## _wp08250

## Source details

**Canonical URL:** [_wp08250](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08250.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08250.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08250.pdf.json)

---

### Introduction — context and objectives
- The global liquidity crisis beginning with the U.S. subprime mortgage meltdown renewed debate over policy responses to financial crises, including use of guarantees on bank deposits.
- Bank runs are highlighted as destabilizing phenomena; Diamond and Dybvig (1983) show runs can occur in equilibrium because of banks’ illiquid balance sheets and self-fulfilling depositor beliefs.
- Policymakers often face uncertainty about causes of runs and potential outcomes if runs escalate; blanket guarantees are sometimes used to restore confidence but carry sizable fiscal contingencies and moral hazard.
- Over the last two decades more than a dozen countries have used blanket guarantees during banking distress; recent examples cited include Denmark, Germany, Iceland, and Ireland.
- Deposit insurance coverage increases cited: the US raised coverage from US$100,000 to US$250,000 per bank per depositor; European Union countries raised coverage from €50,000 to €100,000.

### Research scope and methodology
- Sample and focus:
  - Examines experience of 42 episodes of banking crises.
  - Identifies 14 episodes in which explicit blanket guarantees were extended.
  - Uses liquidity support extended by the monetary authority as a proxy for depositor confidence.
  - Focuses on explicit blanket guarantees defined as authorities’ explicit announcements protecting banks’ liabilities.
- Measurement:
  - Liquidity support measured as claims of the monetary authorities on the banking sector, expressed as a percentage of the total deposits and foreign liabilities of the banking system.
  - Currency pressures measured by a weighted average of percentage change in foreign reserves, interest rates, and the nominal exchange rate, with inverse of the standard deviation of each series as weights; index normalized to start at 100.
- Additional analyses:
  - Evolution of deposits and foreign liabilities examined to complement liquidity-support analysis.
  - Distinguishes explicit full-system guarantees, non-explicit guarantees (e.g., Norway), and limited/institution-specific guarantees (e.g., Northern Rock).

### Country experiences — characteristics of guarantees
- Coverage and duration:
  - Guarantees are usually comprehensive, including foreign and domestic currency liabilities, but commonly exclude subordinated debt (exception: Ireland includes dated subordinated debt) and liabilities to related parties.
  - Duration varies from as little as 11 months to over 8 years (examples: Mexico and Turkey with long-lasting guarantees).
  - Some guarantees were preceded by partial coverage announcements (examples: Thailand, Korea, Nicaragua).
- Notable country examples and features:
  - Mexico: first step toward an unlimited guarantee in December 1993; guarantee lifted gradually starting January 1999.
  - Turkey: May 5, 1994 announcement covering all savings deposits; guarantee was never lifted and was reinforced in December 2000.
  - Nordic crises: Finland and Sweden used explicit guarantees; Norway’s authorities issued assurances that may have been interpreted as blanket protection though not a conventional explicit guarantee.
  - Institutional or “de facto” protections occurred in several cases (examples: Norway late 1980s, Ecuador 1996, Peru 1999, Guatemala 2006, Bear Stearns 2008).

### Effectiveness of blanket guarantees — empirical patterns
- Core hypothesis:
  - Successful blanket guarantees should reduce deposit withdrawals and ease banks’ demand for central bank liquidity support.
- Visual and case evidence (20 episodes analyzed in section III):
  - First 15 cases: announcements covering all institutions (Table 1 plus Norway).
  - Last 5 cases: guarantees limited to a subset of the system.
- Observed patterns in liquidity support following guarantee announcements:
  - Statistically significant and quantitatively important slowdown in provision of liquidity support on average.
  - Cases with substantial reversals in liquidity support after announcement: Malaysia, Honduras, Mexico, Finland, Sweden, Turkey.
  - Cases where liquidity support decelerates or levels off (interpreted as partial success): Thailand, Nicaragua, Jamaica.
  - Cases where liquidity support continued to rise after announcement (possible loss of credibility or confounding factors): Indonesia, Japan, Ecuador.
    - Japan: fall in deposits stopped after the announcement, but foreign credit to banks continued to decline, sustaining liquidity pressures.
    - Ecuador: introduction of a tax with the guarantee affected financial transactions and triggered deposit runs immediately after the blanket guarantee.
    - Indonesia: other political and policy shocks preceding the guarantee undermined confidence and intensified the crisis.
- Foreign liabilities:
  - Foreign liabilities are largely insensitive to blanket guarantee announcements in virtually all cases examined.
  - Non-resident creditors often continue withdrawing resources after guarantees are announced.
  - In some cases capital outflows subsided temporarily (Jamaica, Thailand, Malaysia), but usually only for short periods (at most two months).
  - Countries where foreign liabilities declined sharply after guarantees often experienced significant currency pressures and declines in foreign reserves (examples: Finland, Indonesia, Japan, Korea, Norway, Sweden, Turkey). Korea had an earlier external-liability guarantee in August 1997 that was reinforced in October 1997.

