## _wp12163

## Source details

**Canonical URL:** [_wp12163](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12163.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12163.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12163.pdf.json)

---

### Key distilled findings and summary statistics
- Total identified banking crises, 1970–2011: 147 banking crises, of which 13 are borderline events.
- Other crisis counts, 1970–2011: 218 currency crises and 66 sovereign crises.
- Recent global wave of crises has (thus far) affected mostly advanced economies.
- Advanced economies tend to:
  - Use monetary and fiscal policies more extensively during banking crises than emerging and developing countries.
  - Exhibit fiscal outlays associated with financial sector interventions that are about half that in emerging and developing countries, despite relatively larger banking systems in advanced economies.
  - Experience larger output losses and increases in public debt than emerging and developing countries.
- A large gap between market and book values of bank equity in recent crises points to significant recapitalization needs in some countries.

### Definition and identification of systemic banking crises (operational criteria and thresholds)
- A banking crisis is considered systemic if both conditions are met:
  1) Significant signs of financial distress in the banking system (significant bank runs, losses in the banking system, and/or bank liquidations).
  2) Significant banking policy intervention measures in response to significant losses in the banking system.
- Policy interventions considered significant if at least three out of six measures have been used:
  1) Extensive liquidity support (5 percent of deposits and liabilities to nonresidents).
  2) Bank restructuring gross costs (at least 3 percent of GDP).
  3) Significant bank nationalizations.
  4) Significant guarantees put in place.
  5) Significant asset purchases (at least 5 percent of GDP).
  6) Deposit freezes and/or bank holidays.
- Additional sufficient conditions for systemic classification:
  - A share of nonperforming loans above 20 percent or bank closures of at least 20 percent of banking system assets.
  - Fiscal restructuring costs of the banking sector exceeding 5 percent of GDP.
- Implementation notes and quantitative clarifications:
  - Liquidity support is extensive when the ratio of central bank claims on the financial sector to deposits and foreign liabilities exceeds 5 percent and more than doubles relative to its pre-crisis level.
  - For euro area countries, liquidity support is extensive if in a given semester the increase in this ratio is at least 5 percentage points.
  - Bank restructuring costs are measured as gross fiscal outlays (exclude liquidity assistance from the treasury) and are significant when they exceed 3 percent of GDP.
  - Significant asset purchases include those implemented through the treasury or the central bank and are defined as those exceeding 5 percent of GDP.
  - Significant guarantees indicate full protection of liabilities or guarantees extended to non-deposit liabilities (raising deposit insurance coverage alone is not included).
  - Significant nationalizations are takeovers of systemically important financial institutions where the government takes a majority stake.

### Recent crises, dating, and notable case developments (2007–2011)
- Dataset reports both starting date (first signs of significant distress) and date when crisis became systemic (when systemic definition first met).
- Starting dates for recent crises: United States and United Kingdom started in 2007; Nigeria in 2009; all other listed recent cases started in 2008.
- Crises that reached systemic proportions later:
  - 2009: Denmark, Germany, Greece, Ireland, Mongolia, and Ukraine.
  - 2010: Kazakhstan.
  - 2011: Nigeria and Spain.
- Systemic cases and main measures (selected list with start; systemic year; key measures):
  - Austria (2008; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Belgium (2008; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Denmark (2008; 2009) — extensive liquidity support; significant guarantees; significant nationalizations.
  - Germany (2008; 2009) — extensive liquidity support; significant guarantees; significant nationalizations.
  - Greece (2008; 2009) — extensive liquidity support; significant guarantees; significant restructuring costs.
  - Iceland (2008; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Ireland (2008; 2009) — extensive liquidity support; significant guarantees; significant restructuring costs; significant asset purchases; significant nationalizations.
  - Kazakhstan (2008; 2010) — extensive liquidity support; significant restructuring costs; significant nationalizations.
  - Latvia (2008; 2008) — extensive liquidity support; significant guarantees; significant nationalizations.
  - Luxembourg (2008; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Mongolia (2008; 2009) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Netherlands (2008; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Nigeria (2009; 2011) — extensive liquidity support; significant guarantees; significant restructuring costs; significant asset purchases; significant nationalizations.
  - Spain (2008; 2011) — extensive liquidity support; significant guarantees; significant restructuring costs.
  - Ukraine (2008; 2009) — extensive liquidity support; significant restructuring costs; significant nationalizations.
  - United Kingdom (2007; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant asset purchases; significant nationalizations.
  - United States (2007; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant asset purchases; significant nationalizations.
- Borderline 2008 cases (had not met the three-measure systemic threshold by the cutoff): France, Hungary, Italy, Portugal, Russia, Slovenia, Sweden, Switzerland.
- Case-specific updates:
  - Ireland’s NAMA acquired loans with a face value of €74 billion, qualifying as significant asset purchases.
  - Italy added to borderline systemic list because in 2011 liquidity support surpassed the study’s threshold.
  - Nigeria added as systemic after 2009–2011 policy actions and fiscal costs exceeded thresholds.

