, Inter-American Development Bank).

### Regulation, supervision, deposit insurance, and resolution
- Barth, James R., Gerard Caprio, and Ross Levine, 2001, “Banking Systems Around the Globe: Do Regulations and Ownership Affect Performance and Stability?” in Prudential Supervision: What Works and What Doesn’t, ed. by R. Mishkin (Chicago: University of Chicago Press).
- Barth, James R., Gerard Caprio, and Ross Levine, 2004, “Bank Regulation and Supervision: What Works Best?,” Journal of Financial Intermediation, Vol. 13, No. 2, pp. 205-48.
- Caprio, Gerard, and Daniela Klingebiel, 1996, “Dealing with Bank Insolvencies: Cross Country Experience,” Policy Research Working Paper No. 1620 (Washington: World Bank).
- Caprio, Gerard, and Maria Soledad Martinez-Peria, 2000, “Avoiding Disaster: Policies to Reduce the Risk of Banking Crises,” Discussion Paper (Cairo, Egypt: Egyptian Center for Economic Studies).
- Cull, Robert, Lemma W. Senbet, and Marco Sorge, 2005, “Deposit Insurance and Financial Development,” Journal of Money, Credit and Banking (forthcoming).
- Demirgüç-Kunt, Aslí, and Enrica Detragiache, 2002, “Does Deposit Insurance Increase Banking System Stability? An Empirical Investigation,” Journal of Monetary Economics, Vol. 49, pp. 1373–406.
- Demirgüç-Kunt, Aslí, and H. Huizinga, 2004, “Market Discipline and Deposit Insurance,” Journal of Monetary Economics, Vol. 51, No. 2.
- Honohan, Patrick, and Daniela Klingebiel, 2003, “The Fiscal Cost Implications of an Accommodating Approach to Banking Crises,” Journal of Banking and Finance, Vol. 27, pp. 1539–60.
- Lindgren, Carl-Johan, and Tomás J. Baliño, Charles Enoch, Anne-Marie Gulde, Marc Quintyn, and Leslie Teo, 1999, “Financial Sector Crisis and Restructuring: Lessons from Asia,” Occasional Paper No. 188 (Washington: International Monetary Fund).
- Kane, Edward J., 1989, The S&L Insurance Mess: How Did it Happen?, Urban Institute Press (Washington: Urban Institute).
- Sheng, Andrew, 1995, “Bank Restructuring” (Washington: The World Bank).

### Foreign banks, globalization, and financial integration
- Arteta, Carlos, 2003, “Are Financially Dollarized Countries More Prone to Costly Crises?” International Finance Discussion Paper No. 763 (Washington: Board of the Governors of the Federal Reserve System).
- Arteta, Carlos, and Barry Eichengreen, 2002, “Banking Crises in Emerging Markets: Presumptions and Evidence,” in Financial Policies in Emerging Markets, ed. by M. Blejer, and M. Skreb (Cambridge, Massachusetts: MIT Press).
- Claessens, Stijn, Aslí Demirgüç-Kunt, and Harry Huizinga, 2001, “How Does Foreign Entry Affect Domestic Banking Markets?” Journal of Banking and Finance, Vol. 25, No. 5, pp. 891–911.
- Dages, B. Gerald, Linda Goldberg, and Daniel Kinney, 2000, “Foreign and Domestic Bank Participation in Emerging Markets: Lessons from Mexico and Argentina,” Federal Reserve Bank of New York Economic Policy Review, Vol. 6, No. 3, pp. 17–36.
- De Nicolo, Gianni, Patrick Honohan, and Alain Ize, 2003, “Dollarization of the Banking System: Good or Bad?” World Bank Policy Research Working Paper No. 3116 (Washington: World Bank).
- Detra­giache, Enrica, and Poonam Gupta, 2004, “Foreign Banks in Emerging Market Crises: Evidence from Malaysia,” Working Paper No. 04/129 (Washington: International Monetary Fund).
- Peek, Joseph, and Eric S. Rosengren, 2000, “Implications of the Globalization of the Banking Sector: The Latin American Experience,” Federal Reserve Bank of Boston, New England Economic Review, pp. 45–62 (September–October).
- Claessens, Daniela Klingebiel, and L. Laeven, 2003, “Resolving Systemic Crises: Policies and Institutions” (unpublished; Washington: World Bank).
- Demirgüç-Kunt, Aslí, Ross Levine, and Hong-Ghi Min, 1998, “Opening to Foreign Banks: Issues of Stability, Efficiency and Growth,” in The Implications of Globalization of World Financial Markets, ed. by A. Meltzer (Korea: Bank of Korea).