### Fiscal impact and interactions with liquidity support
- Key patterns:
  - Blanket guarantees increase fiscal costs, but previous quantitative estimates may have overstated their independent impact.
  - Guarantees tend to be used in severe crises when liquidity support and fiscal costs are already high.
  - Guarantees add substantially to fiscal costs primarily when used in combination with extensive liquidity support.
- Proposed mechanism:
  - Excessive liquidity provision can fuel pressure on the currency, increase inflation, and deteriorate balance sheets of borrowers, banks, and government, creating self-reinforcing deterioration that raises ultimate cleanup costs.

### Empirical analysis: sample, variables, and regression approach
- Final regression sample:
  - 20 crisis episodes described earlier (excluding those corresponding to the ongoing 2007-2008 crisis) plus 22 additional cases from the Laeven and Valencia (2008) crisis database.
  - Two additional cases included not in Laeven and Valencia (2008): Honduras (1999) and Turkey (1994).
- Dependent variables:
  - First difference in the natural logarithm of the liquidity support indicator.
  - First difference in the natural logarithm of banks’ foreign liabilities.
- Key regressors:
  - Dummy for announcement (and onwards) of blanket guarantees.
  - Dummy for date of approval (and onwards) of IMF programs.
  - Dummy for announcement (and onwards) of bank restructuring policies.
  - First difference in the natural logarithm of the currency pressures index.
- Estimation:
  - Fixed effects estimator to control for unobserved country-specific factors.
  - Blanket guarantee indicator defined four ways: (a) entire banking system covered; (b) includes announcements covering only specific institutions; (c) and (d) as a and b with revised dates for Finland and Mexico.
  - Some regressions use lagged policy indicators to address timing mismatch.

### Regression findings — liquidity support
- Overall:
  - Announcement of a blanket guarantee is negatively associated with the expansion of liquidity support (consistent with improved public confidence).
- Statistical significance:
  - First four regressions: results not statistically significant.
  - Last four regressions (lagged specifications): columns (5) and (6) are significant and show greater quantitative importance; similar for columns (7) and (8).
- Coverage effect:
  - Blanket guarantees covering all banks (definitions a and c) have a larger quantitative impact than those limited to specific institutions (definitions b and d).
- IMF program dummy:
  - Positive sign on the IMF program variable (surprising): suggests liquidity pressures accelerate after IMF program announcement, likely because IMF involvement often coincides with rapidly deteriorating confidence prior to reversal by concrete actions.
- Bank restructuring policies:
  - Announcement of bank restructuring policies (most notably bank recapitalization schemes) has a highly statistically significant and quantitatively larger negative coefficient than the blanket guarantee indicator.
  - Example coefficient reported: Bank restructuring policies coefficient: -0.129*** (standard errors 0.038–0.041 depending on specification).
- Currency pressures index:
  - Coefficient is highly statistically and quantitatively significant.
  - Example from Table 3: ∆ Log of currency pressures index coefficient is 0.366** (standard error 0.171).
- Short-run vs medium-term effects:
  - Short run dummy: value 1 during first three months following announcement.
  - Medium-term dummy: value 1 from the fourth month onwards.
  - Results show a much larger impact of guarantees on liquidity support after the third month than during the first three months; statistical significance achieved only when the blanket guarantee covers all banks.
  - Short-run effect examples: blanket guarantee (a) (short run effect) 0.004 (0.073) — weak or sometimes positive.
  - Medium-term effect examples: blanket guarantee (a) (medium term effect) -0.108* (0.055) and -0.110** (0.050) in alternative specifications.

### Regression findings — foreign liabilities
- Main result:
  - Banks’ foreign liabilities do not respond positively to blanket guarantees; blanket guarantee and IMF program variables enter with negative signs, implying the decline in foreign liabilities accelerated after guarantee announcement.
- Example coefficients (Table 5):
  - Blanket guarantee (a) coefficient on ∆ Log of foreign liabilities: -0.0277*** (standard error 0.008).
  - Other definitions also negative and significant: -0.0247***, -0.0249***, -0.0223***.
  - Lagged blanket guarantee effects are smaller and less consistently significant (e.g., Lagged Blanket guarantee (a) -0.0208* (0.010)).
- Short-run vs medium-term (Table 6):
  - Short-run effects more negative and significant: Blanket guarantee (a) (short run effect) -0.037*** (0.009).
  - Medium-term effects smaller and less consistently significant: Blanket guarantee (a) (medium term effect) -0.0230** (0.010) or -0.015 (0.010) in alternative splits.
- Interpretation:
  - Non-resident creditors face lower exit costs and/or greater potential losses if guarantees are not honored, so they may withdraw despite guarantees.
  - Banks may reduce foreign liabilities to lower foreign exchange risk, sometimes using central bank resources, which can further reduce measured foreign liabilities.
  - Declines in foreign liabilities help explain why liquidity pressures may remain or rise after blanket guarantee announcements.