### Policy response patterns and interpretation
- Cross-country pattern:
  - Advanced economies rely more on countercyclical fiscal policy and generally have more space to use monetary policy, explaining greater use of macroeconomic policies during banking crises.
  - Greater reliance on macroeconomic policies in advanced economies is consistent with relatively lower fiscal outlays directed at financial sector interventions compared with emerging and developing countries (about half).
- Tradeoffs and consequences:
  - Expansionary macroeconomic policies indirectly support banks by enhancing growth prospects but risk slowing down actual bank restructuring.
  - Large gap between market and book values of bank equity in recent crises implies substantial recapitalization needs.
- Real effects:
  - Advanced economies suffer larger output losses during banking crises; deeper banking systems can make a banking crisis more disruptive.
  - Relatively larger increase in public debt in advanced economies is related to larger output losses and greater use of countercyclical fiscal policy.

### Monthly and regional patterns
- Monthly timing:
  - Banking crises tend to start in the second half of the year, with pronounced September and December effects.
- Geographic distribution:
  - Banking crises are a worldwide phenomenon; dataset documents regional distribution and countries with multiple systemic crises during 1970–2011.

### Frequency of systemic banking crises and crisis cycles
- Total counted systemic banking crises since 1970: 147, of which 13 are borderline events.
- Crises occur in waves; notable clusters:
  - Marked pick up in crisis activity in the early 1980s.
  - Three clusters during the 1990s: transition economies, Latin America (Tequila crisis), East Asia (Asian financial crisis).
  - Early 2000s relatively calm; ended with most recent wave, the largest number of crises since 1970.
- Credit boom association:
  - Out of 129 banking crisis episodes with available credit data, 45 episodes were preceded by a credit boom.
  - Credit boom definition: years when the deviation of credit-to-GDP ratio relative to its trend is greater than 1.5 times its historical standard deviation and its annual growth rate exceeds 10 percent, or years during which the annual growth rate of the credit-to-GDP ratio exceeds 20 percent. A country-specific cubic trend is computed over the preceding 10-year period.

### Currency and sovereign debt crises; twin and triplet crises
- Currency crises (1970–2011): 218 identified, of which 10 episodes occur during 2008–2011.
- Currency crisis definition: nominal depreciation vis-à-vis the U.S. dollar of at least 30 percent that is also at least 10 percentage points higher than the rate of depreciation in the year before.
- Sovereign debt crises and restructurings (1970–2011): 66 episodes, of which 3 episodes during 2008–2011.
- Greece restructuring in first half of 2012 yields one additional sovereign debt crisis case for 2012 (not included in 1970–2011 counts).
- Triplet crises (banking + currency + sovereign debt): count = 8 cases.
- Twin crises involving currency crises are most common; twin crises involving both banks and sovereign debt are least common.
- Twin/triplet timing: banking crisis in year T combined with currency (sovereign debt) crisis during [T-1, T+1]; triple crisis defined analogously with both currency and sovereign crises occurring in [T-1, T+1].

### Sequencing: banking crises and subsequent currency/sovereign crises
- Currency crises relative to banking crises:
  - 16% of banking crises are preceded by a currency crisis in the same country within three years prior to the starting year of the banking crisis.
  - 21% of banking crises are followed by a currency crisis within three years following the starting year of the banking crisis.
- Sovereign debt crises relative to banking crises:
  - 1% of banking crises are preceded by a sovereign debt crisis within three years prior to the start of the banking crisis.
  - 5% of banking crises are followed by a sovereign debt crisis within three years of the onset of the banking crisis.
- Conclusion: currency crises and especially sovereign debt crises tend to follow banking crises.

### Policy responses during banking crises (up to end-2011)
- Data coverage: policy responses updated through end-2011; deposit insurance presence data collected for about half the crisis episodes.
- Use of specific policy tools:
  - Deposit freezes: rare, most frequently used by emerging economies.
  - Guarantees on bank liabilities: more common among advanced economies; guarantees in recent crises on average more targeted than blanket guarantees in past crises.
  - Deposit insurance: in 70 percent of episodes for which data were collected, a deposit insurance scheme was already in place when the crisis erupted.
  - Losses imposed on bank creditors: only in 40 percent of cases losses are imposed on bank creditors.
  - Recapitalization packages and extensive liquidity support: more common in advanced countries, but only marginally so.
  - Nationalizations: equally common in advanced and other economies.
- Macroeconomic policy indices:
  - Expansionary fiscal and monetary policy indices take value 1 if expansionary, 0 otherwise.
  - Proxy for discretionary fiscal policy: difference between increase in public debt and fiscal cost of bank intervention packages; median of this variable is close to 7 percent.
  - Fiscal policy considered expansionary when this variable exceeds its mean by half a standard deviation.
  - Expansionary monetary policy defined when increase in reserve money is half a standard deviation above its mean.
  - Both expansionary monetary and fiscal policies were more commonly used in advanced economies; difference much more pronounced for fiscal policy.