### Exchange rates, capital flows, dollarization, and twin crises
- Calvo, Guillermo, 1996, “Capital Flows and Macroeconomic Management: Tequila Lessons,” International Journal of Finance and Economics, Vol. 1, No. 3, pp. 207–24.
- Calvo, Guillermo,1999, “Testimony on Full Dollarization, Presented before a Joint Hearing of the Subcommittees on Economic Policy and International Trade and Finance, U.S. Congress (April).
- Díaz-Alejandro, Carlos, 1985, “Good-Bye Financial Repression, Hello Financial Crash,” Journal of Development Economics, Vol. 19, pp.1–24.
- Eichengreen, Barry, Andrew. Rose, and C. Wyplosz, 1995, “Exchange Market Mayhem: The Antecedents and Aftermath of Speculative Attacks” (Berkeley, California: University of California).
- Eichengreen, Barry, and Albert Fishlow, 1998, “Contending with Capital Flows: What is Different About the 1990s?” in Capital Flows and Financial Crises, ed. by M. Kahler, pp. 23–68, (Ithaca: Cornell University Press).
- Eichengreen, Barry, and Andrew Rose, 1998, “Staying Afloat When the Wind Shifts: External Factors and Emerging-Market Banking Crises,” NBER Working Paper No. 6370 (Cambridge, Massachusetts: National Bureau of Economic Research).
- Eichengreen, Barry, and R. Hausmann, 1999, “Exchange Rates and Financial Fragility,” NBER Working Paper No. 7418 (Cambridge, Massachusetts: National Bureau of Economic Research).
- Glick, Reuven, and Michael Hutchison, 2001, “Banking and Currency Crises: How Common are the Twins?,” in Financial Crises in Emerging Markets, ed. by R. Glick, R. Moreno, and M. Spiegel (Cambridge: Cambridge University Press).
- Gonzáles-Hermosillo, Brenda, Ceyla Pazarbasioglu, and Robert Billings, 1999, “Determinants of Exante Nanking System Distress: a Macro-Micro Empirical Exploration of Some Recent Episodes,” Working Paper 99/33 (Washington: International Monetary Fund).
- Mundell, Robert, 1961, “A Theory of Optimum Currency Areas,” American Economic Review, Vol. 51, pp. 717–25.
- De Nicolo, Gianni, Patrick Honohan, and Alain Ize, 2003, “Dollarization of the Banking System: Good or Bad?” World Bank Policy Research Working Paper No. 3116 (Washington: World Bank).

### Political economy, governance, and institutional factors
- Bongini, Paola, Stijn Claessens, and Giovanni Ferri, 1999, “The Political Economy of Distress in East Asian Financial Institutions” (unpublished; Washington: World Bank).
- Brown, Craig O., and Serdar Dinç, 2004, “The Politics of Bank Failures: Evidence from Emerging Markets” (Grand Rapids: University of Michigan Business School).
- Caprio, Gerard, and Lawrence Summers, 1993, “Finance and Its Reform: Beyond Laissez-Faire’, Policy Research Working Paper No. 1171 (Washington: World Bank).
- Keefer, Phillip, 2001, “When do Special Interests Run Rampant? Disentangling the Role of Elections, Incomplete Information, and Checks and Balances in Banking Crises” (Washington: World Bank).
- Kroszner, Randall S., 1997, “The Political Economy of Banking and Financial Regulation in the United States,” in The Banking and Financial Structure in the NAFTA Countries and Chile, ed. by G. M. von Furstenberg (Boston, Kluwer Academic Publishers).
- La Porta, Rafael, Florencio Lopez-de-Silanes, and Andrei Shleifer, 2002, “Government Ownership of Banks,” Journal of Finance, Vol. 57, pp. 265–301.
- Mishkin, Frederic S., 1996, “Understanding Financial Crises: A Developing Country Perspective,” NBER Working Paper No. 5600 (Cambridge Massachusetts: National Bureau of Economic Research).
- Stiglitz, Joseph E., 1994, “The Role of State in Financial Markets,” in Proceedings of the World Bank Annual Conference on Development Economics, ed. by M. Bruno, and Boris Pleskovic (Washington: World Bank).