### Fiscal costs of blanket guarantees — empirical correlations and regressions
- Pairwise correlations and figures:
  - Pairwise correlation between the use of a blanket guarantee and fiscal outlays: about 0.3 (noted as 0.31 if Norway treated as using a guarantee, 0.37 if Norway not treated as using a guarantee).
  - Figure 2 and Laeven and Valencia (2008) sample: blanket guarantee users generally have higher gross fiscal costs but with large variation; examples: Finland, Honduras, Sweden contained fiscal costs despite guarantees.
- Table 7 averages:
  - Average fiscal cost (percent of GDP) when Blanket Guarantee = Yes and Extensive liquidity support = Yes: 26.8
  - Average fiscal cost (percent of GDP) when Blanket Guarantee = Yes and Extensive liquidity support = No: 7.7
  - Average fiscal cost (percent of GDP) when Blanket Guarantee = No and Extensive liquidity support = Yes: 14.6
  - Average fiscal cost (percent of GDP) when Blanket Guarantee = No and Extensive liquidity support = No: 6.2
- Regression analysis (Table 8) on Log(fiscal costs):
  - Blanket guarantee coefficient varies across specifications (examples: 0.518, 0.315, 0.402, 0.694**; standard errors range from 0.327 to 1.007).
  - Extensive Liquidity Support coefficient often positive and significant (example: 0.935*** (0.320) in Panel A column 1).
  - Crisis Intensity shows positive association with fiscal costs in several specifications (example: 0.265* (0.153)).
  - Interaction terms (blanket guarantee × extensive liquidity support, blanket guarantee × crisis intensity) reduce the standalone significance of the blanket guarantee variable in extended specifications.
- Interpretation:
  - The simple positive correlation between guarantees and fiscal costs largely reflects that guarantees are used in severe crises and often together with extensive liquidity support.
  - After allowing for interaction effects with crisis intensity and liquidity support, blanket guarantees do not always add significantly to fiscal costs in regression results.
  - Other factors increase fiscal costs: pre-existing banking-sector problems, delayed timing of guarantees, institutional or legal impediments (examples: Indonesia, Turkey, Ecuador).

### Timing, credibility, and real effects
- Timing:
  - Using guarantees late—after liquidity support is extensive—tends to associate with higher fiscal costs.
  - Early use (examples: Mexico 1993, Honduras 1999, Turkey 1994) sometimes produced early improvement in public confidence, but pressures often returned absent credible follow-up policy actions (noted exception: Honduras).
- Credibility and accompanying policies:
  - Credibility of the guarantee and credible policy actions addressing undercapitalization and macroeconomic imbalances are critical to success.
  - Where guarantees were accompanied by credible bank restructuring and macroeconomic policies, effectiveness improved.
- Real effects on output:
  - Bordo et al. (2001) find neither positive nor negative effects from blanket guarantees on output losses (as cited in text).

### Policy implications and recommendations
- Credibility is essential: blanket guarantees need to be credible to restore depositor confidence.
- Timing matters: the fiscal impact of guarantees may be reduced by using them as a breathing device before market conditions deteriorate substantially.
- Avoid combining guarantees with extensive liquidity support unless necessary; combination tends to be ineffective at restoring confidence and raises fiscal cost.
- Emphasize bank restructuring and macroeconomic stabilization:
  - Public confidence responds strongly to clear and comprehensive bank restructuring policies.
  - Sooner implementation of credible bank restructuring and macroeconomic policies reduces the need for blanket guarantees.
- Moral hazard and containment:
  - Blanket guarantees create moral hazard; banks may take greater risks and depositors cease screening banks.
  - Regulatory scrutiny and supervisory strengthening are necessary to contain moral hazard.
- Cross-border considerations:
  - Guarantees may not prevent non-resident withdrawal of foreign liabilities; additional measures may be needed to stabilize foreign funding.
  - Example: Irish guarantee effective Sept 29, 2008 triggered deposit flows from the United Kingdom.

### Conclusions (as presented in the source)
- Blanket guarantees can be effective in slowing deterioration in public confidence, conditional on credibility.
- Blanket guarantees tend to increase fiscal costs of resolving banking crises; a substantial part of these fiscal costs arises because guarantees are used in severe crises and together with extensive liquidity support.
- Foreign liabilities appear insensitive to the announcement of blanket guarantees; many countries observe continued declines in foreign liabilities after announcement.
- The success of a guarantee is significantly enhanced if accompanied by credible policy actions that address underlying banking and macroeconomic problems.

*Source: Excerpt from IMF Working Paper _wp08250.*

### References..............................................................................................................

### References

### Introduction — context and objectives
- The global liquidity crisis beginning with the U.S. subprime mortgage meltdown renewed debate over policy responses to financial crises, including use of guarantees on bank deposits.
- Bank runs are highlighted as destabilizing phenomena; Diamond and Dybvig (1983) show runs can occur in equilibrium because of banks’ illiquid balance sheets and self-fulfilling depositor beliefs.
- Policymakers often face uncertainty about causes of runs and potential outcomes if runs escalate; blanket guarantees are sometimes used to restore confidence but carry sizable fiscal contingencies and moral hazard.
- Over the last two decades more than a dozen countries have used blanket guarantees during banking distress; recent examples cited include Denmark, Germany, Iceland, and Ireland.
- Deposit insurance coverage increases cited: the US raised coverage from US$100,000 to US$250,000 per bank per depositor; European Union countries raised coverage from €50,000 to €100,000.