### Measured outcomes of banking crises (definitions and medians)
- Outcome variables measured (definitions preserved as in source):
  - Fiscal costs: direct fiscal outlays due to financial sector rescue packages from start of crisis up to end-2011 (in percent of GDP).
  - Output losses: cumulative loss in income relative to a pre-crisis trend over [T, T+3], expressed as percent of trend real GDP. Trend computed by HP filter (λ=100) over [T-20, T-1] (or shorter if needed).
  - Increase in public debt: change in public debt-to-GDP ratio over [T-1, T+3].
  - Peak NPLs: peak nonperforming loans over [T, T+5] (or [T, latest data available]), in percent of total loans.
  - Peak liquidity support: highest level of central bank claims against financial institutions normalized by deposits and foreign liabilities; liquidity support computed as difference between peak and pre-crisis level.
  - Monetary expansion: change in monetary base between its peak during the crisis and its level one year prior to the crisis, expressed in percentage points of GDP.
  - Crisis duration: difference between end and start years, measured in years; truncated at 5 years.
- Table 2: Median outcomes (1970–2011) — All / Advanced / Emerging / Developing (in percent unless otherwise noted)
  - Output loss: All 23.0; Advanced 32.9; Emerging 26.0; Developing 1.6.
  - Increase in debt: All 12.1; Advanced 21.4; Emerging 9.1; Developing 10.9.
  - Monetary expansion: All 1.7; Advanced 8.3; Emerging 1.3; Developing 1.2.
  - Fiscal costs (in percent of GDP): All 6.8; Advanced 3.8; Emerging 10.0; Developing 10.0.
  - Fiscal costs (in percent of financial system assets): All 12.7; Advanced 2.1; Emerging 21.4; Developing 18.3.
  - Duration (in years): All 2.0; Advanced 3.0; Emerging 2.0; Developing 1.0.
  - Peak liquidity (in percent of deposits and foreign liabilities): All 20.1; Advanced 11.5; Emerging 22.3; Developing 22.6.
  - Liquidity support (in percent of financial system assets): All 9.6; Advanced 5.7; Emerging 11.1; Developing 12.3.
  - Peak NPLs (in percent of total loans): All 25.0; Advanced 4.0; Emerging 30.0; Developing 37.5.
- Notable patterns and country highlights:
  - Output losses and increases in public debt tend to be larger in advanced economies.
  - For crises that started in 2007 onwards, median output loss reaches 25 percent; non-crisis countries exhibit median output loss of 0 percent.
  - Fiscal costs are larger in developing and emerging economies, both as percent of GDP and percent of financial system assets.
  - In advanced economies, fiscal outlays associated with financial sector interventions constitute only a fraction of overall increase in public debt; discretionary fiscal policy and automatic stabilizers play larger roles.
  - Advanced economies show larger monetary expansion and longer crisis durations; monetary expansion difference between advanced vs. emerging/developing countries is now significant.

### Costliest crises and market-to-book gaps
- Fiscal-cost top-ten examples (In percent of GDP): Indonesia 1997 (57); Argentina 1980 (55); Iceland 2008 (44); Jamaica 1996 (44); Thailand 1997 (44); Chile 1981 (44); Ireland 2008 (43); Macedonia 1993 (32); Turkey 2000 (32); Korea 1997 (31).
- Increase in public debt top-ten (In percent of GDP): Guinea-Bissau 1995 (143); Congo, DR 1991 (130); Burundi 1994 (121); Thailand 1997 (109); Jordan 1989 (106); Ireland 2008 (106); Iceland 2008 (106); Indonesia 1997 (102); Tanzania 1987 (98); Nigeria 1991 (98).
- Output loss top-ten (In percent of GDP): Kuwait 1982 (143); Congo, DR 1991 (130); Burundi 1994 (121); Thailand 1997 (109); Jordan 1989 (106); Ireland 2008 (106); Latvia 2008 (103); Cameroon 1987 (102); Lebanon 1990 (98); Ecuador 1982 (98).
- Ireland is the only country in the sample to appear among the top-ten costliest crises along all three dimensions; Ireland’s crisis (starting 2008) was ongoing at time of reporting.
- Iceland: fiscal costs 44.2 percent of GDP; increase in public debt 61.2 percent of GDP within four years.
- Market-to-book gaps for distressed banks (end-2010) — median gap for banks with market-to-book values below one: about 3.3 percent of GDP.
  - Country examples (Difference between market and book value of distressed banks, In percent of GDP / In percent of total banking assets):
    - Austria 4.7 / 2.9
    - Belgium 6.7 / 2.4
    - France 5.9 / 2.0
    - Germany 1.6 / 1.2
    - Greece 7.7 / 3.2
    - Ireland 6.1 / 3.0
    - Italy 7.8 / 4.7
    - United Kingdom 6.1 / 1.5
    - United States 1.5 / 1.1
  - Note: aggregates for some countries based on less than 5 banks.