### Firm-level, microeconomic, and sectoral studies of crises
- Borensztein, Eduardo, and Jong-Wha Lee, 2002, “Financial Crisis and Credit Crunch in Korea: Evidence from Firm-Level Data,” Journal of Monetary Economics, Vol. 49, pp. 853–75.
- Borio, Claudio, and Philip Lowe, 2002, “Assessing the Risk of Banking Crises,” BIS Quarterly Review, pp. 43–54 (Basel, Switzerland: Bank for International Settlements).
- Dollar, David, and Mary Hallward-Driemeier, 2000, “Crisis, Adjustment, and Reform in Thai Industrial Firms,” The World Bank Research Observer, Vol. 15, pp.1–22 (Washington: World Bank).
- Domaç, Ilker, and G. Ferri, 1999, “The Credit Crunch in East Asia: Evidence from Field Findings on Bank Behaviour and Policy Issues” (unpublished: Washington: World Bank).
- Domaç, Ilker, and Maria Soledad Martinez Peria, 2003, “Banking Crises and Exchange Rate Regimes: Is There a Link?” Journal of International Economics, Vol. 61, No. 1, pp 41–72.
- Ramos, Alberto M., 1998, “Capital Structures and Portfolio Composition During Banking Crisis—Lessons from Argentina 1995,” Working Paper 98/121 (Washington: International Monetary Fund).
- Sachs, Jeffrey, Aaron Tornell, and Andrés Velasco, 1996, “Financial Crises in Emerging Markets: The Lessons from 1995,” Brookings Papers on Economic Activity, Vol. 1, pp. 147–98.
- Ghosh, Swati, and Atish Ghosh, 1999, “East Asia in the Aftermath: Was There a Crunch?,” Working Paper No. 99/38 (Washington: International Monetary Fund).
- Borio, Claudio, and Philip Lowe, 2002, “Assessing the Risk of Banking Crises,” BIS Quarterly Review, pp. 43–54 (Basel, Switzerland: Bank for International Settlements).

### Methodological contributions and reviews
- Beck, Thorsten, Aslí Demirgüç-Kunt, and Ross Levine, 2004, “Bank Regulation, Concentration and Crises” (unpublished; Washington: World Bank).
- Boyd, J., P. Gomis, S. Kwak, and B. Smith, 2000, “A User’s Guide to Banking Crises,” (unpublished; University of Minnesota).
- Demirgüç-Kunt, Aslí, 1989, “Deposit-Institution Failures: A Review of Empirical Literature,” Economic Review, Federal Reserve Bank of Cleveland, Vol. 25, No. 4.
- Ghosh, Swati, and Atish Ghosh, 1999, “East Asia in the Aftermath: Was There a Crunch?,” Working Paper No. 99/38 (Washington: International Monetary Fund).
- Mishkin, Frederic S., 1996, “Understanding Financial Crises: A Developing Country Perspective,” NBER Working Paper No. 5600 (Cambridge Massachusetts: National Bureau of Economic Research).
- World Bank, 2001, “Finance for Growth: Policy Choices in a Volatile World, Policy Research Report (Washington: World Bank).

*Source: _wp0596 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp0596.pdf_