### Research scope and methodology
- Sample and focus:
  - Examines experience of 42 episodes of banking crises.
  - Identifies 14 episodes in which explicit blanket guarantees were extended.
  - Uses liquidity support extended by the monetary authority as a proxy for depositor confidence.
  - Focuses on explicit blanket guarantees defined as authorities’ explicit announcements protecting banks’ liabilities.
- Measurement:
  - Liquidity support measured as claims of the monetary authorities on the banking sector, expressed as a percentage of the total deposits and foreign liabilities of the banking system.
  - Currency pressures measured by a weighted average of percentage change in foreign reserves, interest rates, and the nominal exchange rate, with inverse of the standard deviation of each series as weights; index normalized to start at 100.
- Additional analyses:
  - Evolution of deposits and foreign liabilities examined to complement liquidity-support analysis.
  - Distinguishes explicit full-system guarantees, non-explicit guarantees (e.g., Norway), and limited/institution-specific guarantees (e.g., Northern Rock).

### Country experiences — characteristics of guarantees
- Coverage and duration:
  - Guarantees are usually comprehensive, including foreign and domestic currency liabilities, but commonly exclude subordinated debt (exception: Ireland includes dated subordinated debt) and liabilities to related parties.
  - Duration varies from as little as 11 months to over 8 years (examples: Mexico and Turkey with long-lasting guarantees).
  - Some guarantees were preceded by partial coverage announcements (examples: Thailand, Korea, Nicaragua).
- Notable country examples and features:
  - Mexico: first step toward an unlimited guarantee in December 1993; guarantee lifted gradually starting January 1999.
  - Turkey: May 5, 1994 announcement covering all savings deposits; guarantee was never lifted and was reinforced in December 2000.
  - Nordic crises: Finland and Sweden used explicit guarantees; Norway’s authorities issued assurances that may have been interpreted as blanket protection though not a conventional explicit guarantee.
  - Institutional or “de facto” protections occurred in several cases (examples: Norway late 1980s, Ecuador 1996, Peru 1999, Guatemala 2006, Bear Stearns 2008).

### Effectiveness of blanket guarantees — empirical patterns
- Core hypothesis: successful blanket guarantees should reduce deposit withdrawals and ease banks’ demand for central bank liquidity support.
- Visual and case evidence (20 episodes analyzed in section III):
  - First 15 cases: announcements covering all institutions (Table 1 plus Norway).
  - Last 5 cases: guarantees limited to a subset of the system.
- Observed patterns in liquidity support following guarantee announcements:
  - Statistically significant and quantitatively important slowdown in provision of liquidity support on average.
  - Cases with substantial reversals in liquidity support after announcement: Malaysia, Honduras, Mexico, Finland, Sweden, Turkey.
  - Cases where liquidity support decelerates or levels off (interpreted as partial success): Thailand, Nicaragua, Jamaica.
  - Cases where liquidity support continued to rise after announcement (possible loss of credibility or confounding factors): Indonesia, Japan, Ecuador.
    - Japan: fall in deposits stopped after the announcement, but foreign credit to banks continued to decline, sustaining liquidity pressures.
    - Ecuador: introduction of a tax with the guarantee affected financial transactions and triggered deposit runs immediately after the blanket guarantee.
    - Indonesia: other political and policy shocks preceding the guarantee undermined confidence and intensified the crisis.
- Foreign liabilities:
  - Foreign liabilities are largely insensitive to blanket guarantee announcements in virtually all cases examined.
  - Non-resident creditors often continue withdrawing resources after guarantees are announced.
  - In some cases capital outflows subsided temporarily (Jamaica, Thailand, Malaysia), but usually only for short periods (at most two months).
  - Countries where foreign liabilities declined sharply after guarantees often experienced significant currency pressures and declines in foreign reserves (examples: Finland, Indonesia, Japan, Korea, Norway, Sweden, Turkey). Korea had an earlier external-liability guarantee in August 1997 that was reinforced in October 1997.

### Fiscal impact and interactions with liquidity support
- Blanket guarantees increase fiscal costs, but previous quantitative estimates may have overstated their independent impact.
- Guarantees tend to be used in severe crises when liquidity support and fiscal costs are already high.
- Guarantees add substantially to fiscal costs primarily when used in combination with extensive liquidity support.
- Mechanism proposed:
  - Excessive liquidity provision can fuel pressure on the currency, increase inflation, and deteriorate balance sheets of borrowers, banks, and government, creating self-reinforcing deterioration that raises ultimate cleanup costs.

### Policy implications and recommendations
- Credibility is essential: blanket guarantees need to be credible to restore depositor confidence.
- Timing matters: the fiscal impact of guarantees may be reduced by using them as a breathing device before market conditions deteriorate substantially.
- Avoid combining guarantees with extensive liquidity support unless necessary; combination tends to be ineffective at restoring confidence and raises fiscal cost.
- Emphasize bank restructuring and macroeconomic stabilization:
  - Public confidence responds strongly to clear and comprehensive bank restructuring policies.
  - Sooner implementation of credible bank restructuring and macroeconomic policies reduces the need for blanket guarantees.