### Detailed fiscal outlays and recoveries for 2007–2011 (selected summary)
- Median fiscal cost for recent episodes (excluding borderline cases): 4.7 percent of GDP (2.8 percent with borderline episodes).
- Examples of gross outlays (In percent of GDP) and notable net outcomes:
  - Austria total fiscal outlays 4.9.
  - Belgium total fiscal outlays 6.0; total asset guarantees 7.7.
  - Greece total fiscal outlays 27.3 (includes 2012 IMF program recapitalization package estimate of 23.0); liquidity 1.9.
  - Iceland total fiscal outlays 44.2; recoveries 23.7; net outlays 20.5.
  - Ireland total fiscal outlays 40.7; asset purchases 20.3; asset guarantee NAMA 19.1.
  - Netherlands total fiscal outlays 12.7; recoveries 7.1; net outlays 5.6; total asset guarantees 10.8.
  - United Kingdom total fiscal outlays 8.8; recoveries 2.2; net outlays 6.6; significant asset guarantees 14.5.
  - United States total fiscal outlays 4.5; recoveries 2.4; net outlays 2.1; purchases: 13.0.
- Net fiscal recoveries examples:
  - Switzerland: mandatory convertible notes gross outlays 1.1; recoveries 1.5; net outlays -0.4.
  - United States: Capital Purchase Program (CPP) 1.5 gross outlays and 1.5 recoveries (net 0).
- Asset guarantees and purchases were a major component of interventions in many advanced economies.

### Cross-country aggregation: euro area vs United States (illustrative)
- Aggregated outcomes (In percent of GDP unless otherwise noted):
  - Euro area: Output loss 23.0; Increase in debt 19.9; Monetary expansion 8.3; Fiscal costs 3.9; Fiscal costs (in percent of financial system assets) 1.7; Peak liquidity 19.3; Liquidity support 13.3; Peak NPLs 3.8.
  - United States: Output loss 31.0; Increase in debt 23.6; Monetary expansion 7.9; Fiscal costs 4.5; Fiscal costs (in percent of financial system assets) 2.1; Peak liquidity 4.7; Liquidity support 4.7; Peak NPLs 3.9.
- Interpretation: crisis severity across fiscal costs, output losses, and increases in public debt comparable between euro area aggregate and the United States; liquidity support much larger in the euro area, reflecting the role of the Eurosystem (including ELA and LTRO). Monetary expansion similar thus far.

### Conclusions and research questions
- The updated IMF systemic banking crisis database:
  - Adds recent crises, updates output losses and fiscal costs for ongoing crises, and dates banking crises at monthly frequency.
- Key observations:
  - Traditionally costly banking crises were associated with emerging economies, but recent cases also involve advanced economies.
  - Recent advanced-economy crises have featured aggressive use of macroeconomic policies alongside relatively slow bank restructuring.
- Questions for further research:
  - Has there been a systematic change increasing fragility of banking systems in advanced economies with deeper financial markets and higher quality institutions?
  - What is the optimal policy mix and the implications of the pace of bank restructuring for recovery in advanced economies?

*Source: Laeven and Valencia, Working Paper updating the banking crises database (1970–2011), content as provided.*

### 2011. The data show some striking differences in policy responses between advanced and

### 2011. The data show some striking differences in policy responses between advanced and emerging economies as well as many similarities between past and ongoing crises

### Key distilled findings and summary statistics
- Total identified banking crises, 1970–2011: 147 banking crises, of which 13 are borderline events.
- Other crisis counts, 1970–2011: 218 currency crises and 66 sovereign crises.
- Recent global wave of crises has (thus far) affected mostly advanced economies.
- Advanced economies tend to:
  - Use monetary and fiscal policies more extensively during banking crises than emerging and developing countries.
  - Exhibit fiscal outlays associated with financial sector interventions that are about half that in emerging and developing countries, despite relatively larger banking systems in advanced economies.
  - Experience larger output losses and increases in public debt than emerging and developing countries.
- A large gap between market and book values of bank equity in recent crises points to significant recapitalization needs in some countries.

### Definition and identification of systemic banking crises (operational criteria and thresholds)
- A banking crisis is considered systemic if both of the following conditions are met:
  1) Significant signs of financial distress in the banking system (significant bank runs, losses in the banking system, and/or bank liquidations).
  2) Significant banking policy intervention measures in response to significant losses in the banking system.
- Policy interventions are considered significant if at least three out of the following six measures have been used:
  1) Extensive liquidity support (5 percent of deposits and liabilities to nonresidents).
  2) Bank restructuring gross costs (at least 3 percent of GDP).
  3) Significant bank nationalizations.
  4) Significant guarantees put in place.
  5) Significant asset purchases (at least 5 percent of GDP).
  6) Deposit freezes and/or bank holidays.
- Additional sufficient conditions for systemic classification:
  - A share of nonperforming loans above 20 percent or bank closures of at least 20 percent of banking system assets.
  - Fiscal restructuring costs of the banking sector exceeding 5 percent of GDP.
- Implementation notes and quantitative clarifications:
  - Liquidity support is considered extensive when the ratio of central bank claims on the financial sector to deposits and foreign liabilities exceeds 5 percent and more than doubles relative to its pre-crisis level.
  - For euro area countries, liquidity support is considered extensive if in a given semester the increase in this ratio is at least 5 percentage points (to account for Eurosystem data confounding).
  - Bank restructuring costs are measured as gross fiscal outlays (exclude liquidity assistance from the treasury in this measure) and are considered significant when they exceed 3 percent of GDP.
  - Significant asset purchases include those implemented through the treasury or the central bank and are defined as those exceeding 5 percent of GDP.
  - Significant guarantees indicate full protection of liabilities or guarantees extended to non-deposit liabilities (actions that only raise deposit insurance coverage are not included).
  - Significant nationalizations are takeovers of systemically important financial institutions where the government takes a majority stake.