*Source: Excerpt from IMF Working Paper _wp08250 - References.*

### 1997. The guarantee was then extended to deposits in November 1997 as pressures built up.

### _wp08250 - 1997. The guarantee was then extended to deposits in November 1997 as pressures built up.

### Background and specific episodes
- November 1997: guarantee extended to deposits as pressures built up.
- A month later, panic spread when it became known that usable reserves were much smaller and external short-term debt much larger than markets had believed.
- Official foreign reserve data overstated usable reserves by including illiquid deposits at offshore Korean banks and omitted foreign debt contracted by offshore entities, underestimating external indebtedness substantially.
- Jamaica: an April 1997 transfer of about 11 billions of domestic currency units from the central bank’s claims on banks to FINSAC required an adjustment to the liquidity support series.
- Paraguay and Dominican Republic: interventions uncovered unrecorded deposits; eventually all depositors were compensated up to US$15,000 per depositor after a law was passed.
- Other notable country notes:
  - Czech Republic: no significant change detected following guarantee; deposit insurance coverage temporarily raised from CZK$100,000 to CZK$4,000,000 for affected banks.
  - Chile: full protection announced for depositors of intervened banks but undermined by government intent to impose a 30 percent loss to depositors and foreign creditors of banks in liquidation; foreign creditors cut lines until assured no losses would be imposed.

### Empirical analysis: sample and variables
- Final regression sample: 20 crisis episodes described earlier (excluding those corresponding to the ongoing 2007-2008 crisis) plus 22 additional cases from the Laeven and Valencia (2008) crisis database.
  - Note: Laeven and Valencia (2008) database includes 42 crisis episodes; United States and United Kingdom (2007-08) not considered because crises were still unfolding.
  - Two additional cases included not in Laeven and Valencia (2008): Honduras (1999) and Turkey (1994).
- Dependent variable: first difference in the natural logarithm of the liquidity support indicator (claims of the monetary authorities on the banking sector expressed as a percentage of the total deposits and foreign liabilities of the banking system).
- Key regressors:
  - Dummy for announcement (and onwards) of blanket guarantees.
  - Dummy for date of approval (and onwards) of IMF programs.
  - Dummy for announcement (and onwards) of bank restructuring policies.
  - First difference in the natural logarithm of the currency pressures index.
- Estimation approach: fixed effects estimator to control for unobserved country-specific factors (e.g., preexistence of a deposit insurance scheme).

### Regression specifications and measurement issues
- Blanket guarantee indicator defined four ways:
  - definition a: entire banking system covered.
  - definition b: includes episodes where announcement covered only specific institutions.
  - definitions c and d: same as a and b but with revised dates for Finland (February 1993 formally; first step August 1992) and Mexico (December 1993 implicit).
- Some regressions lagged the blanket guarantee and IMF program indicators to address timing mismatch between end-of-period monthly data and intra-month policy actions.
- Concern: end-of-month liquidity support stocks may include liquidity drawn before guarantees were announced; lagged specifications attempt to address this.

### Regression findings — liquidity support
- Overall: announcement of a blanket guarantee is negatively associated with the expansion of liquidity support (consistent with improved public confidence).
- Statistical significance:
  - First four regressions: results not statistically significant.
  - Last four regressions (lagged specifications): columns (5) and (6) are significant and show greater quantitative importance; similar for columns (7) and (8).
- Coverage effect:
  - Blanket guarantees covering all banks (definitions a and c) have a larger quantitative impact than those limited to specific institutions (definitions b and d).
- IMF program dummy:
  - Positive sign on the IMF program variable (surprising): suggests liquidity pressures accelerate after IMF program announcement, likely because IMF involvement often coincides with rapidly deteriorating confidence prior to reversal by concrete actions.
- Bank restructuring policies:
  - Announcement of bank restructuring policies (most notably bank recapitalization schemes) has a highly statistically significant and quantitatively larger negative coefficient than the blanket guarantee indicator.
  - Interpretation: demonstrating clear commitment to resolve banking-sector problems restores confidence faster and reduces the need for blanket guarantees.
- Currency pressures index:
  - Coefficient is highly statistically and quantitatively significant.
  - Mechanism: anticipation of currency weakening leads depositors to rebalance portfolios; banks may take long positions in foreign assets, exerting pressure on currency and domestic liquidity, possibly inducing increased central bank liquidity support.
- Fixed effects used to control for combined unobserved country-specific influences.

### Short-run vs medium-term effects of blanket guarantees
- Two dummy variables created:
  - Short run effect: value 1 during first three months following announcement.
  - Medium-term effect: value 1 from the fourth month onwards.
- Results:
  - Clear pattern of a much larger impact of guarantees on liquidity support after the third month than during the first three months.
  - Statistical significance achieved only when the blanket guarantee covers all banks.
- Puzzle: effect larger in medium-term than short-run; authors defer explanation until examining foreign liabilities behavior.

### Foreign liabilities (banks’ liabilities to non-residents)
- Dependent variable: first difference in the natural logarithm of banks’ foreign liabilities.
- Findings:
  - Banks’ foreign liabilities do not seem to respond positively to blanket guarantees; blanket guarantee and IMF program variables enter with negative signs, implying the decline in foreign liabilities accelerated after guarantee announcement.
  - Possible rationales:
    - Non-residents have a wider range of assets and/or lower exit transaction costs.
    - Non-residents face greater loss if they stay and guarantee is not honored, so they may withdraw despite guarantees.
    - Banks may decrease foreign liabilities to reduce foreign exchange risk, using central bank resources, further reducing measured foreign liabilities.
  - Currency pressures index: statistically insignificant in these regressions but quantitatively important in explaining foreign liabilities behavior.
- Consequence: decline in foreign liabilities helps explain why liquidity pressures may remain or rise after blanket guarantee announcements.

### Policy implications and interpretation
- Blanket guarantees tend to be associated with a slowdown or sharp decline in liquidity support, but effectiveness varies:
  - More effective when coverage includes all banks.
  - Less effective or undermined when other factors exist: limited political commitment (Indonesia), financial transactions tax (Ecuador), political crisis (Turkey).
  - Often coincident with other measures (e.g., IMF program approval, bank restructuring) that may drive improvement in confidence.
- Credible bank restructuring and concrete commitment to resolve banking problems are more effective than guarantees alone in restoring confidence.
- Addressing macroeconomic imbalances (captured by currency pressures) is crucial to alleviate banking-sector stress and reduce reliance on central bank liquidity support.
- Guarantees may not prevent non-resident withdrawal of foreign liabilities; additional measures may be needed to stabilize foreign funding.

*Source: _wp08250 - 1997. The guarantee was then extended to deposits in November 1997 as pressures built up.*

### introduction of a blanket guarantee.  Finally, the bank restructuring policies variable enters

### _wp08250 - introduction of a blanket guarantee.  Finally, the bank restructuring policies variable enters

### Effect on public confidence and liquidity support
- Blanket guarantees tend to improve public confidence in the banking system conditional on credibility and accompanying policies.
- Empirical regression highlights:
  - Bank restructuring policies coefficient: -0.129*** (column examples; standard errors 0.038–0.041 depending on specification), indicating a quantitatively larger and statistically stronger impact on liquidity support than blanket guarantees.
  - In Table 3, the ∆ Log of currency pressures index coefficient is 0.366** (standard error 0.171) across specifications; constant terms around 0.0655*** to 0.0723*** (standard errors 0.012–0.015).
- Short-run vs medium-term effects on liquidity support (Table 4):
  - Short-run effect of blanket guarantees is weak or sometimes positive (e.g., blanket guarantee (a) (short run effect) 0.004 (0.073)).
  - Medium-term effects show reductions in liquidity pressures: blanket guarantee (a) (medium term effect) -0.108* (0.055) and -0.110** (0.050) in alternative specifications.
- Interpretation:
  - Foreign creditors’ runs accelerate immediately after announcement of a blanket guarantee, increasing short-run liquidity pressures.
  - Over the medium term, runs ease somewhat, consistent with the view that non-residents who preferred to run did so immediately.

### Impact on foreign liabilities
- Blanket guarantees are ineffective at restoring foreign creditor confidence; many countries observe continued declines in foreign liabilities after announcement.
- Regression evidence (Table 5):
  - Blanket guarantee (a) coefficient on ∆ Log of foreign liabilities: -0.0277*** (standard error 0.008).
  - Other blanket guarantee definitions (b, c, d) also show negative and statistically significant short-run coefficients (e.g., -0.0247***, -0.0249***, -0.0223***).
  - Lagged blanket guarantee effects are smaller and less consistently significant (e.g., Lagged Blanket guarantee (a) -0.0208* (0.010)).
- Short-run vs medium-term (Table 6):
  - Short-run effects are more negative and significant (e.g., Blanket guarantee (a) (short run effect) -0.037*** (0.009)).
  - Medium-term effects are smaller in magnitude and less consistently significant (e.g., Blanket guarantee (a) (medium term effect) -0.0230** (0.010) or -0.015 (0.010) in alternative splits).

### Fiscal costs of blanket guarantees
- Blanket guarantees are associated with higher fiscal outlays in restructuring episodes, but causality is nuanced.
- Empirical correlations and figures:
  - Pairwise correlation between the use of a blanket guarantee and fiscal outlays: about 0.3 (noted as 0.31 if Norway treated as using a guarantee, 0.37 if Norway not treated as using a guarantee).
  - Laeven and Valencia (2008) sample-based evidence and Figure 2 show darker columns (blanket guarantee users) generally with higher gross fiscal costs, but with large variation (examples cited: Finland, Honduras, Sweden contained fiscal costs despite guarantees).
  - Table 7 averages:
    - Average fiscal cost (percent of GDP) when Blanket Guarantee = Yes and Extensive liquidity support = Yes: 26.8
    - Average fiscal cost (percent of GDP) when Blanket Guarantee = Yes and Extensive liquidity support = No: 7.7
    - Average fiscal cost (percent of GDP) when Blanket Guarantee = No and Extensive liquidity support = Yes: 14.6
    - Average fiscal cost (percent of GDP) when Blanket Guarantee = No and Extensive liquidity support = No: 6.2
- Regression analysis (Table 8) on Log(fiscal costs):
  - Blanket guarantee coefficient varies across specifications (example: 0.518, 0.315, 0.402, 0.694** in different columns; standard errors range from 0.327 to 1.007).
  - Extensive Liquidity Support coefficient often positive and significant (e.g., 0.935*** (0.320) in Panel A column 1).
  - Crisis Intensity shows positive association with fiscal costs in several specifications (e.g., 0.265* (0.153)).
  - Interaction terms (blanket guarantee × extensive liquidity support, blanket guarantee × crisis intensity) reduce the standalone significance of the blanket guarantee variable in extended specifications.
- Interpretation:
  - The simple positive correlation between guarantees and fiscal costs largely reflects that guarantees are used in severe crises and often together with extensive liquidity support.
  - After allowing for interaction effects with crisis intensity and liquidity support, blanket guarantees do not always add significantly to fiscal costs in regression results.
  - Other factors increase fiscal costs: pre-existing banking-sector problems, delayed timing of guarantees, institutional or legal impediments (examples: Indonesia, Turkey, Ecuador).