### Recent crises, dating, and notable case developments (2007–2011)
- Crisis counts and dating refinements:
  - The dataset reports both the starting date (first signs of significant distress) and the date when the crisis became systemic (when the systemic definition was first met).
  - Starting dates for recent crises: United States and United Kingdom started in 2007; Nigeria in 2009; all other listed recent cases started in 2008.
  - Crises reached systemic proportions in later years in several instances:
    - 2009: Denmark, Germany, Greece, Ireland, Mongolia, and Ukraine became systemic.
    - 2010: Kazakhstan became systemic.
    - 2011: Nigeria and Spain became systemic.
- Table 1 (Systemic Banking Crises, 2007–2011) highlights (systemic cases listed in source):
  - Austria (start 2008; systemic 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Belgium (2008; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Denmark (2008; 2009) — extensive liquidity support; significant guarantees; significant nationalizations.
  - Germany (2008; 2009) — extensive liquidity support; significant guarantees; significant nationalizations.
  - Greece (2008; 2009) — extensive liquidity support; significant guarantees; significant restructuring costs.
  - Iceland (2008; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Ireland (2008; 2009) — extensive liquidity support; significant guarantees; significant restructuring costs; significant asset purchases; significant nationalizations.
  - Kazakhstan (2008; 2010) — extensive liquidity support; significant restructuring costs; significant nationalizations.
  - Latvia (2008; 2008) — extensive liquidity support; significant guarantees; significant nationalizations.
  - Luxembourg (2008; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Mongolia (2008; 2009) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Netherlands (2008; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant nationalizations.
  - Nigeria (2009; 2011) — extensive liquidity support; significant guarantees; significant restructuring costs; significant asset purchases; significant nationalizations.
  - Spain (2008; 2011) — extensive liquidity support; significant guarantees; significant restructuring costs.
  - Ukraine (2008; 2009) — extensive liquidity support; significant restructuring costs; significant nationalizations.
  - United Kingdom (2007; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant asset purchases; significant nationalizations.
  - United States (2007; 2008) — extensive liquidity support; significant guarantees; significant restructuring costs; significant asset purchases; significant nationalizations.
- Borderline cases in 2008 (had not met the three-measure systemic threshold by the source cutoff):
  - France, Hungary, Italy, Portugal, Russia, Slovenia, Sweden, Switzerland.
- Case-specific update: Ireland’s National Asset Management Agency (NAMA) acquired loans with a face value of €74 billion, qualifying as significant asset purchases according to the study’s definition.
- Italy was added to the list of borderline systemic banking crises because in 2011 liquidity support to the banking system surpassed the study’s threshold.
- Nigeria was added as systemic after 2009–2011 policy actions and fiscal costs exceeded thresholds (2009 liquidity support and guarantees; 2010 asset management company; 2011 transfer of nonperforming loans and large fiscal costs).

### Policy response patterns and interpretation
- Cross-country pattern:
  - Advanced economies rely more on countercyclical fiscal policy and have generally more space to use monetary policy, explaining greater use of macroeconomic policies during banking crises.
  - Greater reliance on macroeconomic policies in advanced economies is consistent with relatively lower fiscal outlays directed at financial sector interventions compared with emerging and developing countries (about half).
- Tradeoffs and consequences:
  - Expansionary macroeconomic policies indirectly support banks by enhancing growth prospects but risk slowing down actual bank restructuring.
  - The observed large gap between market and book values of bank equity in recent crises implies substantial recapitalization needs.
- Real effects:
  - Advanced economies suffer larger output losses during banking crises; deeper banking systems can make a banking crisis more disruptive (citing Kroszner, Laeven, and Klingebiel, 2007).
  - The relatively larger increase in public debt in advanced economies is related to larger output losses and greater use of countercyclical fiscal policy.

### Monthly and regional patterns
- Monthly timing:
  - Banking crises tend to start in the second half of the year, with pronounced September and December effects (Figure 1 in source).
- Geographic distribution:
  - Banking crises are a worldwide phenomenon; the dataset documents regional distribution and countries with multiple systemic crises during 1970–2011 (Figure 2 in source).

*Source: Laeven and Valencia, Working Paper updating the banking crises database, content as provided.*

### 2011. Many countries experienced more than one crisis over this period, but only two

### _wp12163 - 2011. Many countries experienced more than one crisis over this period, but only two

### Frequency of Systemic Banking Crises and Crisis Cycles
- Total counted systemic banking crises since 1970: 147, of which 13 are borderline events.
- Monthly frequency depiction reference: JanFebMarAprMayJunJulAugSepOctNovDec (figure context).
- Crises occur in waves; notable clusters:
  - Marked pick up in crisis activity in the early 1980s.
  - Three clusters during the 1990s: transition economies, Latin America (Tequila crisis), East Asia (Asian financial crisis).
  - Early 2000s relatively calm; ended with the most recent wave, the largest number of crises since 1970.
- Credit boom association:
  - Out of 129 banking crisis episodes with available credit data, 45 episodes (or about one in three) were preceded by a credit boom.
  - Credit boom definition (following Dell’Ariccia et al. (2012)): years when the deviation of credit-to-GDP ratio relative to its trend is greater than 1.5 times its historical standard deviation and its annual growth rate exceeds 10 percent, or years during which the annual growth rate of the credit-to-GDP ratio exceeds 20 percent. A country-specific cubic trend is computed over the preceding 10-year period.

### Currency and Sovereign Debt Crises; Twin and Triplet Crises
- Currency crises (1970–2011): 218 identified, of which 10 episodes occur during 2008–2011.
- Currency crisis definition: nominal depreciation vis-à-vis the U.S. dollar of at least 30 percent that is also at least 10 percentage points higher than the rate of depreciation in the year before.
- Sovereign debt crises and restructurings (1970–2011): 66 episodes identified, of which 3 episodes during 2008–2011.
- Greece restructuring in first half of 2012 yields one additional sovereign debt crisis case for 2012 (not included in 1970–2011 counts).
- Twin and triplet crises:
  - Triplet crises (banking + currency + sovereign debt) are rare: count = 8 cases.
  - Twin crises involving currency crises (banking+currency or currency+sovereign debt) are most common.
  - Twin crises involving both banks and sovereign debt are least common.
  - Twin crisis timing definition: banking crisis in year T, combined with currency (sovereign debt) crisis during [T-1, T+1]. Triple crisis defined analogously with both currency and sovereign crises occurring in [T-1, T+1].

### Sequencing: Banking Crises and Subsequent Currency/Sovereign Crises
- Currency crises relative to banking crises:
  - 16% of banking crises are preceded by a currency crisis in the same country within three years prior to the starting year of the banking crisis.
  - 21% of banking crises are followed by a currency crisis within three years following the starting year of the banking crisis.
- Sovereign debt crises relative to banking crises:
  - 1% of banking crises are preceded by a sovereign debt crisis within three years prior to the start of the banking crisis.
  - 5% of banking crises are followed by a sovereign debt crisis within three years of the onset of the banking crisis.
- Conclusion: currency crises and especially sovereign debt crises tend to follow banking crises.

### Policy Responses During Banking Crises (up to end-2011)
- Data coverage: policy responses updated through end-2011; deposit insurance presence data collected for about half the crisis episodes.
- Use of specific policy tools (differences by income group summarized qualitatively and visually in Figure 6):
  - Deposit freezes: rare, most frequently used by emerging economies.
  - Guarantees on bank liabilities: more common among advanced economies; guarantees in recent crises on average more targeted than blanket guarantees in past crises.
  - Deposit insurance: in 70 percent of episodes for which data were collected, a deposit insurance scheme was already in place when the crisis erupted.
  - Losses imposed on bank creditors: only in 40 percent of cases losses are imposed on bank creditors, indicating implicit guarantees remain important.
  - Recapitalization packages and extensive liquidity support: more common in advanced countries, but only marginally so.
  - Nationalizations: equally common in advanced and other economies.
- Macroeconomic policy indices and measurement:
  - Constructed expansionary fiscal and monetary policy indices taking value 1 if expansionary, 0 otherwise.
  - Proxy for discretionary fiscal policy: difference between increase in public debt and fiscal cost of bank intervention packages; median of this variable is close to 7 percent.
  - Fiscal policy considered expansionary when this variable exceeds its mean by half a standard deviation.
  - Expansionary monetary policy defined when increase in reserve money is half a standard deviation above its mean.
  - Both expansionary monetary and fiscal policies were more commonly used in advanced economies; difference much more pronounced for fiscal policy.
  - Advanced economies better able to finance large deficits, allowing automatic stabilizers and discretionary fiscal packages.