### Timing, credibility, and real effects
- Timing:
  - Using guarantees late—after liquidity support is extensive—tends to associate with higher fiscal costs.
  - Early use (examples: Mexico 1993, Honduras 1999, Turkey 1994) sometimes produced early improvement in public confidence, but pressures often returned absent credible follow-up policy actions (noted exception: Honduras).
- Credibility and accompanying policies:
  - Credibility of the guarantee and credible policy actions addressing undercapitalization and macroeconomic imbalances are critical to success.
  - Where guarantees were accompanied by credible bank restructuring and macroeconomic policies, effectiveness improved.
- Real effects on output:
  - Bordo et al. (2001) find neither positive nor negative effects from blanket guarantees on output losses (as cited in text).

### Policy implications and risks
- Moral hazard and behavior:
  - Blanket guarantees create moral hazard; banks may undertake riskier activities or "gamble for resurrection" while guarantees are in place.
  - Depositors cease screening banks, reducing market discipline; cross-border deposit migration can occur (example: Irish guarantee effective Sept 29, 2008 triggered deposit flows from the United Kingdom).
- Recommended practice:
  - Blanket guarantees should only be used as temporary measures.
  - Regulatory scrutiny and supervisory strengthening are necessary to contain moral hazard.
  - Preferably announce guarantees before a meltdown to reduce costs associated with extensive liquidity support, but ultimate outcome depends on credibility, accompanying policies, and shock severity.
- Specific concern:
  - Continued outflows of foreign liabilities after guarantees are announced may exacerbate liquidity pressures and should concern policymakers.

### Conclusions (as presented in the source)
- Blanket guarantees can be effective in slowing deterioration in public confidence, conditional on credibility.
- Blanket guarantees tend to increase fiscal costs of resolving banking crises; a substantial part of these fiscal costs arises because guarantees are used in severe crises and together with extensive liquidity support.
- Foreign liabilities appear insensitive to the announcement of blanket guarantees; many countries observe continued declines in foreign liabilities after announcement.
- The success of a guarantee is significantly enhanced if accompanied by credible policy actions that address underlying banking and macroeconomic problems.

*IMF staff paper: _wp08250 - introduction of a blanket guarantee.  Finally, the bank restructuring policies variable enters*

### References

### _wp08250 - References

### Theoretical and empirical studies
- Bernanke, B. (1983), “Nonmonetary Effects of the Financial Crisis in Propagation of the Great Depression,” American Economic Review 73(3), 257–276.  
- Bordo, M., B. Eichengreen, D. Klingebiel, M. Martinez-Peria, and A. Rose (2001), “Is The Crisis Problem Growing More Severe?”, Economic Policy 16, 53-82.  
- Dell’Ariccia, G., E. Detragiache, and R. Rajan (2005), “The Real Effect of Banking Crises,” working paper 05/63, International Monetary Fund.  
- Diamond, D. and P. Dybvig (1983), “Bank Runs, Deposit Insurance, and Liquidity”. Journal of Political Economy 91, 401-419.  
- Gorton, G. (1985), “Bank Suspension of Convertibility”, Journal of Monetary Economics 15, 177-193.  
- Klein, M., J. Peek, and E. Rosengren (2002), “Troubled Banks, Impaired Foreign Direct Investment: The Role of Relative Access to Credit,” American Economic Review 92, 664–682.  
- Kroszner, R., L. Laeven, and D. Klingebiel (2007), “Banking Crises, Financial Dependence, and Growth”, Journal of Financial Economics 84, 187-228.  
- Saunders, A. and B. Wilson (1996), “Contagious Bank Runs: Evidence from the 1929-1933 Period”. Journal of Financial Intermediation 5, 409-423.  
- Peek, J., and E. Rosengren (1997), “The International Transmission of Financial Shocks: The Case of Japan,” American Economic Review 87, 495–505.  
- Peek, J., and E. Rosengren (1999), “Japanese Banking Problems: Implications for Lending in the United States,” New England Economic Review, (Jan), 25–36.  
- Peek, J., and E. Rosengren (2000), “Collateral Damage: Effects of the Japanese Bank Crisis on Real Activity in the United States,” American Economic Review 90, 30–45.  