### Measured Outcomes of Banking Crises (definitions and medians)
- Outcome variables measured:
  - Fiscal costs: direct fiscal outlays due to financial sector rescue packages from start of crisis up to end-2011 (in percent of GDP).
  - Output losses: cumulative loss in income relative to a pre-crisis trend over [T, T+3], expressed as percent of trend real GDP. Trend computed by HP filter (λ=100) over [T-20, T-1] (or shorter if needed).
  - Increase in public debt: change in public debt-to-GDP ratio over [T-1, T+3].
  - Peak NPLs: peak nonperforming loans over [T, T+5] (or [T, latest data available] for recent episodes), expressed as percent of total loans.
  - Peak liquidity support: highest level of central bank claims against financial institutions normalized by deposits and foreign liabilities; liquidity support computed as difference between peak and pre-crisis level.
  - Monetary expansion: change in monetary base between its peak during the crisis and its level one year prior to the crisis, expressed in percentage points of GDP.
  - Crisis duration: difference between end and start years, measured in years; truncated at 5 years.
- Table 2: Median outcomes (1970–2011) — All / Advanced / Emerging / Developing (in percent unless otherwise noted)
  - Output loss: All 23.0; Advanced 32.9; Emerging 26.0; Developing 1.6.
  - Increase in debt: All 12.1; Advanced 21.4; Emerging 9.1; Developing 10.9.
  - Monetary expansion: All 1.7; Advanced 8.3; Emerging 1.3; Developing 1.2.
  - Fiscal costs (in percent of GDP): All 6.8; Advanced 3.8; Emerging 10.0; Developing 10.0.
  - Fiscal costs (in percent of financial system assets): All 12.7; Advanced 2.1; Emerging 21.4; Developing 18.3.
  - Duration (in years): All 2.0; Advanced 3.0; Emerging 2.0; Developing 1.0.
  - Peak liquidity (in percent of deposits and foreign liabilities): All 20.1; Advanced 11.5; Emerging 22.3; Developing 22.6.
  - Liquidity support (in percent of financial system assets): All 9.6; Advanced 5.7; Emerging 11.1; Developing 12.3.
  - Peak NPLs (in percent of total loans): All 25.0; Advanced 4.0; Emerging 30.0; Developing 37.5.
- Notable patterns and country highlights:
  - Output losses and increases in public debt tend to be larger in advanced economies.
  - For crises that started in 2007 onwards, median output loss reaches 25 percent; non-crisis countries exhibit median output loss of 0 percent.
  - Fiscal costs are larger in developing and emerging economies, both as percent of GDP and percent of financial system assets.
  - In advanced economies, fiscal outlays associated with financial sector interventions constitute only a fraction of overall increase in public debt; discretionary fiscal policy and automatic stabilizers play larger roles.
  - Advanced economies show larger monetary expansion and longer crisis durations; monetary expansion difference between advanced vs. emerging/developing countries is now significant.
- Costliest crises (summary highlights from figures and tables):
  - Fiscal-cost top-ten examples (In percent of GDP): Indonesia 1997 (57); Argentina 1980 (55); Iceland 2008 (44); Jamaica 1996 (44); Thailand 1997 (44); Chile 1981 (44); Ireland 2008 (43); Macedonia 1993 (32); Turkey 2000 (32); Korea 1997 (31).
  - Increase in public debt top-ten (In percent of GDP): Guinea-Bissau 1995 (143); Congo, DR 1991 (130); Burundi 1994 (121); Thailand 1997 (109); Jordan 1989 (106); Ireland 2008 (106); Iceland 2008 (106); Indonesia 1997 (102); Tanzania 1987 (98); Nigeria 1991 (98).
  - Output loss top-ten (In percent of GDP): Kuwait 1982 (143); Congo, DR 1991 (130); Burundi 1994 (121); Thailand 1997 (109); Jordan 1989 (106); Ireland 2008 (106); Latvia 2008 (103); Cameroon 1987 (102); Lebanon 1990 (98); Ecuador 1982 (98).
  - Ireland is the only country in the sample to appear among the top-ten costliest crises along all three dimensions; Ireland’s crisis (starting 2008) was ongoing at time of reporting.
  - Iceland: fiscal costs 44.2 percent of GDP; increase in public debt 61.2 percent of GDP within four years; large financial system size contributes to high fiscal cost relative to GDP.
- Market-to-book gaps for distressed banks (end-2010) — median gap for banks with market-to-book values below one: about 3.3 percent of GDP.
  - Country examples (Difference between market and book value of distressed banks, In percent of GDP / In percent of total banking assets):
    - Austria 4.7 / 2.9
    - Belgium 6.7 / 2.4
    - France 5.9 / 2.0
    - Germany 1.6 / 1.2
    - Greece 7.7 / 3.2
    - Ireland 6.1 / 3.0
    - Italy 7.8 / 4.7
    - United Kingdom 6.1 / 1.5
    - United States 1.5 / 1.1
  - Note: aggregates for some countries based on less than 5 banks.

### Detailed Fiscal Outlays and Recoveries for 2007–2011 (selected summary)
- Median fiscal cost for recent episodes (excluding borderline cases): 4.7 percent of GDP (2.8 percent with borderline episodes).
- Examples of gross outlays (In percent of GDP) and notable net outcomes:
  - Austria total fiscal outlays 4.9.
  - Belgium total fiscal outlays 6.0; total asset guarantees 7.7.
  - Greece total fiscal outlays 27.3 (includes 2012 IMF program recapitalization package estimate of 23.0); liquidity 1.9.
  - Iceland total fiscal outlays 44.2; recoveries 23.7; net outlays 20.5.
  - Ireland total fiscal outlays 40.7; asset purchases 20.3; asset guarantee NAMA 19.1.
  - Netherlands total fiscal outlays 12.7; recoveries 7.1; net outlays 5.6; total asset guarantees 10.8.
  - United Kingdom total fiscal outlays 8.8; recoveries 2.2; net outlays 6.6; significant asset guarantees 14.5.
  - United States total fiscal outlays 4.5; recoveries 2.4; net outlays 2.1; purchases: 13.0 (asset purchases).
- Net fiscal recoveries examples:
  - Switzerland: recapitulation mandatory convertible notes gross outlays 1.1; recoveries 1.5; net outlays -0.4.
  - United States: Capital Purchase Program (CPP) 1.5 gross outlays and 1.5 recoveries (net 0).
- Asset guarantees and purchases were a major component of interventions in many advanced economies.

### Cross-country Aggregation: Euro Area vs United States (illustrative)
- Aggregated outcomes (In percent of GDP unless otherwise noted):
  - Euro area: Output loss 23.0; Increase in debt 19.9; Monetary expansion 8.3; Fiscal costs 3.9; Fiscal costs (in percent of financial system assets) 1.7; Peak liquidity 19.3; Liquidity support 13.3; Peak NPLs 3.8.
  - United States: Output loss 31.0; Increase in debt 23.6; Monetary expansion 7.9; Fiscal costs 4.5; Fiscal costs (in percent of financial system assets) 2.1; Peak liquidity 4.7; Liquidity support 4.7; Peak NPLs 3.9.
- Interpretation: crisis severity across fiscal costs, output losses, and increases in public debt comparable between euro area aggregate and the United States; liquidity support much larger in the euro area, reflecting the role of the Eurosystem (including ELA and LTRO). Monetary expansion similar thus far.

### Conclusions and Research Questions
- Updated IMF systemic banking crisis database: adds recent crises, updates output losses and fiscal costs for ongoing crises, and dates banking crises at monthly frequency.
- Key observations:
  - Traditionally costly banking crises were associated with emerging economies, but recent cases also involve advanced economies.
  - Recent advanced-economy crises have featured aggressive use of macroeconomic policies alongside relatively slow bank restructuring.
- Questions for further research:
  - Has there been a systematic change increasing fragility of banking systems in advanced economies with deeper financial markets and higher quality institutions?
  - What is the optimal policy mix and the implications of the pace of bank restructuring for recovery in advanced economies?

*Source: Authors’ calculations and data compiled in the IMF update of Laeven and Valencia’s systemic banking crises database (1970–2011).*

### References

### _wp12163 - References

### References

- Abbas, S. Ali, Nazim Belhocine, Asmaa El Ganainy, and Mark Horton, 2010, “A Historical Public Debt Database,” IMF Working Paper No. 10/245.  
- Beim, David and Charles Calomiris, 2001, Emerging Financial Markets. Appendix to Chapter 1. New York: McGraw-Hill/Irwin Publishers.  
- Claessens, Stijn, Ceyla Pazarbasioglu, Luc Laeven, Marc Dobler, Fabian Valencia, Oana Nedelescu, Katharine Seal, 2011, “Crisis Management and Resolution: Early Lessons from the Financial Crisis,” IMF Staff Discussion Note No. 11/05.  
- Dell’Ariccia, Giovanni, Deniz Igan, Luc Laeven, and Hui Tong, 2012, “Policies for Macrofinancial Stability: How to Deal with Credit Booms,” IMF Staff Discussion Note No. 12/06.  
- Frankel, Jeffrey and Andrew Rose, 1996, “Currency Crashes in Emerging Markets: An Empirical Treatment,” Journal of International Economics, Vol. 41, pp. 351–366.  
- Kaminsky, Graciela and Carmen Reinhart, 1999, “The Twin Crises: The Causes of Banking and Balance-of-Payments Problems,” American Economic Review, Vol. 89, pp. 473–500.  
- Kroszner, Randall, Luc Laeven, and Daniela Klingebiel, 2007, “Banking Crises, Financial Dependence, and Growth,” Journal of Financial Economics, Vol. 84, pp. 187–228.  
- Laeven, Luc and Fabian Valencia, 2008, “Systemic Banking Crises: A New Database,” IMF Working Paper No. 08/224.  
- Laeven, Luc and Fabian Valencia, 2010, “Resolution of Banking Crises: The Good, the Bad, and the Ugly,” IMF Working Paper No. 10/44.  
- Reinhart, Carmen and Kenneth Rogoff, 2009, This Time is Different: Eight Centuries of Financial Folly, Princeton University Press.  
- Reinhart, Carmen and Kenneth Rogoff, 2011, “From Financial Crash to Debt Crisis,” American Economic Review, Vol. 101, pp. 1676–1706.  
- Sturzenegger, Federico and Jeromin Zettelmeyer, 2006, Debt Defaults and Lessons from a Decade of Crises. Cambridge: MIT Press.  
- World Bank, 2002, Global Development Finance. Appendix on Commercial Debt Restructuring. Washington, D.C.: World Bank.

*Source: _wp12163 - References*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12163.pdf_