### Cross-country databases, methods, and policy frameworks
- Laeven, L. and F. Valencia (2008), “Systemic Banking Crises: A New Database”. IMF Working Paper No. 08/224, International Monetary Fund: Washington, DC.  
- Honohan P. and L. Laeven (2005), Systemic Financial Distress: Containment and Resolution, Cambridge, UK: Cambridge University Press.  
- Hoelscher, D. and M. Quintyn (2003), “Managing Systemic Banking Crises”. IMF Occasional Paper No. 224, International Monetary Fund: Washington, DC.  
- Honohan, P. and D. Klingebiel (2003), “The Fiscal Cost Implications of an accommodating approach to banking crises”. Journal of Banking and Finance 21, 1539-1560.  
- Kane, E. and D. Klingebiel (2004), “Alternatives to Blanket Guarantees for Containing A Systemic Crisis”. Journal of Financial Stability 1, 31-63.  
- Kim, S.-J. and A. Mody (2004), “Managing Confidence in Emerging Market Bank Runs”. IMF Working Paper 04/235, International Monetary Fund, Washington, DC.  

### Country case studies and crisis reports
- Barandiarán E. and L. Hernández (1999), “Origins and Resolution of a banking crisis: Chile 1982-86”, Central Bank of Chile Working Papers #57.  
- Batunanggar, S. (2002), “Indonesia’s Banking Crisis Resolution: Lessons and The Way Forward”, mimeo, Center for Central Banking Studies (CCBS), Bank of England.  
- De La Torre, A., R. García-Saltos, and Y. Mascaró (2001), “Banking, Currency, and Debt Meltdown: Ecuador Crisis in the Late 1990s”. Unpublished mimeo, World Bank.  
- Enoch, C., B. Baldwin, O. Frécaut, and A. Kovanen (2001), “Indonesia: Anatomy of a Banking Crisis. Two Years of Living Dangerously 1997-1999”. IMF Working Paper No. 01/52, International Monetary Fund: Washington, DC.  
- Jácome, L. (2004), “The Late 1990s Financial Crisis in Ecuador: Institutional Weaknesses, Fiscal Rigidities, and Financial Dollarization at Work”. IMF Working Paper 04/12, International Monetary Fund: Washington, DC.  
- Jácome, L. (2008), “Central Bank Involvement in Banking Crises in Latin America”. IMF Working Paper 08/135, International Monetary Fund: Washington, DC.  
- Nyberg, P. and V. Vihriälä (1994), “The Finnish Banking Crisis and Its Handling”. Bank of Finland Discussion Paper No. 7/94.  
- Özatay, F. and G. Sak (2003), “Banking Sector Fragility and Turkey’s 2000-2001 Financial Crisis”. Central Bank of Turkey Discussion Paper, December.  
- Sacasa, N. (2001), “Informe de Gestión Año 2001”. Superintendencia de Bancos y de Otras Instituciones Financieras de Nicaragua.  
- Sanhueza, G. (2001), “Chilean Banking Crisis of the 1980s: Solutions and Estimation of the Costs”. Working Paper of Central Bank of Chile No. 104, Santiago: Central Bank of Chile.  
- Siamwalla, A. (2000), “Anatomy of the Thai Economic Crisis”. In: Peter C. Warr (Ed.), Thailand Beyond the Crisis. Routledge, London.  
- Thorvald, M., J. Solheim, and B. Vale (2004), “The Norwegian Banking Crisis”. Norges Bank Occasional Paper No. 33.  
- FINSAC (1998), “Intervention Activities”. Annual Report, available from www.finsac.com  

### IMF country reports and selected issues papers
- IMF (1995), “The Mexican Banking System”. IMF Country Report SM/95/153, International Monetary Fund: Washington, DC.  
- IMF (1996), “Czech Republic - Recent Economic Developments”. IMF Country Report SM/96/286, International Monetary Fund: Washington, DC.  
- IMF (1998), “Financial System Developments”. IMF Country Report SM/98/166, International Monetary Fund: Washington, DC.  
- IMF (1999), “Republic of Korea: Economic and Policy Developments”. IMF Country Report SM/99/285, supplement 1, International Monetary Fund: Washington, DC.  
- IMF (2000), “Overview of Paraguay’s Banking Crisis
”. IMF Selected Issues Paper SM/00/12, International Monetary Fund: Washington, DC.  
- IMF (2002), “Lessons from the Jamaican Financial Sector Crisis”. IMF Selected Issues Paper SM/02/241, International Monetary Fund: Washington, DC.  
- Jácome, L. (2004), “The Late 1990s Financial Crisis in Ecuador: Institutional Weaknesses, Fiscal Rigidities, and Financial Dollarization at Work”. IMF Working Paper 04/12, International Monetary Fund: Washington, DC.  

### Historical and institutional studies
- Haber, S. (2005), “Banking With and Without Deposit Insurance: Mexico’s Banking Experiments, 1884-2004”. In: A. Demirguc-Kunt, E. Kane, and L. Laeven (Eds.), Deposit Insurance around the World: Issues of Design and Implementation, Cambridge, MA: MIT Press.  
- Klein, M., J. Peek, and E. Rosengren (2002), “Troubled Banks, Impaired Foreign Direct Investment: The Role of Relative Access to Credit,” American Economic Review 92, 664–682.  
- Hoelscher, D. and M. Quintyn (2003), “Managing Systemic Banking Crises”. IMF Occasional Paper No. 224, International Monetary Fund: Washington, DC.  

*Source: _wp08250 - References*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08250.pdf_